Exotic improvements

With many VMs running on top of a hypervisor, flushing the TLB each time you switched to another VM or to the hypervisor was a costly operation. So the TLBs of the AMD Barcelona cores got a new VM specific tag, called the Address Space IDentifier (ASID). An ASID field was added to each TLB tag.  Intel followed this example more than a year later with its Nehalem CPU. The ASIDs allow the TLB to keep track of which TLB entry belongs to which VM. The result is that a VM switch does not flush the TLB. There's more info on this here.

Since context switching does not only happen between CPUs but also between processes (each process has its own virtual address space), this feature might also be handy for a “native” OS. Some CPU families already incorporated ASIDs a long time ago. Guess. Of course, the famous Alpha CPUs EV4 got this back in 1993. Operating systems will have to support this first, and we don’t expect major performance boosts. Completely exotic is the support for 1 GB pages. 2 MB instead of 4 KB pages make a lot of sense to avoid overwhelming the TLB. But 1 GB pages seem a bit exaggerated until we count our internal RAM in TBs. I was told that a few very exotic HPC applications would be sped up with this. 

And then are of course the obligatory new instructions. Six new instructions have been added to accelerate encryption/decryption using AES. The question remains if the performance improvements are worth the extra debugging time. In some rare cases they might be, but we would be lying if we would say we are enthusiastic about yet another SSE instruction.

The SKUs

Below you find an overview of Intel’s newest line-up. We did not include the AMD’s alternatives as the newest AMD Opteron “Magny-cours” will be launched in a few weeks. The current best AMD Opteron, the six-core Opteron 2435 at 2.6 GHz costs $989. It is clear that this pricing will have to be adjusted downwards with the appearance of both Magny Cours and the new six-core Xeon 5600. We expect the Opteron 2435 to compete with one of the quadcore Xeon E5620 to E5640 models.

Processor Cores TDP Clock Speed Price Notes
Intel Xeon W5680 6 130W 3.3GHz $1663  
Intel Xeon X5670 6 95W 2.93GHz $1440  
Intel Xeon X5660 6 95W 2.80GHz $1219  
Intel Xeon X5650 6 95W 2.66GHz $996  
Intel Xeon X5677 4 130W 3.46GHz $1663 Clockspeed optimized quad-core
Intel Xeon X5667 4 95W 3.06GHz $1440 Clockspeed optimized quad-core
Intel Xeon E5640 4 80W 2.66GHz $744 Clockspeed bin higher than E5540 (2.53)
Intel Xeon E5630 4 80W 2.53GHz $551 Clockspeed bin higher than E5530 (2.40)
Intel Xeon E5620 4 80W 2.40GHz $387 Clockspeed bin higher than E5540 (2.26)
Intel Xeon L5640 6 60W 2.26GHz $996 Two extra cores, same TDP as L5520
Intel Xeon L5630 4 40W 2.13GHz $551 Lowest TDP Rating (5500: 60W)
Intel Xeon L5620 4 40W 1.86GHz $440 No Hyper Threading
Intel Xeon E5507 4 80W 2.26GHz $276 45nm Nehalem, Clockspeed bin higher than E5506 (2.13)
Intel Xeon E5506 4 80W 2.13GHz $219 45nm Nehalem, Clockspeed bin higher than E5505 (2.00)
Intel Xeon E5503 4 80W 2.00 $188 45nm Nehalem, Clockspeed bin higher than E5502 (1.86)


Pricing stays the same as the Xeon X5500 series. Intel adjusted its Xeon lineup to better address some niche markets. For the HPC folks with poor thread scaling but with high performance demands there is the rather expensive X5677 quadcore at 3.46 GHz. The cost sensitive market is addressed by the E5620, E5630 and E5640 quadcores. Those parts get a speedbump for the same price.

One of the most interesting offerings is the L5630. Chances are high that you are not quickly CPU power limited once you buy a server based on Westmere cores. One CPU can still cope with 8 threads, realistically address up to 72 GB of RAM per CPU (144 GB maximum) and needs only 10W per core.

Index Benchmark Configuration
Comments Locked


View All Comments

  • Wireloop - Saturday, March 27, 2010 - link

    After watching vApus' result for both Intel and AMD gear, the natural conclusion drawn is that Hyper-V is more optimized for the Opteron architecture than ESX since the latter achieves a lower Geometric Mean VM rate (on that platform).

    I guess it has something to do with maneuver of data into the L3 cache which is a critical condition for high multithreaded performance on the AMD platform. If so, my kudos to Microsoft.
  • mgbell - Friday, March 19, 2010 - link

    Hey Anand,
    I think you should do set up a test pitting the Xeon line against their perspective i7 counterparts and run some workstation type tests. I would be very interested in any testing that had to do with video encoding/rendering. I am a video editor and would love to see a side by side comparison with a xeon sytem of the same speed against a core i7 system. Also just for fun turn off the second processor or turn it on so we can see what kinds of rendering benefits a second processor with 4/6 cores (8/12 threads) would gain.

  • lemonadesoda - Sunday, March 21, 2010 - link

    I very much agree. It would be interesting to run a typical "enthusiast" or "workstation" application/benchmark just to see how it compares.

    I would like to see a Cinebench R10 comparison, a Everest PhotoWorxx, and a Fritz Chess Benchmark. Possibly a video encoding benchmark too.

    A lot of enthusiasts run dual Xeons as workstations... you cant predict what software they will be running, but the above 3 tests are good general comparatives.

    There are also servers providing other services like OCR or PDF generation. These Oracle database benchmarks are useful, but represent only one type of server/workstation use.
  • damianrobertjones - Thursday, March 18, 2010 - link

    I'm sitting here at the end of and ADSL line with a fresh WIndows XP machine, all updates, new Kaspersky install.

    While waiting for an app to install I've visited this page....

    Bang. Kaspersky popped up with a warning

    Trojan downloader.java.agent.aw from www.googleadsenstats.ru/useralexey/files/gsb50.jar/Appletx.class

    Do you have something against ie8 as this doesn't happen with Opera?

  • itsmeagain - Wednesday, March 17, 2010 - link

    Any chance you could throw a couple of these in a mac pro and give us a preview?
  • Shadowmaster625 - Wednesday, March 17, 2010 - link

    The E5503 looks like the most reasonable and appealing server processor for those of us that live in the real world. Yet there are no benchmarks...
  • Lukas - Thursday, March 18, 2010 - link

    The 550x CPUs are crap. They don't have HyperThreading or TurboBoost. The only reason they exist is for a cheap entry price tag. If you don't need a lot of CPU (e.G. unvirtualized LOB software), better go with a 34xx series Xeon. A lot cheaper than the 55xx series.
  • majortom1981 - Tuesday, March 23, 2010 - link

    they also exist for government and public service contracts . We got a z600 with 4 gig ram ,1 5504 xeon, and an 80 gig 10k rpm enterprise sata drive (also nvida gpu) for $700. For just $239 i can add another 5504 .
  • pvdw - Wednesday, March 17, 2010 - link

    How come only Windows servers are being used. What about RHEL with a Tomcat or JBOSS bench (surely such exists).
  • Lukas - Thursday, March 18, 2010 - link

    Probably because the benchmarkers are not familiar with those platforms? Doing benchmarks on a platform about which you don't know enough will not give you any usable results.

Log in

Don't have an account? Sign up now