Home > Intel® IT Galaxy > United Kingdom > Blog > Tags > nehalem

UK

6 Posts tagged with the nehalem tag

Clarksfield Mobile Nehalem Intel® Core™i7 mobile: a proper piece of Kitt.

 

Mobility is one of things that Intel does best. It’s shameless self-congratulation for me to say it but it is a revolution that Intel has really had a large hand in delivering. I am so insanely grateful that I have a work laptop that is less than an inch thick, weighs very little (even with an extra battery) and is packed with good connectivity and other features such as vPro for manageability. It’s cool and I love it – when I’m given cause to think about it.

When I’m not made to think about it, I do what most of us do: take it for granted. No real shame in that at all, but it presents a problem of sorts when it comes to events like IDF where people come with expectations of the next big thing in mobile computing. Improvement is taken for granted, and the weight of expectation now hangs in the auditorium – what will Intel bring to the mobility party this year?

Well hopefully we did not disappoint with the announcement of a new mobile CPU and platform, even if in the space of a few slides it seemed to be referred to by three different monikers: mobile Nehalem, Clarksfield and Core i7 mobile. All three are correct of course, and whoever wrote those slides please take note: Core i7 mobile would have been fine and consistent. But let that not detract from a great proc, and one that packs a punch: 8 threads and turbo boost capability of 9 clock speed bins: over 1GHz - yes, 1GHz - to you and me.

But how to use such computing power? Well, the keynote that followed the mobility session was software and one of the first demonstrations of the power of threaded software was a demo of Cakewalk, an application that will take all the threads you can give it. And then some. Intel has been talking up its rock stars recently in case you missed it, but Cakewalk was able to boast actual rock stars. Can’t argue with that. My description here will not do the demo, the software, or the end product justice, but to think that on a notebook (or desktop for that matter) you can layer 140 different tracks for one song, and do it all real time on a Core i7-based machine. That is rock and roll; end of.

So Core™i7 mobile is here and I can’t wait to get my hands on one. A laptop with a quad-core processor that has over a gigahertz of turbo clock speed is compelling stuff and its 32nm cousins promise even more goodness when they take their bow next year. But for now, I’m happy that this thing is finally here – this is the downside of knowing these thin s in advance, you get to wanting one way before it’s possible. When the time comes I’ll be using it for photography rather than playing at being Brian Eno, but that’s the beauty of personal computing: choose what you want to go and do, go and do it. On a laptop.

0 Comments Permalink

Yes, you read that correctly, and no, nothing to do with Golden Gate.

Welcome to the world of embeddded.

I have a slightly absurd fondness for the Intel® Xeon® processor, one that goes beyond mere enthusiasm for an Intel product from an Intel employee. Intel exec Mooly Eden joked here earlier today that we are addicted to the internet. He said it in jest but he is probably not too far from the truth. After all we’ve seen surveys suggesting that more people would rather give up sex before the internet. There are 100,000 new websites springing into life every day so the internet is a growth phenomenon that, even after all this time looks like it is only getting started in the growth stakes. Server processors such as Xeon power the internet, and I get passionate about the fact that people who love the internet should know more about how it is delivered to them and the masses and masses of servers that exist to sate their desire for content. And Servers have Xeons in them for the large part.

Now, of course the hordes of Amazon shoppers and Youtubers couldn’t care less about what lies behind the internet processing and storing all the data and that will never change. Deep down I know that. So not many people may care about the one of the few more esoteric products to emerge from IDF, a product currently codenamed Jasper Forest. So what is this intriguing new product and where can we expect to see it do its stuff? It is an addition to the Intel(r) Xeon(r) Processor family designed for use in embedded applications, principally storage. The buzzwords here (among many) have been innovation and integration. Intel is taking care of the integration piece (and has been doing so for some time) so that developer can take care of the innovation side – not that this product is not innovative. CPU for CPU it actually has more bells and whistles than its standard server CPU sibling, boasting PCI Express 2.0, I/O virtualisation, RAID 5 and 6 and a non-transparent bridge, all integrated (that word again) into the silicon.

I’m here at IDF with some very knowledgeable technology journalists; we know what PCI e is; we know what I/O virtualisation is; we know what RAID is. None of us though, not even the Intellite among us, had any idea what a non-transparent bridge was, or what it was for. So I set to finding out – or getting others to find out. Turns out that in the embedded applications where such a product is used, you’re typically trying to connect together multiple devices, and because PCIe is now integrated into the CPU, it enables developers to eliminate a separate bridging component from the board that would fulfil this function. Okay, maybe not as intriguing as you might have hoped for, but a cost saving is a cost saving, and in a world where performance-per-watt-per-inch governs your design processes, this is a big deal.

This world is entirely unglamorous and completely mystifying to most consumers, but they will depend on devices that contain CPUs such as these every time they buy something on the internet or check their bank balance. In 2010, when Jasper Forest is released to the world, it will be with little fanfare - save its 5 minutes of fame here in San Francisco. It is an example of innovation and integration that goes entirely unnoticed, but that is the way it goes for an embedded CPU sometimes.

0 Comments Permalink

Ever find yourself in a new location staring hopelessly at a map, wondering where you are?  Then to make matters worse, you call someone on your cell phone and can’t describe where you are so they can help? I think we’ve all been there more than once…

Since the Intel Xeon® 5500 processors launched back in March, I’ve been getting a bunch of questions about DDR3 memory and how best to configure your server platforms to optimize performance.  Many times, folks are having a hard time just getting the conversation started, so here are a couple of tips to get you going.  The good thing is that DDR3 memory picks up where DDR2 memory leaves off in terms of speed, so you know you’ll be moving forward!

  1. Figure out how much memory you need.  With multi-core CPUs now mainstream in servers, you need enough memory to keep these compute engines fed.  One metric you might look at is “GB per CPU core” or “GB per socket” for your existing servers, and then project your memory requirements from there.

  1. Start with DDR3 1066 memory, as that will deliver a good balance of memory performance and capacity. 

ð        If you need more bandwidth (and willing to give up some capacity), use DDR3 1333

ð        If you need maximum capacity (and willing to give up some bandwidth), use DDR3 800

  1. Match your CPU to your memory speed because the faster memory does require a faster processor.  Check out page 11 of the product brief for the quick reference table.

  1. Wherever possible, fill up as many memory channels as possible, and populate all channels evenly (same type, size and number of DIMMs). 

ð        Most two-socket Xeon® 5500 platforms will have a total of 6 memory channels, so aligning your memory requirements to a multiple of 6 GB will optimize memory performance for most application environments.  

ð        However, you can mix/match memory types if your requirements call for something that is not a multiple of 6.

  1. For Server application environments, always go with ECC supported memory.  Decide between Registered (RDIMM) and Unbuffered DIMMs with ECC (UDIMM ECC).

ð        RDIMM provide greatest flexibility across DIMM sizes and availability

ð        UDIMM ECC provide a lower cost alternative if you are using 1 GB or 2 GB DIMMs

There are many, many memory configurations possible for the Xeon 5500 platforms, each offering tradeoffs for performance, power, cost, bandwidth, and RAS.  You will still want to check with your system vendor on the specifics, such as memory configurations and DIMM types and options supported for a given server, but hopefully this helps you pointed in the right direction.

If you still need some more help, ask me a question on this blog!

0 Comments Permalink

When the Xeon 5500 platforms launched earlier in the year, we also introduced a server refresh ROI estimator tool to help IT managers make sense of the significant OpEx savings they can achieve by making targeted investments in new server hardware.

We know that server deployment and refresh plans vary widely from customer to customer, so we needed a robust and interactive to help you model your savings opportunity, regardless of your situation.  This tool delivers just that by taking the knowledge of Intel IT and industry leading ROI and TCO consultant Alinean and putting it into an easy-to-use tool!

We identified and were able to model eleven cost and savings categories (both pluses and minuses) in the Server Refresh ROI calculation and make these cost category assumptions able to be included, excluded or modified by you.  You can model and view scenario output real time and print/email reports to share with others.

To share some numbers from the United States, in the first 3 months, there has been nearly 4,000 users of the ROI estimator, and of those users, almost 800 users have printed reports to share with others in their organizations. Here’s some of the encouraging feedback they’ve been hearing from their customers:

·       CIO for major US hospital: “This would help my IT staff justify the financial value of the technology investment they are proposing. This has been a barrier to freeing up capital internally”

·       IT Manager for major US bank: “I used to have regular funding for technology refresh projects. It was a given for my budget.  However, with the increased constraints on capital, I now have to justify this type of spending”

·       Technology Sales Consultant: “This tool helped me work better with my customer to gain a deeper understanding of their server environment and allowed us to jointly identify high ROI investments to improve their infrastructure”

Good news is the tool has continued to evolve based on feedback from the multiple customer engagements to date, and as a result, we have just released an updated version.  Check them out:

Tool Training – How to Use: We heard that the benefits of using the Savings Refresh Estimator spanned many functional roles, making us realize that the use models for this type of tool and what users were looking for would vary dramatically from person to person.  We have a pdf training guide today that can help you get started now.

PowerPoint Output: What would we do without PowerPoint? J We received feedback on the desire to make the output of this tool more sharable inside IT organizations and with business partners in a PowerPoint format as a way to communicate the opportunity and benefits for server refresh investment.  So, we now have a PowerPoint output option in the reports section that breaks down the benefits of server refresh for a variety of audiences from executive staff to facilities to finance.  Everyone inside your business can benefit from server refresh and now you can show them how.

Secure Analysis: We received feedback that many users wanted access off-line either as a way to use in meetings when connectivity was challenged or to protect internal data from exposure online.  We now have the ability for you to run the tool on your laptop to support these use models.

More … More … More Functionality. We heard lots of requests and ideas to expand the level of functionality and analysis capabilities.  We have to balance scope, complexity Keep these requests coming.  The following changes are incorporated into today’s estimator.

·       Virtualization to Virtualization Refresh Scenario – now included

·       Virtualization Loading: Can edit and change VM/server new and old

·       Custom Performance Data – enter you own performance data to better model what you expect to see in your biz

·       Depreciation Cycle – no longer fixed at 4yrs .. can adjust

·       Memory Sizing: information added to allow user analysis

·       Processor Description: allows user to cross reference data to other more familiar terminology.

I encourage you to check out the tool and let us know how it helps you get a better handle on the benefits of server refresh.  Feel free to respond with comments and feedback here.

0 Comments Permalink

Recently the International Supercomputing Conference (ISC09) took place in Hamburg where Intel was a platinum sponsor.  A couple of musings:

 

For Intel, it was great news - we previewed a new high end server processor code-names 'Nehalem-EX'.  In production later in the year, it will have up to 8 cores and 2 threads per core (hyper-threading), 24MB of shared cache, Integrated memory controllers to name a few of the features.  The platform will double the memory capacity - 16 DIMM's per socket,  64 DIMM's per platform (4 socket) and include advanced virtualisation features - sounds like a great platform for running VM's. 

 

But wait, there's more - we also announced an 8 socket platform.  A quick calculation means: 8 cores, 8 sockets, 2 threads per core = 128 threads.  To see what this looks like under Windows Task Manager click: http://www.youtube.com/watch?v=BQ4shSQJTd0&feature=player_embedded

 

This is a high-end server so naturally it has some advanced RAS features.  It's the first Xeon® based platform to have Machine Check Architecture Recovery.  In layman's terms, it means we can detect CPU, memory and I/O errors and then work with the O/S to correct them.  The result is the system recovers from otherwise fatal errors i.e. increased uptime.  Those familiar with Itanium will recognise this technology and coupled with the increase in performance will help IT managers reduce costs if they move away from proprietary expensive RISC based systems.  For those who want to know more about this, checkout the Intel Channel  here:  http://www.youtube.com/watch?v=ztPTYDllwiY&feature=channel_page

 

Secondly, the latest edition of the Top 500 list was published.  Amazingly, there are 399 based Intel systems in the Top 500 - that's almost 80%. We've come along way since the Pentium® Pro!.  For those not entirely familiar with the Top 500 list - it's essentially the most powerful 500 computers on the planet.  Naturally, high performance clusters/computers need more than just a powerful CPU - there's interconnects, I/O, memory latency, code optimisations and more to consider.  Probably another topic in its own right.

 

For those of you who are wondering what the latest top500 list looks like, you can find the latest list here : http://www.top500.org/lists/2009/06 (click on complete list at bottom of page).

 

Ok, so what does this have to do with an Olympic medal table I hear you cry? Well take a look - not a single entry from the UK in the Top 10 unlike our athletes who managed 4th overall in the 2008 Olympics.  In fact, India, Saudi Arabia, China, Canada, France, Japan and Switzerland all have higher ranking machines/clusters.  By the time the Olympics come to London, I'm hoping that the UK will have an entry in the Top 10.  Certainly our athletes at the Beijing Olympics showed we can be competitive on the world stage…it's now down to us engineers.

 

~Iain

2 Comments Permalink

At Intel we’re very fond of starting sentences with the word ‘so’, particularly when we’re talking to each other, and even more so if we happen to be in conversation with people external to Intel. I think we use it to buy ourselves a precious second or two of thinking time before opening our mouths and going on the record.

So… I admit it is pretty pretentious to even consider the correlation between a cornerstone of one of mankind’s greatest thinkers and something so apparently mundane as a CPU, but there are times when an idea occurs that just latches on like a  - insert your own simile here – and just will not leave you alone.

I’m not going to give you a mini bio of Aristotle the man; if you’re reading this you know how to use Wikipedia or Google. He was undeniably a very influential thinker, and in some respects, way ahead of his time, particularly when you consider he was writing/teaching in the mid 300s BC and that many of his world views were held to be universal until the enlightenment in the late 18th Century. Aristotle is also widely considered to be the father of logic, defining the need for, and the methodology of, deductive reasoning. He was the first person to really set down ways of structuring the process of thinking, of reasoning and it is not stretching the point too far that his work was the root of the subsequent developments on which computing logic is based, no matter how far removed it may appear now.

Among the most famous principles of reasoning is the set of rules he laid out in order to determine the inherent nature of an object. To fully understand it we must ask four questions, we must determine its four “causes”. The word ‘cause’ is one that seems to have resulted from translation from ancient greek, and clearly something has been lost in translation, as the word cause does not carry the same meaning as it does today, so much so that it defies succinct translation even today. Rather it is better to put it in the form of 4 questions which, if responded to help you to understand an object in a holistic fashion. Aristotle is said to have used the example of a statue, but these questions were designed to help us understand all objects. Of course, things were much simpler back in Aristotle’s day, but what sort of understanding do we get of an object that is more complex: an Intel quad-core processor for example.

The first question is: from what is it made (what is its material cause)? The main portion of the answer to this question is of course silicon. But we ought not to stop there. Take a modern day processor, it has silver, tin, hafnium (lovely hafnium!). For simplicity’s sake, let’s leave it as silicon. Next, the formal cause, or more simply put: what is it? Well… um… it’s a processor, a quad-core processor. Yes, we could argue it’s a microchip, a semiconductor or even - and I have a feeling Aristotle would like this - a logic device, but to me it is a processor pure and simple. What brought the object into being is the third question, i.e what is its efficient cause. As an Intel employee this is where you start to feel good about the company you work for. It was born out of one of the cleanest, one of the most advanced manufacturing environments in the world, an environment built to hugely exacting requirements that in some ways they are as remarkable as the devices that they are used to produce. I speak of course of a wafer fab, an Intel wafer fab. I’ve never been inside one of these buildings, so they still hold an extra mystique for me. So far, so good, if unremarkable: it’s a processor, made from silicon (and hafnium), in a fab. We probably all knew that.

So what of the fourth cause, or, as Aristotle called it its final cause? Well the answer to this is, like the device itself, infinitely complex. If we left our imagination at home, the answer to the question ‘what is it for’ could simply be left at ‘computing’, or ‘processing’. But that would be to not answer the question properly or fully. The real answer is, if you want to keep it short and sweet, “whatever you want it to be for”. It can help you do whatever you want with your PC, notebook, or server.

This is the crux of the matter. The choices are endless, or at least as endless as the variety of applications and usages that are out there. What is more, the world and its economy are more reliant on these devices than ever before. We are using them every time we search the web, every time we make an online purchase, and many do not have a so much as an inkling that we are using one. In March this year, Intel launched its latest quad-core processor, in the Intel® Xeon® processor family – the 5500 series for servers and workstations. It seems a shame that it is being introduced to a world that is not as ebullient as once it was. But in another way, these circumstances provide Nehalem EP with an opportunity. It is in times of strife that innovation comes to the fore, receiving more focus as we all count on it to deliver us from stagnation.

This is where such a processor, in tandem with a variety of applications can shine. It provides the means for obsolete hardware to be replaced at a cost which is recouped in less than a year, it provides the means for digital artists to express their ideas better and more immediately than ever before, it enables movies to be animated in 3D, it helps find new reserves of oil, and provides the horsepower to design machines that are more energy-efficient and sustainable than before. There are a wealth of documents on this site that will explain the compelling ROI in replacing old, single-core servers with new machines based on the Xeon 5500 series CPU. And what is exciting is that there are people out there who will take advantage of this supremely quick computing power combined with its intelligent performance and put it to new uses, providing a firm with a new competitive advantage. Then other firms will follow suit, and this pattern, as it snowballs, begins to haul us out of the mire. Don’t misunderstand me, not even Xeon 5500 is going to fix this economic situation singlehandedly or speedily, but history teaches us that technology comes to the fore when times are tough, and the better the technology, the more it stands out, and that those who make best use of it, establish themselves as leaders.

So to bring it back to that curious 4th cause: what is it for? Well, with so many possible answers, we can only stick our hand in to all of those notions above and pick one at random or just pick a favourite. Others will pick something that has not occurred to anyone else and will use that to build a business opportunity. My inclination is to say that it is for innovation, for IT to show its value to the business, as contributor to the bottom line. A little trite maybe, but it is true. It is also an answer derived from a certain amount of logic. Surely Aristotle would not want to argue with that.

0 Comments Permalink