Posts Tagged ‘AMD’

h1

ExtremeTech: Shanghai to Istanbul

June 22, 2009

ExtremeTech runs some tests on the AMD Istanbul 6-core processor and compares the 2435 (2.6GHz) part to the 2384 (2.7GHz) part in a drop-in replacement using PassMark and Spec_JBB2005. Testing was performed in the same Supermicro system running an updated (AGESA 3.5.0.0) BIOS supporting Istanbul processors.

“Perhaps, the most tell tale result comes from the BOPS rating scored using SpecJBB2005, which simulates a server’s ability to process JAVA code. Here, there was a 20% increase in performance, with BOPS increasing from 380721 to 471440. That 20% performance boost would [definitely] be noticed on a busy server in a data center.”

- Loyd Case, ExtremeTech.com

While not as thorough as Scott Wassman’s drop-in testing at TechReport (reported earlier this month), ExtremeTech’s results and conclusions were about the same: Istanbul makes a great upgrade processor.

“It all comes down to simple math, where one has to consider the cost of the CPUs and the time needed to perform an upgrade to see if the return on investment is worthwhile. Most will find that in this case, it is…”

- Lloyd Case, ExtremeTech

“And if you have existing, compatible Socket F servers, the Istanbul Opterons should be an excellent drop-in upgrade. They’re a no-brainer, really, when one considers energy costs and per-socket/per-server software licensing fees.”

- Scott Wassman, TechReport

Both ExtremeTech and TechReport make compelling upgrade arguments in their testing. Compared to a new system architecture like Nehalem, it is logistically less disruptive – technologically and economically – to certify a CPU upgrade versus as platform replacement. After internal certification, a BIOS and CPU upgrade takes about 20-minutes per system to implement. In a virtualized datacenter where low-level differences are abstracted-away by the hypervisor certification testing should be much less invasive. Likewise, rolling upgrades in a virtualized datacenter with vMotion technology can provide a non-disruptive path from 4-core to 6-core. As Case puts it:

“Simply put, by just upgrading five servers in a data center, data center managers can eliminate the need to purchase an additional server to meet performance needs.”

- Lloyd Case, ExtremeTech

However, this “upgrade proposition” is a difficult position for AMD as it does little to sell new systems. Historically, CPU upgrades only happen in 10-15% of the installed base, making CPU sales based on BIOS/drop-in upgrades an interesting footnote. Integrators want to move new hardware with Instanbul pre-installed, not sell “upgrade packages.” Perhaps the dynamics of the new economy will drive a statistical anomoly based on the strength of the Istanbul proposition. Datacenter managers face a familiar dilemma with some new twists.


h1

Tyan: New Istanbul BIOS Released

June 16, 2009

Tyan joins the AMD 6-core release cycle with “Istanbul-enabled” BIOS releases six 2P motherboards (and related systems). We here from a reliable source that the Tyan “twin” system (B2985, 2-node, 1U) BIOS is in “alpha” testing and will be released “soon.” Here’s what’s available now:

Dual-Opteron Motherboards (2000-series)

S2912-E, S2915-E, S2927, S2927-E, S2932, S2932-E

Notes

These BIOS releases were only available from Tyan’s FTP site (ftp.tyan.com, anonymous access) until now. As of 6/17/2009 you will find them on the BIOS helper from the product web page.

Also, we were surprised to find that the AGESA for the S4985 series (quad-socket) motherboards are still version 3.3.0.3 and are likely not Istanbul-ready. We do not have the S4985 in our testing lab to confirm of deny this, but we have been told by AMD that 3.5.0.0 is the minimum Istanbul supported AGESA.

h1

First 12-core VMmark for Istanbul Appears

June 10, 2009

VMware has posted the VMmark score for the first Istanbul-based system and it’s from HP: the ProLiant DL385 G6. While it’s not at the top of the VMmark chart at 15.54@11 tiles (technically it is at the top of the 12-core benchmark list), it still shows a compelling price-performance picture.

Comparing Istanbul’s VMmark Scores

For comparison’s sake, we’ve chosen the HP DL385 G5 and HP DL380 G6 as they were configured for their VMmark tests. In the case of the ProLiant DL380 G6, we could only configure the X5560 and not the X5570 as tested so the price is actually LOWER on the DL380 G6 than the “as tested” configuration. Likewise, we chose the PC-6400 (DDR2/667, 8x8GB) memory for the DL 385 G5 versus the more expensive PC-5300 (533) memory as configured in 2008.

As configured for pricing, each system comes with processor, memory, 2-SATA drives and VMware Infrastructure Standard for 2-processors. Note that in testing, additional NIC’s, HBA, and storage are configured and such additions are not included herein. We have omitted these additional equipment features as they would be common to a deployment set and have no real influence on relative pricing.

Systems as Configured for Pricing Comparison

System Processor Speed Cores Threads Memory Speed Street
HP ProLiant DL385 G5 Opteron 2384 2.7 8 8 64 667 $10,877.00
HP ProLiant DL385 G6 Opteron 2435 2.6 12 12 64 667 $11,378.00
HP ProLiant DL380 G6 Xeon X5560* 2.93 8 16 96 1066 $30,741.00

Here’s some good news: 50% more cores for only 5% more (sound like an economic stimulus?) The comparison Nehalem-EP is nearly 3x the Istanbul system in price.

Read the rest of this entry ?

h1

Server Watch: Istanbul, G34, C32, Itanium and Nehalem-EX

May 29, 2009
Istanbul is launching in June, 2009 and will be a precursor to the G34 and C32 platforms to come in Q1/2010. To that end, AMD will be providing an overview of its next generation of Direct Connect Architecture, or DCA 2.0, which which separates Socket-F systems from G34/C32. This overview will be available as a live webcast on June 1, 2009 at 11:00AM Central Time. In advance of the announcement, AMD has (silently) reduced prices for its Opteron processors across the board. This move will place additional pressure on Intel’s Nehalem-EP systems already weakened (virtualization) price-performance.

We expect to hear more news about Istanbul’s availability in keeping with Tyan’s upcoming announcement next week. Based on current technology and economic trends, Istanbul and G34 could offer AMD a solid one-two punch to counter Intel’s relentless “tick-tock” pace. With Nehalem servers sales weak despite early expectations and compounding economic pressures, market timing may be more ideally suited for AMD’s products than Intel’s for a change. As Gartner puts it, “the timing of Nehalem is a bit off, and it probably won’t make much of an impact this year.”

In the meantime, Phil Hughes at AMD has a posted a personal reflection on Opteron’s initial launch, starting with the IBM e325 in 2003, and ending with Opteron’s impact on the Intel Itanium market by year-end (while resisting a reference to “the sinking of the Itanic“). Phil acknowledges Sun’s influence on Opteron and links to some news articles from 2003. See his full post, “The Sun Also Rises,” here… As 64-bit processors go, 2003 was much more the year of the Opteron rather than “the year of the Itanium” (as predicted by Intel’s Paul Otellini.)

Speaking of Itanium, TechWorld has an article outlining how Intel’s upcoming Nehalem-EX – with the addition of MCA technology derived from Itanium – could bring an end to the beleagered proprietary platform. TechWorld cites Insight 64 analyst Nathan Brookwood as saying the new Xeon will finally break Intel’s policy of artificially crippling of the x86 processor which has prevented Xeon from being competitive with Itanium. The 8-core, SMT-enabled EX processor was being demonstrated by IBM in an 8-socket configuration.
h1

Intel’s $1.1B Euro Slap On the Wrist, Must Sell 2.3M Chips

May 13, 2009

May 13th, 2009  – besides being my birthday – marks the day that the European Competition Commission drew a $1.1B Euro fine (about $1.4B US dollars) on Intel for going “to great lengths to cover up its anti-competitive actions” and in the process “harmed millions of European consumers.” This according to the EU commissioner Neelie Kroes, in an address in Brussels today. The fine could have been as large as $4B Euros, and will go to the EU’s annual budget – not consumers.

Commissioner Kroes was seen holding up an Intel PII/PIII processor card (SECC2) during the news conference, giving some scope to what has been a very long and drawn-out process: going back to 2000. At the heart of the matter has been Intel’s “llegal anticompetitive practices to exclude competitors from the market for computer chips called x86 central processing units (CPUs)” – namely AMD. These were apparantly manifested in behind the scenes rebates and discounts in exchange for a reduction or termination of AMD-based products.

In a press release from Intel’s President and CEO, Paul Otellini, the fined chip maker offered this defense:

Intel takes strong exception to this decision. We believe the decision is wrong and ignores the reality of a highly competitive microprocessor marketplace – characterized by constant innovation, improved product performance and lower prices. There has been absolutely zero harm to consumers. Intel will appeal.

Intel must cover their fine immediately with a bank guarantee which will stay sequestered until their appeal is either exhausted or the decision reversed. Based on EU’s hunger for this type of commercial justice, the money could be tied-up for many years. But the question remains, does Intel have a history of anti-competitive behavior beyond the test of rigorous competition?

Intel’s history tells a compelling story: the EU joins Japan (2004) and South Korea (2008) in finding Intel engaged in anti-competitive behavior. The question remains: how will the EU’s decision play in the US courts as AMD’s ongoing antitrust suit (2005) against Intel continues to unfold? Delayed until 2010 due to the lenghty list of depositions scheduled for the case, the EU’s decision will likely do more to tarnish Intel’s new “Promoting Innovation” Campaign than settle the dispute.

So what does Intel need to do to weather the EU’s wrath? In product terms, Intel needs to move 2,262,752 of its Nehalem-EP (5500-series) chips to cover the loss. Based on a predicted 40M unit replacement market in the US, thats less than 5% and it’s under 2.5% of the market if they are 2P systems. However, Intel’s promised a 9:1 value for the replacement with some estimating that number moves to 18:1 with good results for SMT (depending on the workload).

What does this mean from an Intel 5500-series sales perspective? Here’s our estimate, using Intel’s 9:1 and 18:1 math (not forgeting the 4.5:1 for the dual-core):

Nehalem Units Needed Retail Value 9:1 18:1
W5580 12,545 $20,072,000.00 0.56%
X5570 121,713 $168,694,218.00 5.48%
X5560 168,227 $197,162,044.00 7.57%
X5550 174,450 $167,123,100.00 7.85%
E5540 531,715 $395,595,960.00 23.93%
E5530 419,636 $222,407,080.00 18.88%
E5520 183,533 $68,457,809.00 8.26%
E5506 262,704 $69,879,264.00 5.91%
E5504 250,051 $56,011,424.00 5.63%
E5502 106,312 $19,986,656.00 1.20%
L5520 10,516 $5,573,480.00 0.24%
L5506 21,350 $9,031,050.00 0.96%
Total 2,262,752 $1,399,994,085.00 12.97% 73.49%

By these estimates, Intel will need to close 86.5% of the total replacement market to be able to cover the EU fines. All this assumes, of course, that they don’t offer discounts off of their “published” per-1000 chip prices. Good luck, Intel, on an exciting marketing campaign!

h1

The Cost of Benchmarks

May 8, 2009

We’ve been challenged to backup our comparison of Nehalem-EP systems to Opteron Shanghai in price performance based on prevailing VMmark scores available on VMware’s site. In earlier posts, our analysis predicted “comparable” price-performance results between Shanghai and Nehalem-EP systems based on the economics of today’s memory and processors availability:

So what we’ve done here is taken the on-line configurations of some of the benchmark competitors. To make things very simple, we’ve just configured memory and CPU as tested – no HBA or 10GE cards to skew the results. The only exception – as pointed out by our challenger – is that we’ve taken the option of using “street price” memory where “street price” is better than the server manufacturer’s memory price.

Here’s our line-up:

System Processor Qty. Speed (GHz) Speed (GHz, Opt) Memory Configuration Street Price
Inspur NF5280 X5570 2 2.93 3.2 96GB (12x8GB) DDR3 1066 $18,668.58
Dell PowerEdge R710 X5570 2 2.93 3.2 96GB (12x8GB) DDR3 1066 $16,893.00
IBM System x 3650M2 X5570 2 2.93 3.2 96GB (12x8GB) DDR3 1066 $21,546.00
Dell PowerEdge M610 X5570 2 2.93 3.2 96GB (12x8GB) DDR3 1066 $21,561.00
HP ProLiant DL370 G6 W5580 2 3.2 3.2 96GB (12x8GB) DDR3 1066 $18,636.00
Dell PowerEdge R710 X5570 2 2.93 3.2 96GB (12x8GB) DDR3 1066 $16,893.00
Dell PowerEdge R805 2384 2 2.7 2.7 64GB (8x8GB) DDR2 533 $6,955.00
Dell PowerEdge R905 8384 4 2.7 2.7 128GB (16x8GB) DDR2 667 $11,385.00

Here we see Dell offering very aggressive DDR3/1066 pricing [for the R710] allowing us to go with on-line configurations, and HP offering overly expensive DDR2/667 memory prices (factor of 2) forcing us to go with 3rd party memory. In fact, IBM did not allow us to configure their memory configuration – as tested [with the 3650M2] – with their on-line configuration tool [neither did Dell with the M610] so we had to apply street memory prices. [Note: the So here’s how they rank with respect to VMmark:

System VMware Version Vmmark Score Vmmark Tiles Score/Tile Cost/Tile
Inspur NF5280 ESX Server 4.0 build 148592 23.45 17 1.38 $1,098.15
Dell PowerEdge R710 ESX Server 4.0 build 150817 23.55 16 1.47 $1,055.81
IBM System x 3650M2 ESX Server 4.0 build 148592 23.89 17 1.41 $1,267.41
Dell PowerEdge M610 ESX Server 4.0 23.9 17 1.41 $1,273.59
HP ProLiant DL370 G6 ESX Server 4.0 build 148783 23.96 16 1.50 $1,164.75
Dell PowerEdge R710 ESX Server 4.0 24 17 1.41 $993.71
Dell PowerEdge R805 ESX Server 3.5 U4 build 120079 11.22 8 1.40 $869.38
Dell PowerEdge R905 ESX Server 3.5 U3 build 120079 20.35 14 1.45 $813.21

As you can easily see, the cost-per-tile (analogous to $/VM) favors the Shanghai systems. In fact, the one system that we’ve taken criticism for including in our previous comparisons – the Supermicro 6026T-NTR+ with 72GB of DDR3/1066 (running at DDR3/800) – actually leads the pack in Nehalem-EP $/tile, but we’ve excluded it from our tables since it has been argued to be a “sub-optimal” configuration and out-lier. Again, the sweet spot for price-performance for Nehalem, Shanghai and Istanbul is in the 48GB to 80GB range with inexpensive memory: simple economics.

Please note, that not one of the 2P VMmark scores listed on VMware’s official VMmark results tally carry the Opteron 2393SE version of the processor (3.1GHz) or HT3-enabled motherboards. It is likely that we’ll not see HT3-enabled scores nor 2P ESX 4.0 scores until Istanbul’s release in the coming month. Again, if Shanghai’s $/tile is competitive with Nehalem’s today (again, in the 48GB to 80GB configurations), Istanbul – with the same memory and system costs – will be even more so.

Update: AMD’s Margaret Lewis has a similar take with comparison prices for AMD using DDR2/533 configurations. Her numbers – like our previous posts – resolve to $/VM, however she provides some good “street prices” for more “mainstream” configurations of Intel Nehalem-EP and AMD Shanghai systems. See her results and conclusions on AMD’s blog.

h1

Shanghai Economics 101 – Conclusion

May 6, 2009

In the past entries, we’ve looked only at the high-end processors as applied to system prices, and we’ll continue to use those as references through the end of this one. We’ll take a look at other price/performance tiers in a later blog, but we want to finish-up on the same footing as we began; again, with an eye to how these systems play in a virtualization environment.

We decided to finish this series with an analysis of  real world application instead of just theory. We keep seeing 8-to-1, 16-to-1 and 20-to-1 consolidation ratios (VM-to-host) being offered as “real world” in today’s environment so we wanted to analyze what that meant from an economic side.

The Fallacy of Consolidation Ratios

First, consolidation ratios that speak in terms of VM-to-host are not very informative. For instance, a 16-to-1 consolidation ratio sounds good until you realize it was achieved on an $16,000 4Px4C platform. This ratio results in a $1,000-per-VM cost to the consolidator.

In contrast, let’s take the same 16-to-1 ratio on a $6,000 2Px4C platform and it results in a $375-per-VM cost to the consolidator: a savings of nearly 60%. The key to the savings is in vCPU-to-Core consolidation ratio (provided sufficient memory exists to support it). In the first example that ratio was 1:1, but in the last example the ratio is 2:1. Can we find 16:1 vCPU-to-Core ratios out there? Sure, in test labs, but in the enterprise we think the valid range of vCPU-to-Core consolidation ratios is much more conservative, ranging from 1:1 to 8:1 with the average (or sweet spot) falling somewhere between 3:1 and 4:1.

Second, we must note that memory is a growing aspect of the virtualization equation. Modern operating systems no longer “sip” memory and 512MB for a Windows or Linux VM is becoming more an exception than a rule. That puts pressure on both CPU and memory capacity as driving forces for consolidation costs. As operating system “bloat” increases, administrative pressure to satisfy their needs will mount, pushing the “provisioned” amount of memory per VM ever higher.

Until “hot add” memory is part of DRS planning and the requisite operating systems support it, system admins will be forced to either over commit memory, purchase memory based on peak needs or purchase memory based on average memory needs and trust DRS systems to handle the balancing act. In any case, memory is a growing factor in systems consolidation and virtualization.

Modeling the Future

Using data from the Univerity of Chicago and as a baseline and extrapolating forward through 2010, we’ve developed a simple model to predict vMEM and vCPU allocation trends. This approach establishes three key metrics (already used in previous entries) that determine/predict system capacity: Average Memory/VM (vMVa), Average vCPU/VM (vCVa) and Average vCPU/Core (vCCa).

Average Memory per VM (vMVa)

Average memory per VM is determined by taking the allocated memory of all VM’s in a virtualized system – across all hosts – and dividing that by the total number of VM’s in the system (not including non-active templates.) This number is assumed to grow as virtualization moves from consolidation to standardized deployment. Read the rest of this entry ?

h1

AMD’s New Opteron

April 23, 2009

AMD’s announcement yesterday came with some interesting technical tidbits about its new server platform strategy that will affect its competitiveness in the virtualization marketplace. I want to take a look at the two new server platforms and contrast them with what is available today and see what that means for our AMD-based eco-systems in the months to come.

Initially, the introduction of more cores to the mix is good for virtualization allowing us to scale more gracefully and confidently as compared to hyper-threading. While hyper-threading is reported to increase scheduling efficiency in vSphere, it is not effectively a core. Until Nehalem-EX is widely available and we can evaluate 4P performance of hyper-threading in loaded virtual environments I’m comfortable awarding hyper-threading a 5% performance bonus – all things being equal.

AMD's Value Shift

AMD's Value Shift

What’s Coming?

That said, where is AMD going with Opteron in the near future and how will that affect Opteron-based eco-systems? At least one thing is clear: compatibility is assured and performance – at the same thermal footprint – will go up. So let’s look at the ramifications of the new models/sockets and compare them to our well-known 2000/8000 series to glimpse the future.

A fundamental shift away from DDR2 and towards DDR3 for the new sockets is a major difference. Like the Phenom II, Core i7 and Nehalem processors, the new Opteron will be a DDR3 specimen. Assuming DDR3 pricing continues to trend down and the promise of increased memory bandwidth is realized in the HT3/DCA2 and Opteron, DDR3 will deliver solid performance in 4000 and 6000 configurations.

Opteron 6000: Socket G34

From the announcement, G34 is analogous to the familiar 8000-series line with one glaring exception: no 8P on the road-map. In the 2010-2011 time frame, we’ll see 8-core, 12-core and 16-core variants with a new platform being introduced in 2012. Meanwhile, the 6000-series will support 4-channels of “unbuffered” or “registered” DDR3 across up to 12DIMMs per socket (3 banks by 4 channels). Assuming 6000 will support DDR3-1600, the theoretical bandwidth of a 4 channel design would yield memory bandwidths in the 40-50GB/sec range per link (about twice Istanbul’s).

AMD 2010-2013 Road-map

AMD 2010-2013 Road-map

With a maximum module density of 16GB, a 12-DIMM by 4-socket system could theoretically contain 768GB of DDR3 memory. In 2011, that equates to 12GB/core in a 4-way, 64-core server. At 4:1 consolidation ratios for typical workloads, that’s 256 VM/host at 3GB/VM (4GB/VM with page sharing) and an average of 780MB/sec of memory bandwidth per VM. I think the math holds-up pretty well against today’s computing norms and trends. Read the rest of this entry ?

h1

Opteron Turns 6: Plus Istanbul and a New Road-map

April 22, 2009

AMD released an updated technology road-map for it’s Opteron processor family, beginning with the early availability of Istanbul – its Socket-F compatible 6-core processor – shipping for revenue in May and available from OEM’s in June. This information was delivered in a webcast today.

AMD Istanbul 6-core Processor

AMD Istanbul 6-core Processor

“…up to 30 percent more performance within the same power envelope and on the same platform as current Quad-Core AMD Opteron…”

Additionally, AMD updated the availability of its Direct Connect Architecture 2.0 to be available only in the Opteron 4000 and 6000 series (socket C32 and G34, respectively). Companies waiting for the 12-core “Magny-Cours” processor will have to switch to the G34 platform in 2010. AMD announced that it is already shipping this 45nm part to sampling partners, and some customers will receive parts in 2H/2009. Magny-Cours is expected to be available from OEM’s and system vendors in 1H/2010.

Opteron 4000 series is also planned for introduction in 2010 for 1P and 2P servers and designed to address virtualized Web and cloud computing environments. The 4000 series will launch with 4- and 6-core processors…”

AMD believes, with core counts on the rise, dense computing (HPC and data center virtualization or cloud) will rely on the 4000 series and its more “green friendly” low power parts called “EE” offering comparable performance at 40W average power. This will create a differential in the server space between 4000 and 6000 (much like 2000 and 8000 today) but with overlap in the 2P market (unlike 2000/8000). The 6000 series is envisioned as a “high performance computing” part where power sensitivity is not the major concern. Read the rest of this entry ?

Follow

Get every new post delivered to your Inbox.

Join 48 other followers