Prior to the Intel Xeon X5500 Server Platforms*, measuring server power was done via expensive equipment and could only be performed in a discrete fashion. Unless you had tons of monitoring equipment to mash-up your power data - it was a tedious process. Now, using Intel DCM and Node Manager - you can pull multiple servers worth of power info to make some important power decisions in your datacenter.
First of all, you need to baseline your workload. If you're confident that you can replicate workload patterns then you've got a starting point. Otherwise, it's usually a good idea to start monitoring and looking for some cyclical patterns and/or common data points (time, power, thermals, etc) to keep track of.
In this scenario (like in my last blog) we're using a SQL workload which can be modified to run the CPU at high levels for a relatively set amount of time. The base workload runs for 7 min 30 seconds, as shown in the Intel DCM screencap below.
In this test case: Idle power for the 4 servers is 782W, and under load - the power increases to 1174W - which is a delta of 392W. This power increase occurs when work is given to the server and the P/T states react to the workload and increase power/voltage to the system to increase performance. Exactly what we've been used to seeing even since EIST was introduced several years ago.
Now, what I'll show you is something that may be very interesting in scale... I will power cap the servers by 20W each, and set the Intel DCM Power Policy to only allow 1095W for the 4 servers in the rack.
What is awesome here is that we can still finish the workload in the same 7 minutes 30 seconds. So essentially, we have saved 80W of power for each set of 4 servers and still get the same amount of work completed! In a large datacenter this can be HUGE in energy savings.
Let's do some quick math: 20W power savings per serer x 10,000 servers = 20kW power savings and you still get the work done. I hope I just helped some of you server admins get some new ideas on your next "I need a raise" talk with your manager
*your mileage may vary, so test your own workloads and report out!
Why upgrade your hardware when migrating to SAP ERP 6.0?Because it makes simple, practical, business sense that is all.SAP has identified several key reasons why customers are concerned about migration and several among them are as follows:
·Cost, Cost, Cost
oHW infrastructure cost is highlighted as one of the key barriers of migration
·Business Justification
oIs there a compelling business reason to upgrade the hardware?
·Additional risk of business disruption
oMigration of ERP environment is complex enough…how much more risk is there when upgrading your hardware?
From a cost perspective, the perception that hardware is a barrier to migration can be easily overcome.Based on research, the hardware cost as a percentage of the overall migration cost is only about 7%.That means 93% of the cost is in licensing, consulting, etc, etc.HW costs are only the “tip of the iceberg” and the real $ investment lies elsewhere in the equation.
Is there a compelling business reason to upgrade your hardware? Well…frankly, it does not make sense not to do it.One, we showed above that the hardware investment is minimal compared to SW licensing, consulting, service, etc.Two, the hardware requirements of ERP 6.0 are significantly higher than previous versions. ERP 6.0 requires up to 2.5x more CPU performance, 2.5x more memory and 1.5x more I/O!You will need the increased performance and scalability that Intel provides in our microprocessors.While the ERP performance requirements have increased 2.5x, Intel performance with SAP has increased 10X!Oh, btw…energy efficiency does matter and in your new ERP environment you will be able to consolidate servers and save on power and cooling costs.TCO will be significantly reduced and from hardware investment standpoint, you are likely going to recover the cost of the servers in a very reasonable timeframe.
From my discussions with the IT community, their major concern and number one focus area is to prevent business disruption and downtime.This costs companies real and significant money.The fact is that an ERP migration is a complex enough project managing the strategic, functional and technical portions.Adding a server infrastructure change increases fundamental risk.But, the key here is that it is done often and done successfully.Intel IT has published several whitepapers on the subject and communicated “Best Known Methods” to minimize that risk.A quick summary is inserted here:
•Upgrade from SAP* ERP version 4.7 to 6.0, change the DBMS, and perform a Unicode* conversion as well as a hardware upgrade
•Minimize downtime
Benefit to Intel IT:
•SAP ERP 6.0 improves Intel supportability
•Increases ease of integration to SAP NetWeaver* 7.1 Suite
•Provides access to Enhancement Packs and Enterprise Services
•Intel® Itanium®-based servers provide access to 128 GB of memory for database and SAP operations and significantly increased performance from true 64-bit processing
Key Results:
•Reduced downtime of upgrade by 50% by using Intel Architecture
In summary,upgrading your server infrastructure when migrating your ERP environment is a very, very complex task, but form a business perspective, it should be fairly easy to see the true benefits from combining the ERP migration and hardware upgrade at the same time.
Every day, Intel® technology and platforms help companies solve business problems and challenges. Here are a few of the growing number of stories and reasons for choosing Intel processors and technology.
Winning: Humana – Healthcare product and services company
Humana continues to refresh its infrastructure with more powerful, energy-efficient technologies. For Humana, technology is vital for providing information and a full array of health benefit services to members. To replace an outdated facility, the company worked with Intel to design a state-of-the-art data center with a compact, energy-efficient infrastructure that could deliver flexibility and scalability.
·The Intel processor–based virtualized environment helps IT deploy new services quickly and ensure high availability.
·Humana added 25 percent more servers in 56 percent of the previous space while decreasing data center power consumption by 16 percent.
Winning: Emerson Electronics
Emerson reshapes its IT infrastructure for future growth, consolidating approximately 135 data centers down to four using Intel® technology–based servers
·3,600 physical servers are eliminated by virtualizing on Intel processor–based blade servers, for 18:1 consolidation worldwide
·Power-saving processors help make Emerson’s new global production data center in St. Louis 31 percent more energy efficient than traditional data centers
Winning: Türkiye Finans Katılım Bank
Leading Turkish Financial Institution Drives Better Growth and Services with Intel®Technology. Türkiye Finans Katılım Bank makes use of the online Intel Xeon processor-based Server Refresh Savings Estimator
Intel® Xeon® processor-based Server Refresh Savings Estimator¹ sets expectations clearly, predicting 80 per cent reduction in power/cooling requirements, and a 30 per cent increase in system performance already realized. With only 20 per cent of capacity currently utilised, bank has significant headroom for business expansion
Winning: Oracle IT
Oracle uses Intel® Xeon® processor 5500 series–based systems with Intel® Intelligent Power Node Manager to increase rack density and propel business growth. Refreshing its existing dual socket, quad-core servers on a three- to five-year schedule to increase processing capability and energy efficiency, but had no significant power management in use in the data centers.
More processing capability can fit within the data center power envelope because Oracle can actively manage power consumption for individual servers and applications.
·Energy savings of 35 percent are projected with Intel Intelligent Power Node Manager, for reduced operating expenses
·50 percent more servers per rack saves data center space and enables more growth while keeping costs low
Winning: DataPipe®
DataPipe® retains a competitive edge by designing a new facility and refreshing existing data centers with cutting-edge technology that can deliver outstanding processing performance for a broad range of customer applications. Low-voltage Intel® Xeon® processors help DataPipe create a dense, energy-efficient infrastructure for managed IT services.
As I’m new to The Server Room, I offer this brief introduction:I am a marketing manager in Intel’s Software and Services Group – looking after Intel’s collaborative marketing efforts with virtualization solution providers.
A couple weeks ago, Ken Lloyd blogged about the incredible changes in compute capability and performance brought by the Nehalem microarchitecture – and gave credit to the advances in software, too.I’d like to take the conversation a step further:did you know that the launch of VMware™ vSphere 4.0 in April 2009 represented a milestone of collaborative development?The combination of VMware vSphere and Intel Xeon processor 5500 based systems delivers astonishing performance in part because it is the result of a full cycle of collaboration.
Intel has a well established rhythm of technology innovation – and a lot of really smart architects who know a thing or two about cpu design – but we get innovative ideas from the outside, too.Over the years of the VMware alliance, Intel has received (and acted on) many requests for small changes in cpu circuitry…changes that would make virtualizing the cpu easier, more efficient, or add capability. A whole raft of hardware optimizations for virtualization were included in the Nehalem architecture. As Intel started to deliver early silicon for Xeon 5500 based platforms, Intel software engineers worked closely with VMware engineers – optimizing vSphere code to take advantage of the new hardware features to improve performance, increase efficiency, and add new functionality.The results?Check out this video from the launch of VMware™ vSphere 4.0 to see for yourself what “better together” really means.And the cycle continues – what can you imagine in the next round of collaborative innovation??
These are dog years for servers.Pretty much every year Intel introduces a new Xeon processor.Those who have heard the story recognize this as the Tic Tock model.On Tic years the manufacturing process is updated, on Tock years the chip architecture is updated.Every year customers get a boost in performance, and often a cut in power.Typically this boost is in the 50% neighborhood, enough to make it worth the upgrade, and still achievable by engineering teams on a two year cycle.Except, we are in dog years.
The Nehalem – Xeon 5500 – processor broke all prior boundaries on single generation performance gain.Delivering two to three times the compute capacity of the Xeon 5400 (Harpertown) generation.This is a big change, probably a once in a lifetime change – unless that quantum thing happens in my lifetime.Roughly a 10X performance boost in less than 5 years.
During this same five years we have seen virtualization technology go from a lab project – something for test and dev – to mainstream data center process.In 2005 it would have been heresy to suggest virtualizing the corporate ERP.At that point virtualization overhead on the server could be as high as 25% and the entire server was needed to do “real work”.Fast forward to today.Virtualization technology in both the hypervisor and processor have reduced overhead to only a few percent, AND servers are 10X faster.Not only can you virtualize the ERP, you are irresponsibly wasting resources if you do not.Unless your ERP demands have grown 10X in 5 years, your ERP alone won’t even make a new Xeon 5500 system sweat.
If this advancement wasn’t enough, the announcements last month from Intel about the coming Xeon 7500 (4+ socket) processor were amazing.All the benefits of the Xeon 5500, but on steroids.The new biggest leap ever.With up to eight cores and four memory channels per socket, this is a monster.Your ERP system will be barely a blip in perfmon.It isn’t unreasonable that an entire data center for a SMB business could be virtualized onto one of these beasts.And, how big is a Xeon 7500 server?My guess is about the size of a breadbox…
At Intel, we not only pack a lot of performance in a small form factor, we also pack a lot of great demos and theater presentations into our booth at Oracle OpenWorld in San Francisco (South Moscone, booth #1621).We have 5 demos from 5 of our customers—Cisco, Dell, HP, IBM, and Sun—and 3 other demos showcasing Wind River, Intel’s SOA Expressway product, and last, but certainly not least, Intel’s amazing and upcoming Nehalem-EX processor, which you heard Michael Dell praise in his keynote this morning.
Over the course of the three days of our booth at OOW (Monday through Wednesday this week), we will have over 35 brief presentations that will help you plan your requirements for your next generation data center.They are short and sweet, and you can ask all the questions you want.If you simply attend a presentation and get a few more stamps form our demo stations, you can enter to win one of two netbooks that will be given away at the end of each day.
Outside of our booth, you may find us presenting in various partners’ booths and we hope to see you in a session we are having later today (see info below).We had an amazing session yesterday from resident Intel genius, Steve Shaw.The huge room was filled to capacity.At this other session today we will be giving away a netbook.Here are the logistics for today’s session:
ID#: S309892
Title: Ten Ways to Improve J2EE Application Performance on Multicore Systems
Track: Oracle Develop: Enterprise Java and Oracle WebLogic
Date: 13-OCT-09
Time: 17:30 - 18:30
Venue: Hilton Hotel
Room: Yosemite B
We hope to see you around somewhere at Oracle OpenWorld, but if for some reason we miss you entirely, please visit www.intel.com/server for more info on Intel’s fantastic products.Also, please visit Channel Intel on youtube for some videos from the event.
I’m a bit late in relaying my thoughts from Intel’s Developer Forum (IDF), but there was definitely some excitement around virtualization and high performance networking that I wanted to get the word out about!
In the past I’ve shared some details about SR-IOV and the advantages you can gain by being able to present virtual LAN hardware to each Virtual Machine (VM), effectively avoiding the Hypervisor when presenting virtual devices to each VM. The advantage of being able to do this is clear: The less interaction in the networking stack there is from the hypervisor, the less processing overhead is required for the system process the data.
That’s all good because if you have a dual 10 Gigabit adapter, you can segregate those two physical pipes into perhaps 16 virtual pipes that get exposed to 16 VMs. By segregating these LAN pipes at the hardware level with SR-IOV instead of using Hypervisor switching, the performance gains in both CPU utilization as well as maximum total throughput can be very large. There were several demos at IDF with various configurations, but reductions in CPU utilization of 40% were possible coupled with dramatic improvement in throughput!
But there is unfortunately one minor complication that I didn’t mention in my last post on the topic of SR-IOV. There is the little fact that when VMs move between physical boxes (a usage that is highly desired and commonplace these days) you run into some problems with this SR-IOV capability. When the hypervisor owned the network hardware abstraction, the performance was worse, but the functionality was better because you could seamlessly migrate from one box to another and the virtualization application would handle the details. But with SR-IOV, a new layer needs to be added so that the direct hardware connection between the VM and the LAN hardware can be moved to a new box.
The really exciting part of IDF demos that I saw was the demonstration not just of the SR-IOV functionality on multiple hardware and virtualization configurations, but that these demonstrations also showed updated software from two virtualization vendors allowing mobility of the VMs while supporting SR-IOV!
There was a demo on Dell systems showing this fully functional SR-IOV implementation with Citrix’s Virtualization suite. There were two separate demonstrations on Dell systems, with VMWare displaying their new Network Plug-In Architecture (NPIA) solution that allows for the migration of SR-IOV connected VMs seamlessly between servers.
For those hungry for more detail, I’ve included the three SR-IOV demonstration videos here:
The first is the Citrix demonstration on Dell and Intel hardware of SR-IOV with VM mobility:
These next two are two videos are demos on Dell and Intel hardware with VMWare and their NPIA software implementation.
Each virtualization demo shows the massive performance benefits under various workloads when moving from Hypervisor based LAN segregation to SR-IOV implementation. But most importantly, each demonstration proves out the capability to migrate VMs between physical hardware. The only system hardware requirement is that the server itself supports VT-d. If the networking hardware in the newly migrated-to box supports SR-IOV you get better performance, and if not, the solution falls back on the legacy Hypervisor virtualization. Backwards compatibility is maintained!
I didn’t get firm details on when this full support for SR-IOV and migration will be available in Citrix and VMWare’s releases, but the demos looked pretty clean, and hopefully these suites will be available soon with this new functionality. The LAN and Server hardware ecosystems are ready today, and it looks like the software vendors are just around the corner. Virtualization momentum continues!
While virtualization was the big takeaway for me from IDF, there were also several other interesting demos for us networking hounds. I’ve linked a couple videos of them below for anyone still thirsting for more of the latest networking technology and performance details!
The first video is a demonstration of Intel’s 82599 10 Gigabit Ethernet-based adapter card with Fiber Channel over Ethernet (FCoE) support. Storage and Ethernet together at last!
The second video is a demonstration of Intel’s NetEffect 10 Gigabit Ethernet card publishing 1 million messages per second in a simulated NYSE floor trading scenario. Oh yeah, only 35uS of latency. That isfast.
So although I am two weeks after IDF, I hope some of you got a little taste of the networking excitement that took place. Industry wide, hardware and software vendors alike are delivering ultra high performance low latency applications for the financial services industry, as well as mainstream performance increases for virtualization. The performance and technology beat moves forward. Exciting times!
I had never heard of a cloud forest before I went on vacation this past June to Costa Rica where I spent time at the Villa Blanca Resort. Even when we arrived at Villa Blanca, I have to admit I was a little confused. I had expected to see a forest in the clouds, however, I saw a beautiful hill side scattered with a few trees.
However, when we went on our walking tour the next morning, our tour guide walked up to one of the larger trees and says “Welcome to the Cloud Forest. This tree is a perfect example of a cloud forest”. As I looked more closely at the tree, I was amazed at what I saw - this single tree was host to thousands of species of both plants and animals.
Nature is extremely efficient in it’s use of a cloud forest. Likewise, Cloud Computing is an extremely efficient use of computing resources. It is for this reason that Intel IT has developed an enterprise cloud computing strategy focused on building an internal cloud to boost efficiency and flexibility inside of our IT infrastructure. This internal cloud strategy is closely linked to our current use and accelerated plans for virtualization. In addition Intel IT uses the external cloud services selectively for certain applications.
Additionally, we are exploring using rich mobile clients with cloud computing models moving forward to better meet the needs of an ever changing user base, consumerization trends and the need to maintain highly efficient, secure information and application delivery to employees.
To find more discussion, blogs and content relating to cloud computing – in the enterprise or corporate client solution areas – take advantage of these resources.
And .. if you ever have the chance to visit Costa Rica .. visit the cloud forest. It was worth the trip.
I wanted to follow-up on the pre-IDF blog I wrote and what I and Sean conveyed regarding comprehensive IO optimization for enterprise cloud (based on virtualization infrastructure). A blog I owed to those who could not attend the IDF session.
In the last blog we identified 4 important vectors that drive I/O evolution.
1)Balanced system that maps to the increases in CPU performance
2)Scalability
3)Unified fabric
4)Security
In my view I state it an evolution as I feel that is the natural state things will head towards in the (near) future.
In a cloud environment you would anticipate automation and policies determine the consolidation possible on a system. If SSDs get broader adoption and virtualization performance increases due to hardware assists, I/O and fabric could become the bottleneck for the degree of consolidation and efficiency as it cannot map to the increased data rates from the storage and CPU performance.
Ways to address this is either to reduce or eliminate the overheads in the I/O stack caused by software emulation of devices in the VMM. VMDq is an example of a technology that can reduce the overhead or offload some of the VMM tasks through hardware assists in the NIC. Direct assignment with PCI-SIG SR-IOV support is a way to eliminate the overheads by by-passing the VMM. With SR-IOV, a single device can be divided into many logical devices known as Virtual Functions (like a pair or independent transmit receive queue). Each virtual function can be directly assigned to a VM using Intel VT-d thereby bypassing the VMM. This can work with Live VM Migration too. At IDF we showcased 4 demos of prototype SR-IOV software solutions running on Intel Xeon 5500 based hardware with prominent VMM vendors like VMware, Citrix and Redhat that have different hypervisor technology. The networking demos showcased working live migration with SR-IOV and VT-d based direct assignment. Direct assigned VMs could be even moved to an emulated mode and brought back to direct assigned mode. Intel has not only been working with software providers but also with other hardware vendors like LSI to demonstrate this capability. These technologies are as important to storage as networking particularly as SSDs gather steam. You can learn more from these blogs below on the demos.
If those multi 1GbE cables (that make your fabric look like pasta) be replaced by 10GbE and if SR-IOV and VT-d be used for performance, then it answers both the I/O performance and scalability requirements for a flexible datacenter.
Beyond those VT-d provides greater protection by allowing I/O devices to access only the memory regions allocated to them, and SR-IOV allows VMs to access only their portion of the device and restricts access to other Virtual Functions (owned by other VMs) on the I/O device or the entire device itself. Better security through better isolation.
Last but not the least of the requirements is the unified fabric. When IT can use a single I/O device for storage or for LAN traffic, the rigidity associated with provisioning of servers with some number of HBAs and some number of NICs is reduced. The I/O capacity becomes fungible and flexible. FCoE and iSCSI are key technologies in this direction. Adding capability to monitor QoS and shape the traffic makes it a good match for flexible datacenter.
Many of the technologies I discussed above (VT-d, SRIOV, FCoE, iSCSI) are here today… software such as Red Hat Enterprise Linux is already delivering the solution. In my view just a matter of time that ecosystem builds further and hardware is well tuned.
With these in perspective how do you see your datacenter shaping up?
Really good case study for a leading Turkish bank that used the Xeon ROI tool to justify their server refresh with Xeon 5500 and 7400 platforms: http://communities.intel.com/docs/DOC-4114#cf
Well, it's the 3rd and last day of IDF today and the demos have running now for 2 days. You have one moreday to check them out! So let me describe them quickly.
Dell has been a great partner for these demos. We’re showing 2 demos together in booths 709 and 711 in the Virtualization Community, using Dell’s R710 servers, based on the Xeon 5500 platform, and using Intel 82599 (Niantic) with Virtual Machine Direct Connect (VMDc). The 1st demo is with VMware and their Network Plug-In Architecture (NPA) technology.
When you visit the demo, check out the great CPU utilization as well VMotion* among heterogeneous server configurations!
The 2nd demo is delivered with Citrix, showing scalable direct assignment by using XenServer with VT-d and SR-IOV support. The overall performance is really great and live relocation of virtual machines is working nicely.
Another demo (booth 707) is delivered by Red Hat, featuring RHEL 5.4 with KVM (shipping SW with VT-d and SR-IOV support!), Neterion with their 10GbE NIC, all running on an Intel Xeon 5500 server. Look out for the performance value shown for scalable direct assignment!
And finally, a storage (not LAN) demo! Using the same combination of VMM and server, in booth 517, LSI is showing the value of scalable direct assignment for a RAID controller. The performance boost is fantastic!
Check out these demos at the IDF Showcase... I’d love to hear your impressions!
About Hugh Mercer: I am a sales development manager in Intel’s Enterprise Solution Sales group. One of my responsibilities is working with Intel’s Server Platforms Group to indentify, develop and highlight success stories around Intel’s server platforms and technologies.
Every day, Intel® technology and platforms help companies solve business problems and challenges. Here are a few of the growing number of stories and reasons for choosing Intel processors and technology.
·Implemented small server farm. Intel Xeon processor series performed more powerfully than RISC architectures.
·2010 scale out. In 2010, the university plans to implement some 400 more systems with over 20,000 cores powered by the upcoming Intel Xeon processors code-named Nehalem EX
Winning: Alvotech
Alyotech turns to Intel® Xeon® processor 5500 series to deliver insightful design improvements
·Alyotech benchmarked the new processor, developed on 45nm Hi-k next generation Intel® Core™ Microarchitecture, and increased performance by 65 percent over the previous generations, dual-core servers
Winning: Atos Origin
Intel® Xeon® processor 5500 series helps Atos Origin lower total cost of ownership of its data centre environment.
Atos Origin compared the performance of the Intel Xeon processor 5500 series with four cores to that of the previous-generation with just two cores. It found, on average, 2.4x greater transaction throughput running a web server, 1.75x running a database server and 1.25x running an email server.
Winning: Business and Decision Group
Business and Decision Group powers forward with huge virtualization project underpinned by the Intel® Xeon® processor 5500 series.
Early results showed that with the Intel Xeon processor 5500 series they could gain virtualization rates of 20:1 and with a processor load slightly below 55 percent.
Power consumption was reduced by approximately 30 percent compared to the previous generation of processors.
Winning: Onkosh.com
Intel® Xeon® processor 5500 series boots performance of unique Arabic search engine Onkosh.com
·Onkosh.com already witnessed an increase of around 20% in performance. This performance increase was possible due to the new micro-architecture with Intel Turbo Boost
·Onkosh.com is now able to grow about 300% in terms of the ability to crawl and parse new Arabic content automatically discovered on the World Wide Web.
Winning: BMW
Migration to Intel® Xeon® processor 5500 series lowers total cost of ownership and increases flexibility
·BMW Group is deploying Dell PowerEdge* servers powered by the Intel® Xeon® processor 5500 series, which will replace a RISC-based infrastructure that has much higher costs, lower performance and less flexibility
·This allowed BMW Group to increase the workload to more than 80 percent and to significantly decrease the total cost of ownership (TCO).
Winning: Société d'Exploitation des Transports de l'Agglomération Orléanaise (SETAO)
SETAO turns to Intel® Xeon® processor 5500 series to strengthen and build on its service offerings.
·Thanks to the Intel® Xeon® processor 5500 series and VMware hypervisor, SETAO is now able to provide mainframe-class quality of service and ensure easy deployment of new virtual machines and applications while reducing total cost of ownership.” Olivier Parcollet, Chief Technology Officer, SET
·SETAO estimated that it could save approximately 40 percent on energy costs due to the higher server consolidation ratio and greater CPU energy consumption management.
Take a look at this video segment from Paul Ottelini's keynote today at IDF. Very interesting on where the technology is headed and what the consumer wishes technology could do for them. Very cool stuff......
I have a confession to make… Last year was my first IDF. Ever! I had no idea then, that this year I would end up being responsible for a whole track, and sponsoring the Virtualization Community zone. I was lucky that Jake took ownership of the community zone. He assembled a great line-up of demos, from a variety of companies. It should be great, go see!
But this blog is about the Enterprise Cloud track. I set out to make it to represent a theme, rather than a collection of loosely related sessions. In my view, this required a mix of depths – an overview session to explain the concepts alongside deep technical sessions. I also thought it would be a great opportunity to gather some industry leaders beyond Intel to talk about Enterprise Cloud vision and the opportunities it presented for the developers community.
“What is this guy talking about” you must be asking yourself. “What is Enterprise Cloud? Not more hype?!” Well, I think of Enterprise Cloud as a very real vision of the place where actual IT needs meet the aspirations of the Cloud Computing hype.
The Cloud hype is based on some pretty impressive efficiencies that several companies are being told to have achieved. These companies did so by designing custom application to run in their data centers. In some of these cases the data centers and the hardware in them were even custom architected and designed to run these applications. IT wants to gain similar efficiencies. But IT can’t throw away all the legacy applications…
In comes Enterprise Cloud, where IT evolves to gain the efficiencies, without losing the legacy investments…
In the Enterprise Cloud track we’ll cover some of the key technologies that are required for this to happen.
We’ll start with an overview (session ECTS001) on Tuesday at 10:15, where Dylan and I will do an overview of key technology areas: virtualization and performance, Data Center efficiency, evolution of I/O, and security, and why they are critical for the evolution of IT. What will follow are several in-depth sessions that will cover those very topics:
·ECTS002 – will focus on Intel® Trusted Execution Technology and explain how application can protected in the Enterprise Cloud environment. Check this out in Jim's blog
·ECTS004 – will talk about technologies to improve Data Center efficiency. David covers one of those technologies here and check out his other blogs as well.
·ECTS005 – is an in-depth review of Intel’s technologies for virtualization, and will be presented by Intel Fellow Rich Uhlig.
·ECTS006 – will discuss evolution of I/O, which is necessary to enable IT to gain the desired efficiencies. RK gives an excellent preview in his blog here
·We also have a Q&A session on Tuesday evening (ECTQ001) to allow an open unscripted conversation with all the track presenters who will be around on Tuesday.
·Finally, we have a VERY exciting panel (ECTP001, on Tuesday at 5pm). Jake Smith from Intel will lead a discussion with some true industry thought leaders from Cisco, Citrix, Microsoft, Sun, and VMware. The Theme of the panel is “Enterprise Cloud – technologies, usages, and opportunities for the developers community”. This should be an exhilarating hour!
Along with a couple of labs this should a great track. See you at IDF… it all starts tomorrow!!!
I love food! Since I was a kid I’ve loved noodles, especially Italian pasta. I used to think that Spaghetti was the general name for Italian noodles. Learning how to twist Spaghetti on a fork gave a great sense of achievement and joy.
Many years later my wife and I travelled to Rome. Naturally – both of us really love food, we spent a lot of time seeking out restaurants and checking out new food. One of the wonderful dishes we had was Pappardelle (on the left) with Duck Ragu. It was my 1st encounter with Pappardelle –a very wide form of pasta. You get only a few Pappardelle on your plate but it’s still the same amount of pasta. I found it not as practical to twist the Pappardelle around my fork, so I cut them up into smaller pieces to eat them.
I’ve thought about it recently when I looked at the back of a virtualized server. Looks similar, no? J
A typical virtualized server has 8-10 1Gigabit Ethernet (1GbE) ports, and 2 Fibre Channel ports. This makes for a lot of cabling, and many add-in cards. It translates to a lot of cost, power, and complexity (and thus reliability risk) for an IT shop. As a result, there’s a lot of buzz around high-speed networks, specifically 10GbE. That technology presents the opportunity to consolidate all these 1GbE ports to a significantly smaller number of higher bandwidth, i.e. 10GbE ports. It makes for a much tidier server.
Kind of like substituting Pappardelle for SpaghettiJ
In that case, iSCSI or FCoE (Fibre Channel over Ethernet) could be used for the SAN connection, still using the same high-speed ports. Standards like Data Center Bridging (DCB) could add a lossless character to the 10GbE link to make it friendlier to FCoE.
Few new solutions though come without new challenges. The common way for VMs to share I/O devices in a today’s environment in through mediation of the hypervisor, using emulation or para-virtualization. That reduces the effective I/O bandwidth. It also becomes a fairly significant overhead to the server in its own right, reducing the available server capacity for application processing, and it adds latency. With the growing trend in IT to treat virtualization as a default deployment mode for any application, these issues become quite limiting.
We at Intel have thought that the best way to overcome these issues is by using “direct assignment”. Using the Intel® VT-d technology (launched in the Xeon 5500 platform), a VM can be assigned a dedicated I/O device. This nearly eliminates the overheard related to the hypervisor mediation I mentioned above. A side benefit is that it increases the VM to VM isolation and security. But assigning an individual I/O device to one VM is not very scalable…
This is where the PCI-SIG’s SR-IOV (Single Root I/O Virtualization) standard comes into play. This standard allows a single I/O device to present itself as multiple virtual devices. With SR-IOV, each virtual device can be assigned to a VM, adding scalability to the direct assignment model, effectively allowing the physical I/O to be shared yet with greater security and reliability.
Another challenge with the direct assignment model is related to live migration. Hypervisors have typically assumed the SW mediated IOV model. As a result, hypervisors need to be modified to adapt their live migration solutions to direct assignment.
These technologies span many different components of the server platform. Intel® VT-d is necessary, so Xeon 5500 must be used (or later platforms). SR-IOV capable I/O devices – NICs or Storage controllers, are required. BIOS must be modified, as well as hypervisor software. This is pretty heavy lifting.
So you can only imagine how excited I am to be able to showcase 4 different SR-IOV demosat IDF next week! The demos involve 2 server vendors, 3 VMM vendors – 3 different vendors implementing 3 different hypervisor architectures, and 3 different IHVs representing 2 different I/O technologies. We show the performance improvements, as well as VM live-migration. It works!
Come and see it (Booths 517, 707, 709, and 711 in the IDF showcase)!
