At Oracle's Enterprise Solutions Group, we have only just begun to explore the performance and efficiency that Intel's Xeon Processor 5500 Series can bring to Oracle solutions.  We have Xeon 5500 servers in our Enterprise Technology Centers, are testing them with a variety of Oracle products, and will soon put them into production use.  So far, here's what we're seeing:

 

  • Enterprise-class capacity
  • Superior virtualization capabilities
  • Improved space, power and cooling efficiency
  • Great performance for Oracle software

 

As we move forward with Xeon 5500, I'll blog about some of the ways we're using the servers, and some of the specific results we're seeing.  I'll also talk more about some of the best practices we are discovering as we deploy Oracle solutions on Intel.

Datacenter Dynamic Power Management – Intelligent Power Management on Intel Xeon® 5500

 

With newly released Intel Xeon® 5500 Processor family, it comes with a new breed of datacenter power management technology - Intel® Intelligent Power Node Manager (Node Manager in short).

 

As a former datacenter engineering manager, I had personal experience of the management issues at datacenters, especially dealing with power allocations and cooling – we often assumed the worse case scenario as we could not predict when the server power consumption will peak. When it did peak, we had no way to control it. It is like driving with blindfold and hope for the best outcome. The safest bet was to make the road as wide as possible - leave enough headroom for the power budget, so that we would not run into power issues. But it resuled in under utilized power, or stranded power, that is quite a waste.

 

Over the course of last several years, we met with many IPDC (internet portal datacenter) companies. We heard over and over again of their datacenter power management challenges, which was even worse than I experienced. Many of the IPDC companies we talked with leased racks from datacenter service providers under strict power limits per rack. The number of servers per rack they can fit had direct impact to their bottomline. They did not want to under-populate the racks, as they had to pay more rent for the same amount of servers; they could not over-populate the racks as it would be over the power limits. Their power management issues could be best summerized as the following:

·        Over-allocation of power: Power allocation to servers does not match actual server power consumption. Power is typically allocated for worst case scenario based on server nameplate. Static allocation of power budget based on worst case scenario leads to inefficiencies and does not maximize use of available power capacity and rack space.

·        Under-population of rack space: As a direct result of the over-allocation problem, there is a lot of empty space on racks. When the business needs more compute capacity, they have to pay more for additional racks. There are not enough datacenter spaces for them to rent. As a result, they had to go to other cities even other countries – increased operational cost and supporting staff.

·        No capacity planning: There is not effective means to forecast and optimize power and performance dynamically at rack level. To improve power utilization, datacenters needs to track actual power and cooling consumption and dynamically adjust workload and power distribution for optimal performance at rack and datacenter levels.

This is where the Node Manager comes to play. Let’s take a look at what Node Manager and its companion software tool provided by Intel for rack and group level power management – Intel® Data Center Manager (DCM) will do:

Intel® Intelligent Power Node Manager (Node Manager)

Node Manager is an out-of-band (OOB) power management policy engine embedded in Intel server chipsets. Processors carry the capability to regulate their power consumption through the manipulation of the P- and T-states. Node Manager works with the BIOS and OS power management (OSPM) to perform this manipulation and dynamically adjust platform power to achieve maximum performance and power for a single node. Node Manager has the following features:

·        Dynamic Power Monitoring: Measures actual power consumption of a server platform within acceptable error margin of +/- 10%. Node Manager gathers information from PSMI instrumented power supplies, provides real-time power consumption data singly or as a time series, and reports through IPMI interface.

·        Platform Power Capping: Sets platform power to a targeted power budget while maintaining maximum performance for the given power level. Node Manager receives power policy from an external management console through IPMI interface and maintains power at targeted level by dynamically adjusting CPU p-states.

·        Power Threshold Alerting: Node Manager monitors platform power against targeted power budget. When the target power budget cannot be maintained, Node Manager sends out alerts to the management console

Intel® Data Center Manager (DCM)

DCM is software technology that provides power and thermal monitoring and management for servers, racks and groups of servers in datacenters. It builds on Node Manager and customers existing management consoles to bring platform power efficiency to End Users. DCM implements group level policies that aggregate node data across the entire rack or data center to track metrics, historical data and provide alerts to IT managers. This allows IT managers to establish group level power policies to limit consumption while dynamically DCM provides allows data centers to increase rack density, manage power peaks, and right size the power and cooling infrastructure. It is a software development kit (SDK) designed to plug-in to software management console products. It also has a reference user interface which was used in this POC as proxy for a management software product. Key DCM features are:

·        Group (server, rack, row, PDU and logical group) level monitoring and aggregation of power and thermals

·        Log and query for trend data for upto one year

·        Policy driven intelligent group power capping

·        User defined group level power alerts and notifications

·        Support of distributed architectures (across multiple racks)

 

What the combination of DCM and Node Manager will do to datacenter power management? Here is the magic part… With the DCM at group and rack level setting policies, Node Manager can dynamically report the power consumed by a server and adjust it within certain range, so that the overall power consumption of the rack or a particular server group could be managed within a given target. Why this is important? Let me use a real example to explain it:

 

IPDC Company XYZ (a name I cannot disclose in public) has a mission critical workload at their datacenter that runs 24x7 and there are workload fluctuations during the day. The CPU utilization is mostly at 50~60%, with few cases that it will jump to 100%, typical for datacenter operations. To be on the safe side, the current solution is to do a pre-qualification of the Xeon® 5400 server for the worst case at 100% CPU utilization which ran at ~300W. They used 300W for power allocation, which was considered significantly lower than the nameplate value of the power supply (650W).

 

With Xeon® 550, for the same workload at 100% throughput, the platform power consumption goes down to 230W, a 70W reduction from the previous generation CPU – a good reason to switch to a new platform due to the advance intelligent power optimization features on Xeon® 5500. But the story does not end there…

 

 

 

On top of that, we further analyze the effect of power capping using Node Manager and DCM. After many tests, we noticed that if we cap at 170W and the performance of impact for workload at 60% CPU utilization and blow is almost negligible. This means, that we 170W power capping, the platform can deliver the same level of services most of the time, with 50W less (230W-170W) power consumption. For occasional spike that is above 60% CPU utilization, there will be some performance impact. However, since the Company XYZ operates at below 60% CPU utilization most of the time, the performance impacts are tolerable. As a result, we can squeeze more power from the power allocation using the dynamic power management feature of Node Manager and DCM.

 

 

 

What does this mean to the Company XYZ? Well, we can do the math. The rack they lease today has the limit of 2,200W/rack. With the current Xeon® 5400 servers, they can put upto 7 servers per rack at 300W per server. With Xeon® 5500, they can safely put 9 servers at 230W per server – a 28% increase of the server density on the rack. Top it up, by using Node Manager and DCM to manage the power at rack level with power limit of 2,200W and dynamically adjust the power allocation among the servers, we can put at least 12 servers at an average of 170W power allocation per server – a 71% increase of the server density comparing with the situation today! This means a great saving for the Company XYZ. In this case, the power consumption of each server on the rack could go above 170W, or lower than 170W. DCM dynamically adjusts the power capping policy while holding the line for entire rack power consumption below 2,200W.

 

Of course, the power management result varies from workload to workload. There has to be workload-based optimization in order to achieve the best result. Also, we assume that the datacenter should be able to provide sufficient cooling for devices that consume power within the given power limit. Even though, the result we get from this test could not be applied universally to all IPDC customers, we have finally had a platform that can dynamically and intelligently monitor and adjust the platform power based on workload. For datacenter managers, you can manage power at rack level and datacenter level with optimized power allocation to fully utilize the datacenter power. Are you ready to give it a try?

   

 

What is “Data Intensive Computing”? Abstractly, think of the quest to make better decisions out of big data as a filtering or funneling process. Basically the process of taking that huge amount of data and filtering and transforming it down to the right piece(s) of information that you need to make what amounts to a one-bit decision: yes or no, buy or sell, live or die. Data Intensive Computing is this end-to-end process.

I like the phrase “Data Intensive Computing” because it better describes the problem as being data limited rather than the compute limited. Data intensive computing is about efficiently taking vast amounts of structured or unstructured data through a series of processing steps to transform raw data into intelligence, knowledge, experience, and ultimately better decision making. In today’s world, we ideally want this filtering process to occur in sub-seconds, or real time.

To do this in an efficient and scalable way we have to focus on the data itself, transforming and filtering it as quickly as possible, minimizing the movement of data when it is large, and only transporting it in some compressed or size reduced form.

  • Here are some key attributes that should be present in Data Intensive Computing architectures:
  • Fewer more powerful discreet components (CPU, GPU, etc…)
  • Utilize discreet components that have high I/O communication bandwidth
  • Consolidation of application workflows around data (move processing to data not data to processing)
  • Minimizing time-stealing data movement (especially when data is large)
  • Centralize rendering of data for visual analysis (move rendering to data not data to rendering)
  • Delivering visual (size reduced) representations of data to users, rather than the data itself

My next and last Blog post on this topic will be less abstract and more about real world “data intensive” applications.

Yesterday – Intel officially launched the Intel® Xeon® 5500 processor (formerly codenamed “Nehalem”) for servers and workstations. One of the most exciting uses of this new platform will be as a key building block in cloud computing infrastructure. Whether you’ve bought into the hype of cloud computing or are a jaded IT realist – you can’t afford to pass up this list of 10 reasons the Intel Xeon 5500 processor is perfect for the cloud.

 

 

 

 

 

  1. Efficiency. To get the greatest efficiency – the leaders of large-scale Internet providers place their datacenters next to hydroelectric power or other low-cost energy sources. Each watt saved flows straight to the bottom line. Similarly – cloud computing companies intensely scrutinize their server purchases – weighing some variation of this question: how much performance (and by extension, revenue) can I squeeze out of the equipment – versus the cost of procurement and operations. This is the essence of “efficiency”. And now – with Intel’s new Xeon 5500 processor – there’s great news for anyone building efficient cloud infrastructure. The Xeon 5500 can deliver up to 2.25X the computing performance at a similar system power envelope compared to Intel’s previous generation Xeon 5400 series1. (By the way – the Xeon 5400 is no efficiency slouch – as it’s been leading the industry-standard SpecPower results for two socket systems since the benchmark was created.2) Need more evidence of Xeon 5500 efficiency? Look no further than the amazing results announced by IBM – a score of 1860, which is a 64% leap over the previous high score for a two socket system.3 Results like this clearly demonstrate that the Xeon 5500 has the extremely efficient performance that cloud operators are seeking.
  2. Virtualization performance. If a cloud service provider has leveraged a virtualization layer in its architecture - the performance of virtual machines and the ratio of VMs to servers are key concerns. Enter the Xeon 5500 which boasts a stellar jump in virtualization performance, up to 2 times the previous generation Xeon 5400 series4 allowing virtualized clouds to squeeze even more capability out of their infrastructure.
  3. Adaptable. Cloud computing environments tend to be highly dynamic as usage ebbs and flows during the day, some applications scale rapidly while some shut down, and so on. To meet such shifting demand – it’s critical to have adaptable cloud building blocks. And here Intel’s Xeon 5500 shines: this processor has unique new intelligence to increase performance when needed (Intel Turbo Boost) and to reduce power consumption when demand falls (Intel Intelligent Power Management Technology).
  4. Designed for higher operating temperatures. Across the datacenter industry – there’s growing interest in the notion of running datacenters at warmer temperatures to conserve energy. For cloud computing mega-datacenters, this concept has been in practice for several years. But it’s not just the datacenter staff that needs to handle warmer climates - the equipment must tolerate the conditions as well. Intel’s Xeon 5500 has been designed to run at higher temperatures providing one more piece of the puzzle to enable more efficient cloud infrastructure environments5.
  5. 50% lower idle power. Cloud computing providers – like airlines and phone companies – need to run at the highest utilization possible to maintain a healthy P&L. Yet there are times when usage – and thus server utilization – drops and at these times, cloud service providers desire processors with low power consumption. The Xeon 5500 processor now boasts an idle power that’s up to 50% lower than the prior generation systems, reducing energy costs6.
  6. Advanced power management. Intel has incorporated special platform level power technologies into the Xeon 5500 platform – which open new avenues to managing server energy consumption beyond what’s already built into the processor. Intel Intelligent Power Node Manager is a power control policy engine that dynamically adjusts platform power to achieve the optimum performance-power ratio for each server. By setting user-defined platform energy policies – Node Manager can enable datacenter operators to increase server rack density while staying within a given power threshold. While results vary based on the type of application and server – Intel demonstrated up to 20% improvement in rack density by using Node Manager in a recent proof-of-concept with Baidu, a leading search engine7.
  7. High Performance Memory Architecture. Cloud computing and other highly scalable Internet services are often relying on workloads where it makes more sense to keep large volumes of memory in DRAM, close to the CPU, rather than on slower, more distant hard drives. “Memcached” – a distributed caching system used by many leading Internet companies – is but one example. The Intel Xeon 5500 offers several exciting memory architecture benefits over the previous generation: (1) Up to 3.5X the memory bandwidth8 by leveraging an integrated memory controller and Intel Quick Path Interconnect (QPI), (2) supports a larger memory footprint (144GB versus 128GB), and (3) DIMMs and QPI links automatically move to lower power states when not active. In these new caching and distributed workloads, where large memory architectures are crucial, the Intel Xeon 5500 offers real advantages.
  8. Perfect when paired with SSDs. Few technologies get datacenter gurus more excited than solid state drives – which can offer impressive performance gains over their rotating hard drive cousins at far lower energy consumption. But with SSDs that can read 1000 times more data into the CPU versus a HDD – you want a ravenous processing beast to handle the traffic. And – you’re catching on to the blog theme – the Xeon 5500 can provide up to 72% better performance using SSDs than even the previous generation Xeon systems9. Intel Xeon 5500 is truly a perfect engine to complement SSDs.
  9. Ideal for optimized server boards. For cloud infrastructure – where every watt is a pernicious tax – you need more than just an extremely efficient processor such as the Xeon 5500. You also need an optimized server platform that has been stripped of every unneeded feature, configured with world-class energy efficient components, and designed for reduced airflow that minimizes the use of fans. One such product is an Intel server motherboard – codenamed “Willowbrook” which has an impressively low idle power below 70W, considering it’s a dual Xeon 5500 performance rocket10.
  10. A competitive lever for cloud operators. Lastly, for a service provider scaling out its infrastructure – systems based on Intel Xeon 5500 processors could offer a competitive advantage versus service providers whose servers are 2 to 3 years old. Because of the performance leaps in Intel server processors in the past few generations – Intel Xeon 5500 based servers can handle the same performance load as up to three times the number of 3-year old dual core servers11. The benefit is clear: providing the same performance level but with far fewer servers means a leg-up on those service providers with more antiquated, less efficient infrastructure.

 

 

 

If you have made it through this lengthy top 10 list – you should have a better sense for the advantages of Intel’s latest processor for cloud computing environments. Of course, the best way to really see the benefits is to get an Intel Xeon 5500 based system from your preferred vendor and test with your own code.

 

 

 

 

 

1 - 11For Footnotes, Performance Background, and Legal information, please refer to the attached document.

The Intel® Xeon® 5500 Series Processor (aka Nehalem) officially stepped out from behind the curtain onto center stage today.  This processor is an engineering marvel…one that can intelligently provide phenomenal performance on demand, while also sipping power when not in use.

 

Any measure of energy efficiency consists of performance in conjunction with the amount power consumed, so let’s cover these “big” items first.

  • Performance:  As of March 30, 2009, Intel based 2 socket Xeon® 5500 series servers set at least 30 world performance records across a wide range of benchmarks that cover virtually every application type on the market. The performance results, just by themselves, are utterly amazing, and in general they are greater than 2x the Intel® Xeon® 5400 series processors (Harpertown).
  • System level power consumption:  The electricity bill is based on how much power the server consumes, so that is also an important part of the energy efficiency equation. In general, Intel® Xeon® 5500 (Nehalem) based servers consume equal or slightly less power under peak workloads vs. previous generation Intel® Xeon® 5400 based servers.  By increasing performance more than 2x over previous Intel server processors while keeping overall power consumption in check, this is a great recipe for energy efficient performance.  In addition, when servers are at idle or are not fully utilized, customers want them to consume the least amount of power possible. Because of some key new power management features built into Nehalem, system idle power is dramatically lower (up to 50% less) than previous generation Intel® Xeon® 5400 based servers.

 

Now let’s get into three of the “behind the curtain” details of how some of the energy efficiency improvements are achieved.

  1. Power gating:  When a core is inactive, the operating system can request the core to enter a deep C state. Xeon® 5500 series processors supports C6, which is called “power gating”. This essentially puts the core into such a low power state that it consumes very close to 0W when not in use.
  2. DIMM memory power management: Today’s servers often have a lot of DIMMs installed, but leaving them in their full power state all the time isn’t a very wise. The Xeon® 5500 processor can intelligently reduce DIMM power consumption when not active by using techniques such as clock gating (CKE) and putting the DIMMs in “sleep state”, called self-refresh.
  3. Increased # of performance states:  P-states enable the server to proportionally match the power consumption of the server to the desired performance output. For example, if the processor CPU utilization is less, the operating system may request a lower P-state. By doing this, the power consumption of the processor is reduced to match the lower performance required. All this happens dynamically and allows the processor to scale both performance and power up and down to intelligently meet the workload demands.

 

In summary, while it is interesting to get into these “behind the curtain details”, what matters most is the performance and power at a system level. Servers based on Intel® Xeon® 5500 Series Processors represent a quantum leap forward in terms of both performance and energy efficiency! Call up your favorite server vendor and “test drive” one today to see for yourself.  And…once you get your hands on one, let me know what you think.

ChrisPeters

Why Buy for the Little Guy

Posted by ChrisPeters Mar 30, 2009

In my blog titled top 10 reasons to buy in a recession ,I discussed generic reasons to invest.  While small businesses (and to some extent medium businesses) don’t have the scale to take advantage of some of the consolidation and cost savings gains discussed in my “why buy for the big guy” blog, the benefits of replacing older servers remains a strong value for smaller businesses also. 

 

For those business owners who don’t have dedicated IT staff, your technology is depended on for reliable, efficient operation of the business. Technology is depended on to support daily operations and services ... like website operation, email communication, customer management, purchasing and record retention among other things.  If not … then you (as business owner) must turn your attention away from customers and towards your technology platform – not something a you have the time or resources to do.

 

Before reading, you need to know that I’m not a small business owner but I do have a family (in many ways that is my business) and as such I have several computers at home to support the operation of my family.  Recently, I experienced some issues with my existing computers that have put me in the market for replacement technology.  Specifically, when doing my taxes this year and when trying to load TurboTax*, my computer did not have sufficient memory to support the new 2008 software version.  Additionally, my wife and I track our finances with Quicken* and recently the slow performance on our computer has resulted in us spending too much time doing data entry on a slow computer.  In essence, our technology now is limiting us from doing the things we need to do .. so time to upgrade .. and I am shopping for a new desktop.

 

These challenges are similar to what I foresee from small businesses when it comes to technology upgrades.  Here are two examples of customers who, as a result of a growing demands and slowing performance of existing technology, turned to a Xeon-based server to streamline operations, boost reliabiilty, improve customer service, improve competitiveness and open up business possibilities for themselves.

 

ð       Yellow River i-café See How this small i-café went from a situation where a demanding workload caused hardware failures, leading to down time, loss of revenue, and frustrated customers to a situation where Yellow River saw performance gains and head room for growth by upgrading to an new Intel Xeon based server.

 

ð       Lampworks. Read how Jason Harper of Lampworks went from “We knew that we’d hit the wall with our desktop-based server; it couldn’t bear the extra load. Our computer was suddenly a barrier to our growth, rather than a business enabler” ... to ... “We’re extremely excited about the growth possibilities that our new server gives us.”

 

As your technology ages, you have a choice.  Typically standard OEM manufacturer warranties are supported for 3 years with purchase of a new server.  Before you extend the warranty (for $800-1200 per server for additional two more years), evaluate the enhanced performance, improved energy efficiency and capability to replace many servers (either desktop or true server technology) with fewer new servers.  If you are already at the end of your extended warranty, can you run the risk of server failure inside your business or can you afford the unexpected expense of a service call. 

 

Additionally there are some government incentive programs, like the American Recovery and Reinvestment Act  that offer businesses accelerated depreciation (60% first year vs standard 20%) on new computer hardware which can lower you 2009 tax burden and accelerate ROI.

 

Is your current technology holding you back or in your way? ... If so, consider a new server.

 

Chris 

This blog post is meant to discuss some of the considerations for performance tuning your Intel® Xeon® Processor 5500 (“Nehalem-EP”) series based server. I’d like to do this by discussing the un-boxing process.

Step 1. Place the box on the floor

Step 2. Open the box

Step 3. Carefully remove the server.

Step 4. Plug the server into a keyboard, mouse, and monitor.

Step 5. Plug the server into the wall socket.

Step 6. Power on the server.

 

There. You are done tuning your Nehalem-EP based server for performance. “Really?” you ask? Well mostly. There are some considerations and I’ll discuss them. I can speak to this subject as I was asked to tune this class of system using the TPC-Cand TPC-Ebenchmarks.

BIOS / Firmware / Drivers

It is very important to remember to update your system's BIOS, firmware, and OS drivers before you do any deep performance tuning. I cannot over state the importance of this step. Your system's manufacturer should be able to provide the latest BIOS and firmware associated with your server. OS drivers are available through many sources these days. Typically these can be downloaded from OS vendors, hardware vendors, from the Linux open source community, or the platform's manufacturer.

A good example of this is the SATA driver associated with the ICH10. The ICH10 is part of the chipset that supports Nehalem-EP. I recommend going to Intel’s website and using the Intel Matrix Storage Manager driver for the SATA controller.

Understand your system

Last year, Nehalem launched for the desktop market segment. Now it is time for the server market. The Nehalem-EP processor is meant to be used in dual processor (DP) socket systems. Nehalem-EP is the follow on to the Intel Xeon Processor 5400 (“Harpertown”) series. However, Nehalem-EP is really very different from Harpertown. The Nehalem-EP processor is based on the Intel Core i7. The Nehalem-EP processor inherits the same architectural features as the Intel Core i7. Once you understand these features, then you can better tune your system for performance.

L3 Cache:

Nehalem-EP uses a level 3 cache. Depending on which SKU you are using it can be 4MB or 8MB in size. If you are interested in performance, then I would encourage you to pick the larger cache size SKU.

Hyper Threading Technology:

If some threads is good then more threads is better. This is where Hyper threading technologycomes in to play. Nehalem-EP provides this technology out of the box. So, on a typical DP server this will give your system 16 threads of processing goodness.

Intel Quick Path Interconnect:

Nehalem-EP supports a CPU interconnect known as Intel Quick Path Interconnect (QPI). This interconnect is the replacement for the Front Side Bus of old. QPI provides a point to point link to each of the processors and the Intel X58 chipset. The Nehalem-EP supports QPI speeds of up to 6.4GT/s. This provides a theoretical bandwidth of 25.6 GB/s. This is a welcome shift for Intel’s designs for the future.

Turbo Boost Technology:

As with the desktop SKUs of Nehalem, the Nehalem-EP supports Turbo Boost technology. This technology will run the CPU at a higher frequency than its rating. It will increase the frequency in steps of 133MHz until it achieves its upper thermal and electrical design requirements. Turbo Boost Technology is dynamic. In other words, the processor will decrease its core frequency if the temperature is too high. If your application is sensitive to core frequency changes and does not fully utilize all cores, then it may benefit from this technology.

Integrated Memory Controller:

Another key feature of Nehalem based processors is that they have the memory controller integrated into the processor. This allows for much lower memory latencies. The Nehalem-EP supports three channels of DDR3 memory. It is important to talk about DDR3 memory and population on Nehalem-EP based servers. As mentioned before Nehalem-EP supports three channels of memory and supports 800, 1067, and 1333 MT/s memory speeds. Those speeds are dependent on how many channels are populated with DIMMs. For instance, 1333 MT/s is supported in a single DIMM per channel configuration. 1067MT/s is supported in a single DIMM per channel and two DIMMs per channel configuration. 800MT/s is supported in all configurations. These speeds are based on dual ranked DIMMs. If you plan on filling up all the memory slots with as many DIMMS as possible you will end up running at 800MT/s. So, here is the consideration. Does your application need all that memory or could it use less memory running at a higher speed? If the answer is yes to the latter, then perhaps running two DIMMS per channel at 1067 MT/s is the best configuration.

 

To wrap things up here, we have looked at the new and Nehalem architecture, the importance of BIOS/ firmware/ OS drivers, and memory population. Your application's performance will vary, but I hope I have given you some things to narrow down your performance testing. Thanks for taking the time to read this blog post. For more great performance methodology tips please check out Shannon Cepeda’s blogposts on performance tuning.

 

Ready to compete for your chance for some time in the spotlight?  The Intel Datacenter Efficiency Challenge turns the camera on you and gives you the opportuntity to tell the world about your datacenter efficiency plans and compete for prizes.  I have included a quick blurb on the challenge below, including links to the details of the program that are on Facebook.

 

Intel’s Datacenter Efficiency Challenge seeks to identify innovative solutions to reduce energy and costs associated with datacenter operations and to highlight the role technology plays in the overall efficiency and sustainability of organizations.  Using Intel multi-core server technology such as the new Intel® Xeon® processor 5500 series (or other Intel-based server platforms currently available) as a starting point, participants will submit a short video “proposal” describing their plans to redesign their current data center and project their potential savings.

 

Entries will be reviewed within enterprise and small business categories by a panel of Intel and third-party judges, with winners and runners up selected based on the percentage of energy savings and the use of varying technologies. Bonus points will be awarded for design creativity. Grand prize winners will receive an energy efficient netbook/laptop, will be flown to San Francisco for the Intel Developer Forum in the fall where they will meet with senior data center experts and Patrick Gelsinger, and they will have the opportunity to bring their proposal to life with help from an Intel Xeon Processor 5500 series-based server.

 

More information about the competition including rules and judging criteria can be found on the program’s Facebook page at http://www.facebook.com/pages/Intels-Data-Center-Efficiency-Challenge/73261657891. To learn more about the latest in datacenter efficiency best practices please visit: http://www.intel.com/technology/eep/data-center-efficiency/.

 

We are really looking forward to seeing some real innovation & creativity in the proposals and the videos.

 

Best of luck!

 

Dave

 

David E Jenkins

Intel Corp - Server Platforms Group Marketing

 

4/1 edited to add entire FB link

ChrisPeters

A Very Merry Christmas

Posted by ChrisPeters Mar 30, 2009

For readers of my February Blog, I talked about being so excited that i felt like a kid on Christmas morning when it came to our upcoming Nehalem launch and shared a story about some customers I talked with.  Well I can now give you your presents and a little background on the experience I had back in February.

 

 

Time to play with our new technology toys.

 

Chris

I enjoy running, and I typically train with a heart rate monitor to help me stay in a certain zone during my workouts.  When I was out on a run the other day and started to settle into my zone, instead of my mind drifting off to a peaceful place (like it should), for some reason it started drawing parallels about the human heart and Turbo Boost Technology.  I decided to play along as I was my own captive audience.

 

Let’s start with the heart, which as a pump has evolved over a long time to be pretty darn reliable and adapt quickly based on the needs of its owner.  It’s nominally rated at about 70-100 beats per minute (BPM), which is all it needs to do to support most activities during a normal day.  If you take care of it and operate it within spec, it should provide many years of reliable service in that range. 

 

However, we know the heart is capable is much higher rates, and most every day I operate it well above the rated 70-100 BPM during my runs.  As long as I take in enough air, don’t overheat, or don’t cramp up, my heart can maintain these higher rates without much problem.  In fact, if I feel REALLY good on a particular day, I can probably go above my max heart rate, but it’s not recommended and a lot of bad things can happen (a typically accepted max BPM calculation is 220 BPM – your age).

 

How does this relate to Turbo Boost Technology?

 

Xeon® 5500 processors are spec’d at a rated frequency (for example, 2.93 GHz), and the processor and platform are designed to operate for an indefinite period of time at that frequency.  With Turbo Boost, the processor is now able to run higher than rated frequency whenever you need a boost in performance, provided it meets the following conditions:  (1) the operating system requests the extra performance (I want to go out running), and (2) the processor has power, current, or temperature headroom (I’m getting enough air, and not overheating or cramping up).  As long as those conditions are met, the processor will run at those higher frequencies to maximize performance whenever it’s needed, either for short periods, long periods, or somewhere in-between.  When your performance demands drop, the processor frequency drops down to normal.

 

How high can you Turbo?

 

Similar to your maximum heart rate, we need to set Turbo Boost frequency limits in the Xeon® 5500 processors.  For example, the highest Turbo frequency the 2.93 GHz processor can support is 3.33 GHz, which is a 400 MHz jump.  While there still could be platform headroom even at the highest Turbo Boost frequency (I’m still feeling good at my max heart rate), we need to set these limits to ensure the processors will function reliably for a good long time.

 

So let your servers get some exercise with Turbo Boost – they’ll thank you for it.

When Intel launches a new product have you ever wondered what happens in our own IT shop?  We actually do testing and look for things like increased performance, energy efficiency, and total cost of ownership. Check out what we have tested and our findings around the newest server….does it apply to you and your environment?

 

Selecting Server Processors to Reduce Total Cost Read about Intel IT’s analysis that has led them to standardize on high-end processors for two-socket servers for design computing and enterprise server virtualization.  Intel IT analysis includes performance and TCO comparisons of servers based on low- and high-end processors in the Intel® Xeon processor 5500 series

 

Improving EDA Batch Application Performance Read the results of Intel IT’s joint performance assessment with Synopsys of the Intel® Xeon® processor 5500 series running Synopsys applications for silicon design. Results show up to 13.14x improved performance over single-core processors for simulation workloads, and up to11.39x improved performance for OPC workloads.  In addition, the analysis shows that one server based on the new processors can replace up to 13 servers based on single-core processors for electronic design automation simulation workloads, and up to 11 for OPC workloads.

 

Faster Chip Design with Intel® Xeon® Processor 5500 Series Learn about Intel IT’s test results of the Intel® Xeon processor 5500 series that show significantly higher performance for both 64- and 32-bit electronic design automation (EDA) applications over 64-bit Intel Xeon processors with a single core.  Up to 3.77x performance improvement for a single-threaded job; 4.98x for a multi-threaded job; and up to 11.39x throughput improvement for distributed jobs.

ChrisPeters

Why Buy for the Big Guy

Posted by ChrisPeters Mar 30, 2009

Why Invest in IT … for Large Enterprises

 

In my blog titled top 10 reasons to buy in a recession  , I discussed generic reasons to invest.  For large enterprises with a large install base of servers (multiple data centers, row and rows or rooms and rooms of servers), you have the economies of scale on your side.  Most likely, about 40% of your existing servers use single-core processor technology and another estimated 40% based of dual-core processor (source IDC).  Running existing infrastructure on these slower servers is just plan inefficient compared to the new servers available on the new Intel Microarchitecture (Nehalem) – intel's 3rd generation of quad-core processors for 2 socket servers.

 

Based on Intel estimates, replacing nine single-core based servers with one new xeon 5500 can yield up to 90% lower operating costs, delivering a payback on investment in  as short at 8 months (learn more here) … or … by upgrading single-core, dual-core or even the latest quad-core processors can yield performance enhancements that can boost productivity or open up new business opportunities. 

 

Even though this is day of introduction, there are four large companies today that have already identified the benefits of using these new processors.  See their results below

 

ð       Play saw roaming mobile transaction times reduce from 102 minutes to 44 minutes from last years quad-core processors and expects to be able to reduce the cost of running its data centre with these energy efficient servers.

ð       Capgemini tested a virtualization environment and sees ability to help their development team be more productive while strengthening customer offerings … as exhibited by a reduction in response time from 12.46 sec to 5.56 seconds compared to last years quad-core processors

ð       The Technical University of Munich saw processing speeds increase by 66% and experience 4x memory bandwidth for applications leading them and their customers to consider new projects and compute models for their research and business.

ð       Business & Decision saw the ability for 20:1 virtualization ratios with utilization levels at approximately 55%, providing the ability to improve customer service levels, productivity, reduce implementation costs by 50% and anticipates a ROI of < 1year. 

 

The bottom line is that these customers are moving forward with technology investment as a core strategy to boost their business and cut costs – helping them to emerge stronger and more competitive in their industry as economic conditions improve.

 

What could the Intel Xeon processor 5500 series based server do in your business?

 

Chris

By:  Boyd Davis, General Manager, Server Platforms Group Marketing, Intel Corporation

 

 

I’m excited to welcome you to our new Server Solutions Insider blog. Our goal is simple. We want to create a community where industry insiders debate insights and industry newcomers plug into what’s possible with the next generation of server technology. So I hope that you’ll join me and become a Server Solutions Insider.

 

Here are a few of the new exciting capabilities you’ll find:

 

 

  • Industry Perspective – We’re working with the server ecosystem to bring you a wide range of experiences and perspectives.  Join one of the upcoming live chats in The Server Roomon 4/2 @ 10am pacific. Visit Open Port Radio to listen to podcasts from industry insiders, or comment on and contribute to blogs and discussions.
  • Become a Contributor/Submit an Article –Simply register and follow the “write a blog” or “create a document” links in the Action Box.
  • Stay on top of the latest developments and key new content through Twitter.
  • Access our XEON® 5500 Wikithat will include the best content available for Intel® Xeon® 5500 processor based Systems and Solutions including Case Studies, Best Practices, Known Solutions, Tools, Top Blogs, References, and Video links.
  • More frequent polling to better understand your needs and desires and ensure that the community content hits the mark based on the community needs.
  • And of course, we’re maintaining our popular "Ask an Expert" forum and other well established features from The Server Room.

 

 

Click around. Check out the content and tools. Provide your point of view. Share your experiences. Invite others to join the conversation. And most importantly, let us know how we can make this a more valuable forum for you.

 

 

 

 

 

Boyd .

 

 

 

 

Today, at our headquarters in the heart of Silicon Valley, we launched the most important Intel server product since 1995’s Intel® Pentium® Pro processor.

 

In 1995, fewer than 750Ku servers based on Intel architecture were sold, representing only 10 percent of the total revenue for server hardware purchases−most of these were simple department or print servers playing fairly modest roles in the computing hierarchy. The Pentium Pro processor was our first chip optimized for server workloads—you could call it the father of the Intel® Xeon® processor. With its multiprocessing front side bus and the first out of order microprocessor in the industry, we knew the technology was transformative, but we didn’t predict the role that the technology ultimately would play as a driver for growth of the internet.

 

Today, 8 of 10 servers are based on Intel technology. There are still proprietary systems in the market but Intel-based servers are growing every year into a broader range of ever higher end applications. Into this market segment, today we’ve introduced another transformative server technology: the Intel® Xeon® processor 5500 series (code-named Nehalem-EP) designed by the same core team in Oregon that did the original Pentium Pro design work.

 

There are several breakthrough capabilities incorporated into platforms based on the Xeon 5500. First is the awesome raw performance. The performance gains relative to the prior generation are greater than for any Xeon processor we’ve ever delivered. Across a range of enterprise workloads, Xeon 5500 processor-based systems will deliver 70-125 percent higher performance than systems based on the Xeon 5400 series. That’s a stunning 1.7x – 2.25x performance.

 

Xeon 5500-based platforms also deliver dramatic leaps in intelligence and adaptability. Server workloads are diverse, and the physical environment that servers operate in, are increasingly constrained. Breakthroughs in the Xeon 5500 include Intel Turbo Boost Technology, Hyper-Threading Technology, Power Gating, Extended Page Tables, and VT Flex migration. In combination, they enable servers to deliver outstanding results on a broad range of workloads: optimized for parallelism or sensitive to clock frequency, virtualized or native, performance critical or power limited.

 

With this combination of advances, we know the Xeon 5500 will have a profound impact on the server market segment. While we can’t predict exactly how the market will evolve, there are two areas where I believe the new Xeon processor is likely to be most transformative. First is discovery and invention. As the foundation of high performance computing solutions both big and small, the Xeon 5500 delivers the performance to help scientists unravel the mysteries of the universe, as well as speeding time to market for small local manufacturers. We are seeing extraordinary interest in the Xeon 5500 for HPC with well over 100K units already installed in HPC configurations on the day of launch.

 

As the internet expands toward our vision of 15 billion connected devices by 2015, the Xeon 5500 will also be a foundational technology for the transformation of internet infrastructure. Our industry is aligned on a vision for applications abstracted from optimized hardware, available on demand, and scalable to the masses—often called cloud computing. Executing to this vision requires underlying technology that incorporates the adaptability, capability, and intelligence of our newest Xeon processor.

 

Characterizing an introduction as the most important in over a decade is a serious—and debatable—statement. We didn’t make this statement last year and don’t expect to make it next year. The combination of immediate benefit and long-term upside make the Xeon 5500 truly exceptional. What do you think?

 

Pat

Our new product, the Intel Xeon Processor 5500 series, has ushered in what we at Intel call a new generation of intelligent server processors. Before I wrote this blog I had to look up the definition of intelligence (American Heritage Dictionary):

 

 

In•tel•li•gence n 1.a. The capacity to acquire and apply knowledge. b. The faculty of thought and reason.

 

 

In this context, I’d like to discuss two topics. (1) An Intelligent Product (2) An Intelligent Choice

 

 

An Intelligent Product: (the capacity to acquire and apply knowledge)

Key technology enhancements to the Xeon 5500 include a suite of new features and capabilities that enable servers utilizing these new processor to serve a wide range of server usages (from basic business to high performance computing) (from single threaded applications to well threaded applications) (from non virtualized to highly virtualized environments) and makes these servers adaptable to the environment you want to deploy it into.

              

ð       Intel Hyper-Threading Technology is back boosting performance for well threaded applications

ð       Intel Intelligent Power Technology adjusts server power consumption real time to workload

o       Automated Low Power States reduces CPU, Memory and I/O power without impacting performance

o        Integrated Power Gates dynamically turn cpu cores that are not in use to reduce idle power near 10W

ð       Intel Turbo-Boost Technology speeds up your processor when application demands peak

ð       Intel QuickPath Technology provides industry leading server bandwidth (up to 3.5x prior Xeon)

 

 

The benefits for IT and Business?

ð       A server platform that can adapt to your application environment allowing you to deploy it in one environment today with the knowledge you can repurpose it tomorrow, if needed

ð       A server platform that can adapt you changing workload demands over the course of a day, saving power when demands are low and better performance when you need it most

 

 

Read the Intel Xeon processor 5500 series platform brief to learn about these technologies

Visit this video about the new product and the technologies listed above

 

 

An Intelligent Choice: (the faculty of thought and reason)

Economic times are tough and we’re all struggling with spending choices (or not spending) at both a personal and corporate level. However, business spends about 2/3 of their IT budget maintaining existing servers (source IDC). IDC further estimates that 40% of the servers installed today are 4yr+ single core servers with another 40% being 3 year old dual-core. These servers are consuming a lot of valuable resources. With a heavy % of IT budget spent on operating costs, the challenge is that if you cut spending, you are cutting innovation. This limits business competitiveness.

 

 

What is the option? …. Server Refresh. Compared to installed single core Xeon servers, these new Xeon processors enable up to 9x performance per server, a 9:1 server consolidation opportunity (with flat performance), lowering operating costs by an estimated 90% and delivering an estimated up to 8 month payback on investment. That means that an investment in a new server today can pay for itself in less than a year, helping you to self fund more innovation or helping to boost the bottom line of your organization. If your environment is dual-core based, the opportunity is about a 3:1 consolidation opportunity.

 

 

               Download this pdf to understand the 8 month estimate

View a video demonstration highlighting the 9:1 consolidation and 3:1 consolidation

 

 

In summary the Xeon 5500 series is an intelligent product in it’s capability to adapt to both it’s application and user environment and an intelligent choice for IT investment delivering an estimated up to 8 month payback – much better than you can do in the stock market, bank or many other projects.

 

I think that this is the right product at the right time.  What do you think? ... I'd like to hear your reactions.

 

Chris

     

These days it seems I’m overrun with data, not just my professional life but my personal life too. The stress of it is affecting my quality of life and not in a good way. Like Sisyphus I wake each day and go to battle with it knowing that the data is growing every second of every day and by the end of the day I’ve made no overall headway. I battle with massive data because I know there’s a benefit to be gained. New insight, a fresh way of thinking that can spawn a breakthrough, a more accurate computer model to better predict with, basically better decision making. I’m motivated by the potential but fatigued by how ineffective I am coping with the growing data volume. Sound familiar?

The data growth we’ve experienced in the last decade has changed the world. The tools and methodologies for processing data are increasingly inefficient as data size grows and soon will reach the point of being ineffective. I feel a big change is necessary if we’re going to outrun this data avalanche, innovation that departs from the compute-centric approach of the last decade.

I know large research centers operating at petascale levels today can already generate more data than they can effectively store and manage, pressuring researchers to visualize and analyze the data in real-time and foregoing any ability to archive their work or do future analysis without re-computing the solution again.

Let’s try and think differently. The point of Petascale computing, Cloud, or Virtualization is the data or information being produced or more importantly the eventual insight to be gained. If we can’t feed the processors fast enough or effectively analyze the results then we’ve missed the mark. Keep in mind we’re talking about a data size problem that is far out pacing Moore’s law.

This end-to-end processing problem is a “Data Intensive Computing” problem. One we should focus on sooner rather than later.

More to come...

 

 



The HP team is fired up about our new line of Intel Xeon 5500 (Nehalem) -based servers – we call them HP ProLiant G6 – the sixth generation since we started counting back in 2001. We made a big investment to offer these new features across our entire line of rack, tower and blades – 11 new servers in all. Big breakthrough in thermals – and I’m not talking about long johns here – I mean the power draw is half of previous generation thanks to revamped server design including Sea of Sensors, Common Power Slots and the world’s only hardware implementation of Power Capping (great partnership with Intel on that one too).

 

 

In addition, we’ve built on our long history of server management leadership with a trio of new capabilities, ProLiant Onboard Administrator, Virtual Connect Flex-10 and Insight Dynamics. All together these allow you to squeeze every ounce of productivity out of your servers and your support staff.

 

The other major focus for this new server line was tying them directly to new offerings from services. Do you use Starbucks cards? Our Proactive Select debit card approach is similar – but instead of dispensing caramel macchiato we dispense 85 different mission-critical services such as health checks and SAN assessments.

 

 

Lastly, we know times our tough – and for customers wanting to conserve cash (that means you), our HP Financial Services team offers sale-lease back options that will give you a cash-infusion by taking your existing infrastructure off of your balance sheet. In this way, you can lean on HP to help you get the buying power to step up to a new generation of server capability.

 

 

Get a guided backstage tour of HP ProLiant G6 by the people who built it here

 

 

Read more about HP ProLiant G6 in our press release here.

ProLiant_BestSellers300x200.jpg

With 2S Xeon processors delivering outstanding energy efficient performance, I get many questions from customers on “what about the Dunnington-based Servers”, “would it make sense for me to continue to use 4S servers in my IT/Enterprise”, “am I making the right choices with Dunnington” and so on.

 

 

The answer is a yes, even considering our 2S Nehalem-EP based servers expected in Q1 2009.

 

 

Customers make choices based on their business requirements – whether the new platform would be able to meet their IT needs. And in making such a decision in today’s economic context, there is temptation to use a 2S server as it is costs less. However, the choice of the server platform should not be dictated by the price alone; there are other more important considerations that you have to look into before deciding on the right choice.

 

 

One of our customers in banking had a very interesting problem – the system should be capable of delivering a user-acceptable response time while maintain headroom for growth. And the customer was keen to use virtualization running his workloads. With these key parameters (among others) in mind, we sized both 2S and 4S servers for the customer. And it turned out, that both the 2S and 4S servers were able to meet the response time, but the 4S server could deliver the “headroom for growth”. The headroom was based on the bank’s projected business growth in the next 12-24 months, their existing datacenter facility (space, power, cooling) and opportunity to further consolidate applications in a virtualization environment. The bank could have easily settled on a multiple 2S servers running in a virtual pool, but when you factor in “headroom for growth” considerations, the 4S+ Xeon 7400 servers still deliver the performance scalability and expandability that is needed today and tomorrow.

 

The bank finally settled on ten 4S servers and ten 8S servers. These servers had the unique capability to scale (4S to 8S to 16S) within the same OS footprint (while keep costs under control) and also deliver the performance headroom for the bank’s further needs.

 

 

In doing the above exercise for the bank, simple guidelines emerged – a) select the right architecture that scales in the future, b) look at established OS/App technologies such as virtualization to consolidate the environment, c) make decisions based on “your workload” running on the new servers and use published benchmarks as indicatives only.

 

 

Let me know what you think – share your views.

 

Here’s the 9th follow-up post in my 10 Habits of Great Server Performance Tuners series. This one focuses on the ninth habit: Don’t Break the Law.

test2.JPG

 

Amdahl’s Law, I mean. Amdahl’s Law tells you how much improvement you can reasonably expect from tuning a part of your system. It is often used in the context of software optimization and parallelization. Basically, what it says is that the potential improvement (speedup) you will get from applying an optimization with a speedup of X to a fraction F of your system is equal to 1/((1-F) + F/X). More generally, the speedup you will get depends on how much of your system the optimization affects as well as how good the optimization is.

 

For example, say you think you can speed up a function that takes 25% of your application’s execution time. If you can speed it up by 2x, then the potential speedup of your whole application, according to Amdahl’s Law, is 1/((1-.25) + .25/2) or a 1.14x speedup. Knowing something like this means you can evaluate which is more important: a 2x optimization affecting 25% of your code or a 4x optimization affecting 10%. (It’s the 25% one.)

 

Amdahl’s Law can also be used in other situations, such as estimating the potential speedup from parallelization or evaluating system tuning changes. It can be tricky in certain cases, such as when there are overlapping or dependent parts of the system. So use your intuition as well. However, in general using this law can help you to focus on making the common usage case faster for your system.

 

Once you have a good understanding of Amdahl’s Law, you may want to check out Gustafson’s Law and Little’s Law as well. All are commonly used in performance optimization. Being armed with the knowledge of these theoretical basics can help you to sniff out suspicious performance results or conclusions, as Neil J Gunther humorously wrote about here.

 

So stay out of trouble with the law (both the ones I mentioned and the legal kind!), and look for my post on the last habit next month.

 

 

The industry has been waiting for Intel's new product, codenamed Nehalem - and for good reason.  With the recession in full swing and all of us re-thinking our spending habits, we are only spending our hard-earned money on the best and most important products and services ... and everyone is looking for a deal.

 

The Xeon 5500 based server is one of those products that is going to make IT, business leaders, financial accountants and Corporate Executives take notice.

 

Why? ... because an investment in a new server, used to replace nine aging single core servers, can pay for itself in an short as 8 months while dramatically reducing infrastructure space consumption, power & cooling requirements and operating costs.

 

For Finance, that means that a capital investment today, could create a positive cash flow stream within a year

For IT, that means newer, in-warranty servers delivering more performance per server simplifying operations and maintenance, with lower risk of problems

For Facility Managers, that means reduced power / cooling requirements and the ability to consolidate data centers and avoid new construction or retrofits

For the Corporate Executives, this means a more efficient business, with lower costs and more innovation that can spur business growth

 

Read My blog on the Server Room titled "Top 10 Reasons to Buy in a Recession"

View the 8 month payback calculation here

 

Read more about the collaboration of industry solutions in the Server Solutions Insider Forum and view the innovation from intel in the server room

 

Chris

Citrix has anticipated the potential for enhanced performance when using the new Intel Xeon 5500 with XenServer for some time now. We were fortunate to have received one of these gems from Intel in January to do our own comparison tests in our Bedford, MA test lab. We had just completed a series of performance validation tests for virtualizing SQL Server 2008 in a XenServer 5 environment (a very successful test series I might add...look for the technical WP on this shortly at: http://citrix.com/English/ps2/products/documents_onecat.asp?contentid=683148&cid=White+Papers). So we decided to keep the same test environment and repeat a portion of the SQL test series comparing Xeon E5405 vs. Xeon E5570-based XenServer hosts to see what kind of performance improvement we would get.

 

Each of the test servers were configured with a single SQL Server 2008 VM with 4 vCPU cores and 7GB RAM. We used DBHammer for SQL to create a 10 million record database and generated transaction processing workloads to simulate an actual SQL Server 2008 client workload. Workloads were created starting with 200 clients. Each client workload test ran for 30 minute periods. Starting after the first 10 minutes of each test, we began taking measurements using Microsoft performance monitor and continued doing so for the remaining 20 minutes of the test. Measurements of the maximum transactions-per second rate were taken every fifteen seconds until the 30 minute test period had elapsed. Increments of 200 clients were added until the maximum average CPU utilization level of 90 percent was reached. When average CPU utilization of 90 percent was reached, it was determined that the system was considered saturated and testing was ended.

 

 

Given the difference in the clock speeds of the two systems tested, 2.0 GHz and 2.93 GHz respectively, we fully expected to see some level of improved performance, but what we saw was pretty astounding, even to us. The Xeon E5405 XenServer host topped out at 1,600 clients, generating 13,708 maximum transactions per second. The Xeon E5570 XenServer host, by comparison, was able to sustain a load of 2,400 clients, generating 20,978 maximum transactions per second. That's an improvement of 53%!

 

 

Intel E5405 vs E5570 (640x443).jpg

 

The results are pretty clear. Not only is the new Intel Nehalem Xeon 5500 series a powerful addition to the Intel Server Solutions lineup, but it is a clear advantage to customers seeking to virtualize enterprise application platforms like SQL Server 2008 using XenServer.

 

Details of this comparison test will be posted in a whitepaper to

http://citrix.com/English/ps2/products/documents_onecat.asp?contentid=683148&cid=White+Papers on or about April 1.

 

Bernie Hannon

Director of Certification & Testing

Citrix Systems, Xen Products Division

3-25-2009 11-14-57 AM.jpgWith the Nehalem architecture-based processor server/workstation launch just around the corner, customers are demonstrating how quickly they can take advantage of the performance afforded to them by our newest processor. One recent example was a Philips* CT demonstration of a beating heart at the major European IT show, CeBit*.  According to the Philips demonstration, they were able to render the human heart twice as fast using our upcoming Nehalem workstation processors as compared to the previous generation Intel® Xeon® processor 54xx series.  Check out the beating heart demonstration video in this link using our Nehalem architecture-based workstation processor.

For those of us with no medical background,  3D imaging is a tool enabling doctors to visualize the human body without invasive tools.  The patient is scanned, data is acquired, and a 3D model is assembled and they are able to show what going on in the human body without touching it.  CT tools are able to produce thousand of pictures or slices per second.  This results in large datasets that take time to process.  Software running on a server or workstation processes this vast amount of data to generate detailed 3D images and then can be interpreted by medical staff.   The primary challenge is managing large datasets generated by large volume acquisitions while speeding time to diagnosis.

The heart is a particularly challenging to visualize using CT scanning.  It is a very fast moving organ, with little control over its movement because is involuntary, and the heart needs to be frozen in time to get accurate anatomical picture.  Another challenge for volume rendering is high level image quality necessary so that anatomical structures of the heart can be depicted in an accurate way.

One interesting aspect of Philips CT solution is that image rendering takes place in the in the CPU which means that the performance of their solution is not restricted to by the memory available on a graphics card and is not slowed down by transfer of data between the CPU and the graphics card.  The Philips’ rendering software is multi-threaded and is able to take advantage of Intel advances in multi-threading /multi-core processing technology and software optimization tools for best performance improvements.

I believe Philips’ demonstration is just one example of how Nehalem architecture-based processors will provide meaningful advances in 3D visualization, not just in medical fields, but other areas such as manufacturing and digital content creation.

Jimmy Leon

*Other names and brands may be claimed as the property of others.

The promise that Cloud Computing brings is that of a ubiquitous compute infrastructure that provides services accessible over the internet from anywhere, by any device, and as such Cloud Computing is currently getting lots of air-time on the web news services.


The challenge at present is much like in real life every cloud is different offering different services, different tools to manage, different SLA's different charging models etc - basically all clouds are unique.


Before embarking into the great unknown of Could Computing enterprise IT managers need to fully understand what they are being offered and to look closely at their current infrastructure to determine how best to utilise Cloud Computing. Its key to understand what apps and/or workload may be suitable for placing into the cloud, for many IT departments today this starts with assessing non-mission/business critical workloads - initail targets for deploying in the cloud could be travel expenses, job search, benefits packages etc.


Another consideration is whether to move directly to using an external cloud service or to utilise resources within ones own data centres to setup 'internal clouds'. For many IT departments this is a logical extension of the ongoing work to virtualize their existing infrastructure and provides a valuable learning platform before setting out into the great unknown of the external 'public' cloud infrastructure.


Some questions IT managers need to ask when evaluating Cloud Computing offerings


  • What's the SLA being offered - how reliable is the service, what's the guaranteed uptime, what's the response time to fix failures ( nearly every week there are reports of cloud failures ), whats the financial impact of loss of cloud services and how would this be recompensed.
  • What's the application latency/response time
  • How secure is the service - backup policy for data, how secure is my data, what's the isolation between my app/data and other customers in a multi-tenanted cloud.
  • Where is my app/data located - what's the impact on local data protection laws,

& the list goes on . . .


Something else that needs to be thought about when looking at utilizing Cloud computing is the interoperability between cloud vendors?


Today most vendors offering 'Platform as a Service'* have developed their own cloud platforms based around open source or commercially available hypervisors and these require specific tools and stack formatting to utilise the virtualised compute resources. Cloud vendors are also offering higher level application/web frameworks enabling the use of higher levels of abstraction e.g. Python, Ruby or .Net . All of this further drives the uniqueness of every cloud offering. This will no doubt change over time as the major virtualisation vendors make inroads into this space and standards get developed for packaging virtual machines to be loaded into the cloud and services interfaces.


But today there is little/limited interoperability between Cloud providers and this can limit the flexibility and usability of the Cloud.


Theres a good white paper here from Intel's IT folks on developing a Cloud Computing startegy with the enterprise.


So great promises, and lots of challenges, but as the saying goes 'every cloud has a silver lining' - IT just needs to work out how to extract the benefit before the sun comes out and the clouds float away.



* Platform as a Service - rent a 'hardware platform' in the cloud and load on to this hardware your own software stack, usually paid for by the hour for specific amount of CPU/memory/IO capacity as opposed to 'Software as a Service' - some element of software, either a building block to utilise within your own application ( e.g. credit card checking ) or a fully blown application, such as a CRM app, accessible via the web .

That’s right, now you can buy a supercomputer that fits right under your desk. The PSC’s (Personal supercomputer) of today would have been #1 on the Top500 http://www.top500.org in November of 1996. ASCI Red would have been the first system to overtake the performance these small supercomputers can provide today.

So why do you need one of these high performance bad boys? Well, if you are trying to keep up with technology, beat out your competition, and do it at the lowest cost possible then you just better think about buying one. Whether you are in manufacturing, engineering, financial services or life sciences, the benefits offered are huge. You can now simulate vs constructing expensive prototypes, you can do more work at your desk vs. waiting to schedule the job on the oversubscribed cluster and most importantly, it provides the competitive advantage you most dearly need to keep up with your customer’s demands for lower pricing.

There are a couple of very interesting solutions on the market right now that should be considered. One is the Cray CX1. This little monster can support up to 16 quad core processors! With Nehalem soon to launch, that is one heck of a lot of performance.

http://www.cray.com/products/CX1.aspx

The CX1 is also ICR (Intel Cluster Ready) certified. This certification helps to ensure end users the system will provide a positive out-of-box experience. When you install the system and turn it on, it just works. You are maximizing your investment. The last thing a small business needs is to make the investment and then spend days getting the system up and running.

http://software.intel.com/en-us/cluster-ready/

With today’s economy, many businesses are reducing cost and putting off capital expenditures. You have decided to be like many other businesses and wait just one more year to upgrade your current computing system. Your competition decided not to wait. Instead, they purchased a PSC that was Intel Cluster Ready Certified and are now more productive than ever before. They made the investment and are now turning out new designs faster and at a lower cost than ever before. You scratch your head wondering how they do it. As you try to save your business they are growing and winning new business. Sometimes, being aggressive in a difficult time is the prudent thing to do…

Still wondering if this is right for me? Concerned you don’t have the IT staff to support such as beast? You don’t have the budget? Is there software out there I can use? All are good questions/concerns, but the ICR certified PSC minimizes if not eliminates the need for an IT staff. The PSC is one of the most affordable cluster solutions on the market today. It plugs right into your wall socket…you have tamed the beast! As for software, if you are purchasing an ICR certified system, then there are numerous applications available and most likely, one you are already familiar with.

When you are getting ready to make your next workstation or high-end PC purchase, I strongly recommend you consider one of the new kids on the block, the PSC. If it is Intel Cluster Ready Certified, you can rest assured the solution you do buy will just work.

We are starting to see more and more IT related services being sold under the banner of “cloud computing.”  And it is not just coming from the large service providers like Google, Amazon, Yahoo, although they are certainly a big part of it.  Traditional hosting service providers are also starting to evolve their infrastructure to support this model.  A good part of the motivation to move to Cloud services is its potential to bring significant economic benefits to consumers.  We all know that servers are expensive to purchase and operate, and building and managing even a small size data center is no walk in the park.  For consumers of cloud services the pay-as-you-go model rather than having to own and manage their own infrastructure is attractive.  And the smaller consumer you are the more attractive this probably is.  So it is not surprising that we see individuals and small business as the leading adopters of this model.  Actually, it turns out that the business considerations look different if you are a large enterprise, but perhaps more on that at a later time.

But if you are a service provider what drives the economic model for you and how do you benefit?  Clearly, you need to deliver a differentiated and competitive portfolio of services; but a good part of this is also about how you drive down the cost of owning and operating your data center.  I see a difference between the “true cloud” service providers and the traditional hosting service providers in their approach to data center design and operations.  Efficiencies in data center are the key to driving down data center operating expenses.  And optimization is the path to achieving efficiencies.  There’s a lot of buzz in the press and the blogging world about how large Internet Portal Data Centers are optimizing and driving efficiencies.  Smaller cloud service operators and traditional hosting providers should be watching this to ensure not only their competitiveness, but also to ensure their ability to I’ll be sharing my perspectives on how the traditional hosting providers are evolving in upcoming blogs; in the mean time, I’d love to hear from you on what your data center challenges are and if/how you are going about driving efficiency in your operations.

The storage world is getting quite a bit of press lately. As IT organizations seek to reduce costs, streamline their management tasks, save power, increase capacity and access to their data and improve security, new types of storage solutions are being introduced continuously The growth rate of storage revenue is blistering, and the number of companies scrambling for the piece of the pie is impressive.

 

 

As with any hot area of technology, there are a lot of technological developments to keep track of, and of course there are disputes about which new trends will maintain dominance, and which technologies will end up carrying the day. I’ve writtenin the past about some of the potential improvements coming along with the introduction new Data Center Ethernet (DCE) [aka, Data Center Bridging (DCB)] IEEE standard improvements. These layer 2 improvements coupled with Fiber Channel over Ethernet (FCoE) can bring to the storage world a converged story for networking. However, noticeably missing from my previous post was a discussion of iSCSI. The iSCSI standard has been around since 2003, and many vendors including Intel have been selling adapters with various levels of support for iSCSI for quite some time. Microsoft has an iSCSI initiator built into their server OS stack, and VMWare also supports iSCSI in their virtualized system environments. This isn’t a comprehensive list, but just a summary to show that iSCSI is ‘real’.

 

 

So the question is how does FCoE fit in with iSCSI? Are they competing technologies, or complimentary, or neither? Did I leave iSCSI out of my earlier discussion because I believe the technology is doomed? Hardly…

 

 

The basic case for FCoE is pretty straight forward. If you have an FC SAN in your datacenter, as you roll out a new rack (for instance) you can eliminate the top-of-rack FC switch and also the need for two adapters in each of servers (one for Ethernet, and one for FC). There are additional potential savings as you move FCoE further and further into the network, but I think to keep the discussion simple, the key thing to note is “if you have an FC SAN…”; then you care about FCoE and can potentially gain from it. If you don’t, well then FCoE will not necessarily be your technology of choice. FCoE is not an Ethernet capability that will drive you to use Fiber Channel, but it may make expanding and growing your FC network cheaper, lower power, and more convenient, especially with virtualized blade servers.

 

 

To elaborate, today the iSCSI SAN market is growing quickly, but it is growing in areas where FC is not generally a popular solution. It is usually chosen where storage over IP on existing Ethernet was viewed as acceptable given the cost tradeoffs. In contrast, FC is deployed where key IT knowledge, and its higher cost hurdles are still overcome by its performance and high reliability; this is generally only the case in fairly large datacenters. iSCSI on the other hand tends to be used in smaller enterprises and potentially in Tier 2/3 datacenters. The question of iSCSI or FC/FCoE is likely not getting decided by the network card in the host initiator. It is decided by organizational needs, and the cost structure / ROI of the storage target deployment including IT personal / expertise. If an organization has FC deployed, FCoE makes sense. If you are designing a new datacenter, FCoE isn’t exactly going to drive your decision (although you may still want to use it). The better way to consider FCoE is that it is another storage over Ethernet option, specifically for FC (to add to iSCSI and NAS). So decide which is the best storage system solution for your Data Center and know that you can connect any of these solutions over Ethernet to your servers.

 

 

Finally, as the improvements in Ethernet proliferate and these DCE capabilities become mainstream the landscape may change for iSCSI and FC deployments, but the simple fact is that in most cases, the main commonality is that Ethernet is chosen fabric to deliver storage over the network. Whether the storage protocol that is used is iSCSI, or FCoE will be determined by a lot of things within an individual companies’ IT shop. The idea that iSCSI and FCoE are a pair of competitive technologies has been raised in various places around the tech industry, but in reality, these arguments are driven more by people who have some bias toward one solution or the other, it is not fundamentally driven by the end user demands for storage. Both technologies have their place, and both will be supported in varying ways by Intel as well as other ecosystem vendors in the future.

 

 

So iSCSI or FCoE? I think the answer is: Both.

 

 

As we move to 10 Gigabit networking, IT decision makers will have something great with both of these capabilities; convergence, and choice. Only Ethernet can deliver this.

 

 

Ben Hacker

scotthuck

Once in a life time project

Posted by scotthuck Mar 16, 2009

Have you ever worked on a project, or with a team than just “clicked”?  Everyone was excited to be at work.  They worked well together.  And while there were of course many challenges, everyone worked through them and the end result was something special? I worked on one such team back in the mid-90’s.  The team was a silicon design and marketing team here at Intel, and the end result was the P6 processor (the official name was the Pentium Pro processor).  That may be too long ago for many of you to remember this product…I guess I’m showing my age.  But the P6 was a revolutionary product.

 

So what made this team click?  I think it was the fact that we all knew we were doing something special; something that had never been done before. The P6 was the first out-of-order, Super-scale, Super-pipelined, Speculative execution, glue less MP x86 processor every produced.  And the jump in performance as compared to previous generation x86 processors was the biggest ever seen at that time.

 

What is interesting is that most people, who worked on that project, have a very similar view.  All look very fondly back at that project, there were special friendships created during that project which exist today, and even though many years have past, and many project have come and gone since them, the P6 was something special; a once-in-a-lifetime project which most remember as the best project the ever worked on.  I’ve been at Intel for 23 years, and the P6 is still the best project I ever worked on.

 

So why all the reminiscing?  Because the soon to be released Nehalem project (aka Intel(r) Xeon(r) processor 5500 Series for servers and workstations, and also known as the Intel® Core i7 processor for desktop)  has many of the same attributes as the P6 project.  While times have changed, and working in a global environment means you cannot duplicate all the things that made the P6 team so special (in particular a close, tight nit group of people working at a single site), there are many things about the Nehalem project which make it special.  While I can’t disclose all the details just yet, look for follow-on to this blog after the product launch where I can describe some if the things which truly make the Nehalem product revolutionary.  Things that make me just sit back and say wow!  What this team accomplished is truly amazing!

I was thinking about a catchy title when I suddenly recalled the ‘Look Who’s Talking’ movie series from a while back. After all catchy titles are key for blogs!.

Previously I shared some thoughts on overall TCO savings that could be achieved, performance benefits that can be realized and how to migrate from RISC to Intel architectures. We all agree that making a change for the sake of change is never a good thing and justifying a change in the current economic environment can be a challenging path. So let’s look at who is changing and the benefits they are realizing from making a change. (I do apologize for over-use of word change, this is not a political commercial)

  • BMW Group wanted to simplify management of their environment and reduce TCO of their proprietary RISC server infrastructure. BMW moved their SAP environment and achieved 2.75-3xperformance gains and greater energy efficiency and drove down cost.
  • Telefonica a major Telecom Service Provider in Europe migrated their mobile online billing system and achieved a 428%performance gain.
  • Florida Hospital moved their disaster recovery system and got higher availability, reduced recovery time and lower system maintenance costs

Changing architecture does not mean that you have to change the operating system and solution stack. In some cases IT organizations are choosing to retain their Solaris environment.

  • BT Vision wanted to triple their Data Center capacity without increasing their power consumption or consuming more space in their DataCenter. Deployed Solaris on Xeon and achieved 10xfaster performance in Solaris Applications, 25-50%increased availability and 80%savings on their underlying equipment

Hopefully these examples help in some way to show that you will not be the first trailblazer trying out something new and unproven.  IT Organizations have moved and are reaping the benefits of the change.

Finally being March 17th and Irish, I would like to wish you all a Happy St Paddy’s day!

In our previous post we noted that the state of the art power montoring in virtualized environments is much less advanced than power monitoring applied to physical systems.  There is a larger historical context, and economic implications in the planning and operation of data centers that make this problem worth exploring.

 

 

 

Let's look at a similar dynamic in a different context: In the region of the globe where I grew up, water used to be so inexpensive that residential use was not metered.  The water company would charge a fixed amount every month and that was it.  Hence, tenants in an apartment would never see a water bill.  The water bill was a predictable cost component in the total cost of the building and included in the rent.  Water was essentially an infinite resource and reflecting this fact, there were absolutely no incentives in the system for residents to reign in water use.

 

 

 

As the population increased, water became increasingly a more precious and expensive resource.  The water company started installing residential water meters, but bowing to tradition, landlords continued to pay the bills, which was still a very small portion of the overal operating costs.  Tenants still had no incentive to save water because they did not see the water bill.

 

 

 

Today there are very few regions in the world where water can be treated as an infinite resources.  The cost of water increased so much faster than other cost components to the point that landlords decided to expose this cost to tenants.  Hence the practice of tenants paying the specific consumption for the unit they occupy is common today.  Also, because this consumption is exposed at the individual unit level, the historical data can be used as the basis for the implementation of water conservation policies, for instance charging penalty rates for use beyond a certain threshold.

 

 

 

The use of power in the data center has been following a similar trajectory.  For many years the cost of power had been a noise level item in the cost of operating a data center.  It was practical to include the cost of electricity in the bill of the cost of the facilities.  Hence IT managers would never see the energy costs.  This situation is changing as we speak.  See for instance this recent article in Computerworld.

 

 

Recent Intel-based server platforms, such as the existing Bensley platform, and more recently, the Nehalem-EP platform to be introduced in March come with instrumented power supplies that allow the monitoring and control of power use at the individual server level.  This information allows compiling a historical record of actual power use that is much more accurate than the more traditional method of using derated nameplate power.

 

 

The historical information is useful for data center planning purposes by delivering a much tighter forecast, beneficial in two ways: by reducing the need to over-specify the power designed into the facility or by maximizing the amount of equipment that can be deployed for a fixed amount of power available.

 

From an operational perspective we can expect ever more aggressive implementations of power proportional computing in servers where we see large variations between power consumed at idle vs. power consumed at full load.  Ten years ago this variation used to be less than 10 percent.  Today 50 percent is not unusual.  Data center operators can expect wider swings in data center power demand.  Server power management technology provides the means to manage these swings, stay within a data center's power envelope, yet maintain existing service level agreements with customers.

 

 

There is still one more complication:  with the steep adoption of virtualization in the data center in the past two years starting with consolidation exercises, an increasing portion of business is being transacted using virtualized resources.  Under this new environment, using a physical host as the locus for billing power may not be sufficient anymore, especially in multi-tenant environments, where the cost centers for virtual machines running in a host may reside in different departments or even in different companies.

 

 

It is reasonable to expect that this mode of fine grained power management at the virtual machine level will take root in cloud computing and hosted environment where resources are typically deployed as virtualized resources.  Fine grained power monitoring and management makes sense in an environment where energy and carbon footpring is a major TCO component.  To the extent that energy costs are exposed to users along as the MIPS consumed, this information provides the checks and balances and the data to implement rational policies to manage energy consumption.

 

 

 

Based on the considerations above, we see a maturation process for power management practices in a given facility happening in three stages.

 

  1. Stage 1: Undifferentiated, one bill for the whole facility.  Power hogs and energy efficient equipment are thrown in the same pile.  Metrics to weed out inefficient equipment are hard to come by.
  2. Stage 2: Power monitoring at the physical host level implemented.  Exposes inefficient equipment.  Many installations are feeling the pain of increasing energy cost, but organizational inertia prevents passing costs to IT operations.  Power monitoring at this level may be too coarse grained, too little, too late for environments that are rapidly transitioning to virtualization with inadequate support for multi-tenancy.
  3. Stage 3: Power monitoring encompasses virtualized environments.  This capability would align power monitoring with the unit of delivery of value to customers.

As usual, after swimming in the morning, I thumbed through my Blackberry.  On the small glass screen, I saw the email from a friend, “Hi, For: Did you send this email to me?”  I was very puzzled by what she meant that I quickly scrolled down to see the full text below her message: “Dear friend, I would like to introduce a good company who trades mainly in electronic products…  I looked at the “From” line.  It is from my personal email account!  I knew immediately that some hacker hijacked my address book and used my email name to send out spam email.  But how did that happen?  How could I clean up this mess? I suspected that my not-so-strong password was hacked and I corrected it right away.  Since the send box identified who were the recipients. I then sent an email to explain the situation.  My sister-in-law shot me an email afterward: “I though it was a little strange.  

 

This cyber identity theft really makes me mad at the intruder and myself not taking more precaution measures.  I use my web email account everyday, save my personal data in the “cloud”, and provide my VISA card number to purchase online.  With the social media network, I may disclose even more personal information on the web.  This incident wakes me up that I need to protect myself diligently by adopting caution behaviors such as using the strong password or making sure confidential data are encrypted.  I also realize how much trust I have put in the datacenter and service provider that I may not even realize until I am personally affected.  Do the servers enforce strong passwords only?  How do I know the communication between my personal computers and the servers are secured?  Can the service provider be trusted?  It takes both the consumer end and the service providers together to create a secured environment.  Service provides have the fiduciary duty to protect their customers and their investors by focusing on datacenter security issues.  It may take only one security compromise to shake up the trust of the customers. 

 

I have been with Intel’s server group for the last 13 years and experienced many server technologies from form factor to power saving that have transformed the datacenters.  With our upcoming server platforms, we will be placing more focus on helping datacenters to secure their infrastructure.  We would like to see a day that no one will need to send an email to their friends to say: “I didn’t send that spam!”  

 

What is your story and resolution regarding security issues in cyber space and datacenters? 

Ever since Johnny Carson introduced his Top Ten format on the Tonight Show, I have found myself a big fan of this format. (Note: I guess the origin of the top 10 format may not have been new with the Tonight Show but it was my first exposure – I guess I date myself some ).

 

Many times the top 10 format is fun and entertaining, however, my topic today is a little more serious. In January of 2009, I found Tech Republic’s “10 Reasons To Purchase New Hardware During A Recession” on a ZDNet Blog

 

  1. Equipment still wears out
  2. Productivity becomes paramount
  3. Downtime is expensive
  4. Competition suffers too
  5. Manufacturers offer discounts
  6. Consultants more willing to negotiate
  7. Running older hardware longer costs more
  8. Interrupting purchasing cycles is expensive
  9. New applications require greater resources
  10. Employee retention remains a consideration

 

As I spend my days talking IT managers around the world, many of these items resonate with me and whether you are investing to support continued daily business operations (existing hardware is a limiter for you), to improve IT efficiency (reduce operating costs of aging install base) or to business competitiveness (offer new services before your competition does) … many of these Tech Republic Top 10 reasons (see below) were at the heart of their investment strategy.  

 

So before you cut your IT budget in response to economic conditions, consider if purpose driven IT investments might deliver you a competitive advantage by enabling you and your company to do more with less.

 

Tune in next few weeks for a my two part series covering the benefits of IT investments in specific business environments

ð      Small – Medium Business: Why Buy for the Small Guy?

ð      Large Enterprise: Why Buy for the Big Guy?

 

PS: Do you have a favorite top 10 (personal or business)?

 

Chris

Go Beyond the Kernel:

Refocusing HPC Benchmarking on Total Application Performance

 

Want to improve application performance by 10x or 100x? Few HPC customers would say no. Yet in some cases, the promises of tremendous performance improvements from accelerators, attached processors, field-programmable gate arrays, and the like evaporate when total application performance is evaluated. Benchmarks that focus on kernel or even partial application performance provide incomplete picture with respect to the impact on total application benchmarking.  While difficult, HPC customers should look to test total application performance.

 

Why benchmark?

Benchmarking is an essential means for helping end users choose and configure HPC systems. An end user has a problem and needs to know the best way to solve it. More specifically, the end user has a specific workload to run and needs to find hardware that can deliver the best performance, reliability, application portability, and ease of application maintenance. As Purdue University researchers wrote in a recent IEEE article that argued for real application benchmarking, an HPC benchmark should, among other things, produce metrics that help customers evaluate the overall or total time to solution for their problems.

 

The claim of a 10x to 100x improvement from a particular product can easily grab someone’s attention. But what does that 10x measurement really mean? In many cases, these claims are derived from kernel or partial application benchmarking, which might fail to tell the whole story. While an increase in floating-point performance or the addition of a CPU accelerator could deliver a significant improvement for one kernel, the total application improvement depends on additional HPC system elements. As one participant argued in a recent HPC conference reported by IDC, solution time can be represented as an equation:

 

Solution time = processing time + memory time + communication time + I/O time – all four combine to form total application time.  The Caveat Emptor is to make sure you analyze your application; understanding what you are measuring and ensuring that you have the balanced architecture to deliver the best performance.  I am biased but I think Intel’s soon to be announced Nehalem processor delivers just that. 

 

Kernel benchmarking has its place, but benchmarking total (or “real”) application performance is critical for accurately evaluating HPC systems.

Given the recent intense focus in the industry around data center power management and the furious pace of the adoption of virtualization, it is remarkable that the subject of power management in virtualized environments has received relatively little attention.

 

It is fair to say that power management technology has not caught with virtualization.

 

Here are a few thoughts on this particular subject, which I intend to elaborate in subsequent transmittals.

 

For historical reasons the power management technology available today had its inception in the physical world where watts consumed in a server can be traced to the watts that came through the power utility feeds.  Unfortunately, the semantics of power in virtual  machines have yet to be comprehensively defined to industry consensus.

 

For instance, assume that the operating system running  in a virtual image decides to transition the system to the ACPI S3 state, sleep to memory.  What we have now is the state of the virtual image preserved in the image's memory with the virtual CPU turned off.

 

Assuming that the system is not paravirtualized, the operating system can't tell if it's running in a physical or virtual instance. The effect of transitioning to S3 will be purely local to the virtual machine.  If the intent of the system operator was to transition the machine to S3 to save power, it does not work this way.   The virtual machine still draws resources from the host machine and requires hypervisor attention. Transitioning the host itself to S3 may not be practical as there might be other virtual machines still running, not ready to go to sleep.

 

Consolidation is another technology for reducing data center power consumption by driving up the server utilization rates.  Consolidation for power management is a blunt tool, where applications that used to run in a physical server are now virtualized and squished into a single physical host.  The applications are sometimes strange bedfellows.  Profiling might have been done to make sure they could coexist, as a priori, static exercise with the virtual machine instances treated as black boxes. There is no attempt to look at the workload profiles inside each virtualized instance and in real time.  Power savings come from an almost wishful side effect of repackaging applications formerly running in a dedicated server into virtualized instances.

 

A capability to map power to virtual machines, in both directions, from physical to virtual and virtual to physical would be useful from an operational perspective.  The challenge is twofold, first from a monitoring perspective because there is no commonly agreed method yet to prorate host power consumption to the virtual instances running within, and second from a control perspective.  It would be useful to schedule or assign power consumption to virtual machines, allowing end users tomake a tradeoff between power and performance.  Fine grained power monitoring would allow prorating power costs to application instances, introducing useful pricing checks and balances encouraging energy consumption instead of the more common method today of hiding energy costs in the facility costs.

In 1965, Intel co-founder Gordon Moore made a prediction, popularly known as Moore's Law, stating that the number of transistors on a chip will double about every two years. Intel has kept that pace for nearly 40 years. For IT, this translates into a roadmap that enables IT to buy new servers that cost roughly the same as the previous server but performs so much better. Compare Intel’s 4 Socket MP server performance introduced in 2006 (Intel Xeon processor 7000 series) to today’s server introduced in 2008 (Intel Xeon 7400): 3x more performance throughput as measured by SPECint*_rate_base 2000*, 2.4x more ERP users as measured by SAP-SD* and 2x more database transactions as measured by TPC-C*.

Now, introduce a global economic downturn into the mix and suddenly IT is forced to cut costs and projects (i.e. delay or cancel upgrades and non-revenue generating projects). New articles start popping up from magazines like the Economist that take Moore’s Law and propose flipping it on it’s ear: instead of products providing more performance at roughly the same price, provide products that offer the same performance as IT is already experiencing, but now at a lower price. Call it “inverting Moore’s law” where IT takes the dividend it provides in dollars vs. extra performance.

So here’s something to think about: You can also “invert” Moore’s Law by making new targeted IT investments today that offer attractive payback scenarios tomorrow - giving you similar performance but at a much lower cost. With mortgage rates dropping, you may have already benefited from a rapid payback in your personal life (i.e. I recently refinanced a house down from 7% to 5.25% 30-year fixed rate that I had continuously made additional principle payments for. The ~$5k up front investment (i.e. closing costs) will be “paid back” to me after 5 months due to monthly mortgage payment savings.

Here is a server refresh example that explains how you can also get an attractive payback for your IT department.

 

Oracle Database Refresh: Let’s next look at a hypothetical example of an IT department running current Oracle Database Enterprise Edition on 12 servers purchased in early 2006 (dual-core Intel Xeon 7041 based servers introduced in 2005) and assess the total cost of ownership difference in moving to new servers.  We’ll assume the IT manager is paying per processor licensing fees for Oracle Database. We’ll compare the old server equipment to new 4-core Xeon 7440 based servers that offer up to ~3x more database performance (Xeon processor 7400 Series come in flavors of 6-core and 4-core versions).  This should enable consolidation ratios of 3:1, enabling the IT manager to reduce from 12 servers to 4 new servers. 

First the new investment: 4 New Xeon 7400 based servers at roughly $20k each = $80k.  Add another $5k for Network, Server Maintenance and Install Costs.  Remove ~$2k in tax implications associated with the expense in year 0.  Total investment ~$83k. 

Next, let’s look at the savings: The IT Manager is paying $41.8k yearly on Oracle maintenance/support costs x 12 dual-core MP servers today, that is $501k.  The 4 new quad-core servers will have larger Oracle database maintenance/support costs because of the core count ($83k x 4 servers = $334k) but this will still result in $167k SW savings each year (difference between $501k and $334k) which my calculations show about $669k savings over 4 years.  Moving from 12 to 4 servers also reduces about $72k in network, server maintenance, and utility (power/cooling) costs over 4 years as well. In addition to all of these costs savings over 4 years, my calculations show that the original investment of ~$83k has a payback of 9 months.

Targeted IT investments today can offer attractive payback scenarios and cost savings tomorrow - giving you similar performance but at a much lower cost.   Let me know what you think? 

 

 



So, building off Bob's post from September (http://communities.intel.com/openport/thread/1905), I contend that, at least from a performance perspective, with the new capabilities in the next generation of virtualized infrastructure coming this year, the answer is yes!

As we look at the availability of ESX 4.0 from VMWare and servers based on the Intel Nehalem-based Xeon servers with new VT features for CPU, chipset and IO later this year, we're not seeing any of the mainstream applications that can't be virtualized. In the past, some of the mainstream apps that (allegedly) couldn't be virtualized that we've consistently heard are SAP and other complex business processing apps, middle sized databases and large enterprise email systems like Microsoft Exchange. While it's a little early to declare victory, we're thinking the next generation of technology will be more than good enough to run these workloads in most environments. We're currently running testing on the lastest generation infrastructure software and not seeing any reason why most of these apps won't be capable of being virtualized over the next couple of years.

Anyone think differently? Why?

Note, other issues remain:

  • Even if I don't run the applications on the same physical server as other applications, is the virtual infrastructure secure and reliable enough to support these important applications?
  • And, if I try to consolidate the app with other apps, can I be guaranteed that the app won't interfere or be interfered with by other apps. Interference could be either unintentional resource contention or intential security attacks.
  • Do I have the tools and support infrastructure to do such a critical application in a virtual infrastructure.


I'm making no claims on whether these particular challenges have been solved but I would be interested in whether they are real issues for you.

What do you think?

 

Jim Blakley

 

<Note: This is a duplicate to the blog I posted at VMWorld Europe last week. I'll pull over the responses as replies to this>

Virtualization usage models continue to evolve and so does Intel technologies that support it.

Consolidation is the foundation usage model of virtualization. This we term the basic virtualization 1.0 usage model. Workload consolidation with virtualization within a physical server helps IT realize cost savings and also cost avoidance. Intel VT technologies, power efficient performance of Xeon 5400 and Xeon 7400, and reliability capabilities are essentially the building blocks for consolidation.

 

Emerging usage models or virtualization 2.0 go beyond more than cost savings and cost avoidance. They target data center efficiency and business continuity. Virtualization 2.0 hence is a step after 1.0. Once workloads are consolidated the key challenge for IT is to utilize the consolidated resources in the data center in the most efficient manner and making sure that one consolidated workload does not impact the others running on the same platform. Essentially, how can IT move the workload from one system to another is a resource pool (that now spans more than a single physical server) for efficient use of their data center resources to manage application SLAs and manage datacenter power and performance resources; and how can IT ensure business continues to run with better reliability and availability. VM mobility is a corner stone of virtualization 2.0 usage models such as load balancing and virtualization high availability. Intel technologies have also evolved to meet the requirements of these usage models. Some of the important ones are Intel VT FlexMigration and virtualization technologies for I/O (that I have blogged about in the past). Intel VT FlexMigration for instance provides investment protection to IT by allowing resource pooling of multiple generations of Intel Xeon based servers. This way IT does not have to break a resource pool to get the benefit of new energy efficient hardware when they refresh. This allows IT to constantly also refresh without impacting their datacenter architecture, as new processors typically deliver better energy efficient performance than previous generation and newer capabilities to meet the customer demands in this digital age.

 

It is also note worthy that customers buy solutions and not just hardware or software. So it is important that these solutions be efficiently enabled with the ecosystem for customers to take advantage of the benefits. In this video clip, learn from Steve (VMware CTO) and Doug (VP, Intel, Software Solutions Group), how Intel and VMware, are working together to deliver solutions that meet IT demands.

 

 

More online resources to usage models with virtualization and Intel technologies that enable these usage models efficiently:

 

Part1 virtualization usage models

Part2 virtualization usage models

Part3 Intel role in evolving usage models

 

 

 

Filter Blog

By date:
By tag: