My last post generated quite a bit of discussion, some of it based on misunderstandings. In this post I'll try to make a few things more clear. In a previous post, I pointed out that there are a good indications that a dual Nehalem EP has a 40 to 100% advantage over Shanghai (depending on the application, based on the SAP and Core i7 workstation benchmarks).

If Istanbul is introduced in the early part of H2 2009, AMD will have a small window of opportunity of competing with a hex-core versus a quad-core (Intel's Nehalem EP). Time will tell of course how small, large or non-existing this window will be.

In well threaded applications, the best a "hex-core Shanghai" can do is give about a 30-40% boost to performance compared to the current Shanghai, which is most likely not enough to close the gap with the upcoming Nehalem CPU (let alone the 32 nm hex-core version). However, Istanbul is more than a hex-core Shanghai. The improved memory controller and HT-assist can lower the latency of inter-CPU syncing and increase the effective memory bandwidth. For that reason, Istanbul will do better than just "a shanghai with 2 added cores" in many applications such as SAP, OLTP databases, Virtualization scenario's and HPC. Depending on the application, Istanbul might prove to be competitive with the quad-core Nehalem. It is clear that the hex-core "Westmere" which will have a slightly improved architecture will be a different matter.

But back to the "this higher amount of bandwidth will allow the quad Istanbul to stay out of the reach of the dual Nehalem EP Xeons" comment. It is very embarrassing, and simply bad PR if a quad socket platform is beaten by a dual socket platform in any benchmark. This is something we have witnessed in the early SAP numbers. That is why I commented that the improved "uncore" will help the quad socket Istanbul to stay out of the reach of the dual Nehalem EP. I was and am not implying that people who would consider a dual Nehalem EP are suddenly going to consider a quad Istanbul.

It is clear those looking for a 4S and 2S server are in a slightly overlapping but mostly different market. Quad socket is mostly chosen for large back end applications such as OLTP databases or for virtualization consolidation. The number of DIMM slots in that case is a very important factor. However, even with the advantage of having more DIMM slots, better RAS etc., a quad socket platform that cannot outperform a dual socket platform will leave a bad taste in the mouth of potential buyers. It is important that there is a minimal performance advantage.

The fact that the performance/power ratio of such a quad server will be worse than a dual socket server is an entirely different discussion. IBM's market research (see the picture below) shows which form factor is bought mostly for consolidating VMs. As you can see it comes down to some people being convinced that a number of 4-socket rack servers is the best way, others are firm believers that about twice as much low power 2-socket blades is the way to go. It is very hard to convince the latter or former group to switch sides and that is why I feel that 2S and 4S servers are mostly in different markets.

In many cases, the number of virtual machines you can consolidate on one physical server is mostly a function of the amount of RAM. If the number of DIMM slots allows you to consolidate twice as many virtual machines on the quad socket machine, the consumed energy might be better than using two DP machines with the same number of DIMMs.

So despite the fact that the two DP machines have a lot more CPU power, the "scale up" buyers still prefer to go for a large box with more memory; they are not limited by raw CPU power, but by the amount of RAM that they can put in this server. It is these people that AMD will target with their 4S platform, a platform which has - especially for virtualization - a number of advantages over the current Intel 4S "Dunnington" platform... at least until Intel's octal-core arrives. Whether you choose the 2S blades or 4S rack servers depends on whether you believe in the "scale up" or "scale out" philosophy.

The conclusion is that many 4S rack servers are not only bought for raw CPU performance, but for the amount of RAM, their RAS features, and so on. However, it is clear that a 4S server should still outperform 2S servers so that the group of buyers who are believers in the "scale up" philosophy feel good about their purchase.

Comments Locked


View All Comments

  • smith1795 - Thursday, June 4, 2009 - link">">
  • Casper42 - Friday, February 27, 2009 - link

    So perhaps you care to include the so called numbers that IBM has?

    Your comment "Yes, 4S blades exist, but as IBM's numbers show, they are not on the radar when it comes to consolidating VMs." has no merit.

    The Chart simply says "Mostly Racks". That means there are some blades (and "some" in the grand scale of things is still fairly mind boggling).

    The fact is I saw a quote come across my desk for another group of IT people at my office that was for 16 x HP BL680c maxed out with each server containing 4 x Xeon 7450s and 128GB of RAM. They were also able to put 8 Gig NICs and 2 4Gbps FC ports in each blade. This was 1 order of a large group of orders for a new DataCenter that has been designed from the ground up to be treated as a hosting facility and VMs are a large portion of that technology stack.

    So while I would tend to agree that 4P Servers are "mostly" Rack Models, to say they are "not on the radar" is a fallacy.
    You also need to give the Blades a little more time. Rack servers have been around for a very long time. Blades only reached a really mature level a few years ago. Companies usually don't replace their infrastructure, especially the big servers hosting Databases, SAP, VMware and such as you mentioned, but every 4-5 years.

    So if you have some real numbers, show them.
    But keep the conjecture to a minimum if you don't.
  • JohanAnandtech - Friday, February 27, 2009 - link

    Fair enough. I removed that sentence, as it will only detract from the real point I am making in this blogpost.
  • QChronoD - Friday, February 27, 2009 - link

    I'm confused as to what the overall point was....

    Are you trying to say that although AMD's 4S system may be somewhat faster than the Intel 2S, the main reason that you foresee people picking the 4S is due to a vastly increased amount of memory capacity and design continuity with current installations?

    I thought that the newest Intel server chips were using the same architecture as the Core i7? Wouldn't that make it fairly trivial for Intel to use those same chips (with their overwhelming per socket performance) in a 4S design with similar memory capacity?
  • JohanAnandtech - Saturday, February 28, 2009 - link


    1. if 6core Istanbul is launched early, it will have a window of opportunity to keep up with 4core Nehalem. This will allow AMD to be competitive (depending on the app) even in 2S.

    2. Dual Nehalem (at 2.93 GHz) will be as fast/almost as fast as quad Shanghai (at 2.7 GHz) in some apps. Quad Istanbul will open up gap, which is important for AMD to keep the AMD marketshare in 4S, which is very profitable market. This will make AMD very competitive in 4S until Beckton arrives (H1 2010).
  • tshen83 - Saturday, February 28, 2009 - link


    1. Don't start your statement with "if", especially "if" the if is about AMD launching a product early.

    2. You still don't get it. Comparing 4S to 2S Nehalem is flawed in the first place. The fact that 2S is holding up is the miracle of the decade. Quad Istanbul is unlikely to open up the gap significantly to make AMD's 4S solution attractive.

    I really want to find out what you would say "if" AMD went bankrupt in 2009 or early 2010, which is highly probable.
  • winterspan - Saturday, February 28, 2009 - link

    Yes, but the 4-socket version of Nehalem/i7 is totally different from two socket/DP version. It will be an eight-core monolithic monster with 4x QPI links and 4x memory channels...
  • ssj4Gogeta - Friday, February 27, 2009 - link

    Yes, 4S Nehalem _will_ be launched, but it'll be launched later.

Log in

Don't have an account? Sign up now