Cisco’s timing for Unified Computing System - a tad off?

Does anyone else think it is a bad idea for Cisco Systems to enter the server market when the industry is experiencing the most significant year-over-year sales decline in history?

Worldwide server revenue declined 24 percent in the first quarter of 2009 and shipments dropped 24.2 percent compared to the first quarter of 2008, and no one went unscaved; all of the top five global server vendors - IBM, HP, Dell, Sun and Fujitsu/Fujitsu Siemens - saw double-digit revenue declines for the first quarter of 2009, according to Gartner, Inc.

Worldwide: Server Vendor Shipment Estimates, 1Q09 (Units)

Source: Gartner (June 2009)

Meanwhile, Cisco is marketing the hell out of its upcoming Unified Computing System (UCS), which is rumored to start shipping in a couple of weeks. The company has been offering tid bits of information about UCS through webcasts for months to build anticipation for the system. For instance, yesterday. Cisco announced it would offer rackmount servers in addition to blades.

But once the drumroll for UCS dies and the system actually ships, who’s buying?

I would love to be a fly on the wall in a Cisco executive meeting to hear their strategy with UCS. Do Cisco executives really think this is a good time to introduce an entirely new server system? And are they arrogant enough to think they can beat IBM, Dell and HP at their own game?

Intel set to release “Nehalem” Xeon processors

Intel may launch its next generation multi-core Xeon processors, code-named Nehalem, on Monday.

The company sent out invitations to a live webcast on March 30 “for the launch of a groundbreaking new server architecture.” If that doesn’t give it away, some server vendors have already announced products based on the Nehalem processors, including Cisco, which will use the Intel Xeon CPU’s in their upcoming Unified Computing System’s blade servers.  Rackable Systems already introduced CloudRack systems based on Nehalem, and Dell is expected to introduce Nehalem-based systems this week.

In earlier disclosures about Nehalem chips for x86 servers, Intel said the processor will have two, four or eight processing cores and provide better scalability than previous generations. It will also have scalable cache sizes and simultaneous multithreading, or Hyper-threading, which is already available on other Xeon processors.

While Intel prides itself on introducing multi-core processors at a faster pace than competitor AMD, some of the most significant enhancements to the new Xeon processor have existed in AMD chips for years.

For example, one of the major changes with Nehalem is integration of the memory controller into the CPU. This replaces the legacy Front Side Bus, which is a known culprit in traffic bottlenecks issues. AMD has been offering an integrated memory controller –called Direct Connect Architechture — in its Opteron CPUs for years now.

Another feature in Nehalem is the QuickPath Interconnect (QPI), which will give the chip faster access to a lot more bandwidth. This feature is similar to AMD’s HyperTransport technology, which has been around for a number of years as well.

That said, by adding QPI and an integrated memory controller, Nehalem will have access to a lot more bandwidth than its predecessors without relying on tons of cache memory, according to an ARS Technica report on Nehalem.

More importantly, what all of this means for end users is significantly better performance for applications that can take advantage of multithreading and multiple processing cores.

AMD ships more efficient, faster versions of 45nm Opteron CPU

Advanced Micro Devices (AMD) is now shipping seven new versions of its latest Opteron processor, the 45 nanometer quad-core chip, code-named Shanghai; five are high efficiency (HE) and two are designed for higher performance than the standard version of Shanghai.

The new versions of Shanghai are essentially identical to the original flavor, only the HE is more power efficient and the SE offers better performance than the standard versions.

The 45nm Quad-Core AMD Opteron HE processor is 55-watts, compared to the standard 75-watts, and speeds range from 2.1 to 2.3 GHz. A server with an HE version can save 20% more than similarly configured systems during idle times, AMD reported.

The new HE processors are available in eight server systems from HP, Rackable Systems, Dell, Sun Microsystems, and other vendors are expected to start shipping the CPUs by mid-year.

Additionally, two new 45nm Quad-Core AMD Opteron SE processors (2.8 GHz) are designed for performance-intensive workloads; this compares to the standard Shanghai chip speed of 2.7 GHz. The SE chips aren’t conservative on power though; they come in a 105-watt ACP thermal envelope and are aimed at data centers where performance trumps power efficiency, said John Fruehe, AMD’s director of business development for server and workstations.

“Depending on the application, the SE version offers up to 5% better performance [than the standard version], but it also uses more power,” Fruehe said. “The customers that use these chips are less interested in the power efficiency and more interested in the performance, so we don’t do power testing on these.”

The new SE processors are also immediately available in three new systems from HP and other AMD technology partners.

The AMD Opteron pricing model for HE versions range from $316 to  $1,514, and the two SE models are $1,165 and $2,649.

Dell shipping Egenera PAN Manager on PowerEdge servers - finally

Dell announced today it has officially begun shipping Egenera’s Processor Area Network (PAN) Manager software on its PowerEdge servers sold in North America, nine months after originally announcing it would do so.

When Dell first reported it planned to ship Malboro, Mass.-based Egenera’s PAN Manager software, it was slated for availability by June. Dell did not respond to questions regarding the delay by the time of this posting.

But, now that it is shipping, PAN Manager software will extend beyond the hypervisor and virtualizes I/O infrastructure on Dell PowerEdge servers, including Ethernet network interface cards, Fibre Channel, host bus adapters (HBAs), and Ethernet and Fibre Channel switches.

By virtualizing I/O, PAN Manager essentially creates an entire virtual datacenter where nothing is tied to physical hardware, applications or operating systems. This allows IT to allocate compute, storage, and network resources wherever and whenever necessary. The software also manages both physical and virtual resources under one pane of glass. A demo of how it works is available here, on Egenera’s website.

With PAN Manager software, Dell will have a strength against competitors with virtualization management tools, such as Hewlett Packard Co. and its Virtual Connect software, which pools and abstracts the local area network and storage area network (SAN) connections to servers and virtual machines (VMs) in HP’s BladeSystems.

When Dell first announced its partnership with Egenera, Ideas International analyst Jim Burton posted a blog stating, “In today’s market, Dell can compete very effectively with other vendors on simple server virtualization and SANs. But what it lacks is a management tool that can pull everything together into an entirely virtualized datacenter. That is where PAN Manager comes into play. With PAN Manager, Dell leaps over many of its competitors with the ability to create the virtualized datacenter of the future today using inexpensive industry-standard components…We at IDEAS feel the OEM relationship is a win-win for Dell and Egenera, as well as the customers of both companies”

Unlike other software that Dell resells, PAN Manager is integrated with the Dell hardware in the factory. “That means the customer doesn’t have to worry about installing software, and all the pieces work together correctly with multiple vendors. Dell also provides any professional services and offers first line support on Dell / PAN Systems,” said Christine Crandell, Senior Vice President of Marketing for Egenera.

PAN software existed only on Egenera’s BladeFrame products until November 2007, when the company opened it up to third-party hardware. Fujitsu Siemens Computers was the first official OEM, and Egenera PAN Manager is now integrate into its Primergy server line as well.

IDC Server Tracker shows slowing x86 server growth

Framingham, Mass.-based IDC released its Worldwide Quarterly Server Tracker for the second quarter of 2008 (2Q08) today showing that although the overall server market grew in the second quarter of 2008 (2Q08), x86-based systems experienced their slowest growth rate in 23 quarters.

As a whole, the worldwide server market grew 6.4% year over year to $13.9 billion in 2Q08, marking the ninth consecutive quarter of positive revenue growth and the highest Q2 server revenue since 2000.

Unit server shipments grew 11.1% year over year in 2Q08 driven by a hardware refresh cycle and infrastructure expansions, according to IDC.

Although volume systems revenue grew 2.1% in 2Q08, they underperformed the market for the first time since 4Q06, as server OEM’s experienced strong pricing pressure in the marketplace.

Matt Eastwood, group vice president of Enterprise Platforms at IDC, said in a statement, “IDC saw strong growth in blades, Unix systems, and IBM System z demand across the marketplace. Diversity in market demand demonstrates customers do not believe a single standardized infrastructure is capable of meeting all their computing needs.”

x86 Server Market Dynamics

x86-based systems experienced their slowest growth rate in 23 quarters. x86 server market growth slowed in 2Q08 to a rate of 3% year over year ($7 billion worldwide). The 2Q08 was also the first quarter that spending for non-x86 systems outpaced revenue growth for x86-based systems since 4Q00.

Could it be that server physical sprawl is slowing? Now that virtualization is mainstream, have people slowed down on their server hardware acquisitions?

IDC blames the x86 server market slowdown not on virtualization, but on the pricing climate, saying average selling values declined 8.4% year over year in 2Q08.

“The pricing challenges many OEMs experienced, particularly in the x86 server market, is a concern as it may foreshadow a slowdown in market demand as enterprise budgets face further scrutiny in the second half of 2008,” Eastwood stated.

“While all the major vendors exhibited strong unit growth, there was significant price competition throughout the quarter,” stated Jed Scaramella, senior research analyst for Datacenter Trends at IDC. “Low-end volume servers, such as 1- and 2-socket systems, are somewhat viewed as commodities and experienced the most pricing pressure. Additionally, the quarter was made noteworthy by the fact that several of the tier-one vendors began shipping their new systems targeting large-scale datacenters. Typically, these are stripped down servers that are designed to operate at maximum power efficiency. All components and features that are not essential, including server redundancy, are eliminated to reduce the capital expenditure of these datacenter customers.”

Nashua, N.H.-based Illuminata analyst Gordon Haff said speculation about virtualization causing server sales to decline have always swirled, but never seem to manifest.

“I’ve been hearing the “won’t people buy fewer servers?” question every time there were faster processors, more processor cores, etc. for as long as I’ve been an analyst. And the market just keeps on growing,” Haff said.

“What you are seeing here though–and which relates to virtualization to at least some degree–is the relative popularity of larger servers,” Haff said. “Virtualization really helps people more effectively utilize larger servers even when single apps don’t need all the
horsepower. Even within the x86 server space we’ve seen more interest in 4-socket servers after that category was in decline for years.”

Blade Server Market Shows Strong Shipment and Revenue Growth

Although blade revenue decelerated slightly in 2Q08, year-over-year revenue growth of 40.8% in 2Q08 was the third fastest over the past 2 years, IDC reported.

Overall, bladed servers, including x86, EPIC, and RISC blades, accounted for $1.2 billion in the second quarter, or 8.8% of quarterly server market revenue.

HP held the number 1 spot in the blade market with 53.3% market share and IBM held the number 2 position with 24.8% share. Dell and Sun also experienced blade revenue growth in 2Q08.

VMmark a server vendor leapfrog game

This week I wrote a follow up story on VMware Inc.’s virtualization performance benchmarking tool, VMmark, and found it is mainly used by vendors as a way to market their servers.

Server vendors run the VMmark test under a set of guidelines and submit results to VMware for posting. It is my suspicion that vendors play leapfrog with VMmark by looking at exsiting VMmark results and only submitting their performance results when theirs are as good or better.

For instance, IBM submitted a benchmark for its 16-core System x3850 M2
running VMware ESX v3.5, which trumped the other published as of March 2008. IBM then published a press release to brag about the results, but within a few months, Dell submitted results three PowerEdge systems sporting better virtual machine (VM) performance than IBM, and Hewlett Packard (HP) beat them all out with its ProLiant DL585 G5 server results published August 5.

HP also sent out an email to press this week boasting their top 32-core results, but didn’t mention one minor detail; they are the only vendor with results in the 32-core category so far. Sure, they are number one. They are the only one.

System Administrator Bob Plankers sums this game up nicely in his blog with a post called “Why VMmark Sucks.”  Here is what Plankers had to say:

“Having a standard benchmark to measure virtual machine performance is useful. Customers will swoon over hardware vendors’ published results. Virtualization companies will complain that the benchmark is unfair. Then they’ll all get silent, start rigging the tests, scrape and cheat and skew the numbers so that their machines look the greatest, their hypervisor is the fastest. Along the way it’ll stop being about sheer performance and become performance per dollar. Then CapEx vs. OpEx. Watt per tile. Heat per VM. Who knows, except everybody will be the best at something, according to their own marketing department.”

In addition, the benchmark is a real pain to set up and run, and the ‘free’ VMmark software requires other expensive software to work. According to VMware’s website,
VMmark requires  licenses for the following software packages;

• Microsoft Windows Server 2003 Release 2 Enterprise Edition (32-bit)—thre 32-bit copies per tile (two for virtual machines and one for that tile’s client system), and one 64-bit copy per tile (for the Java server virtual machine)
• Microsoft Exchange Server 2003 Enterprise Edition
• SPECjbb2005 Benchmark
• SPECweb2005 Benchmark

Plankers said he won’t be wasting any time or money running VMmark. “Instead, I’ll be in meetings explaining to folks why we are maxed out at 30 VMs per server when the vendor says they’ll run 50. Or why we chose VMware over Xen, when Xen claims 100 on the same hardware. I’ll have to remember the line from the FAQ that says “that VMmark is neither a capacity planning tool nor a sizing tool.”

Which begs the question: if it isn’t for use in sizing or capacity planning, exactly what is it good for?”

VMware says the benchmark is good for users who are making hardware purchasing decisions.

“The intention [of VMmark] is that customers can look at the results and make decisions based on what they see. It isn’t just about the fastest server; it’s about making system comparisons; between blades and rackmounts or a two-core or four-core system. Someone can see how much more performance they get from upgrading to four core processors, for instance,” said Jennifer Anderson, the senior director of research and development at VMware.

This makes sense, but as Plankers said, users should beware of benchmark manipulation by vendors and know that the results do not reflect the same workloads that users will run in their own data center environments.

Virtual machines per server: A viable metric for hardware selection?

When server vendors introduce new blade servers these days, they often mention virtualization in the same breath, often touting the number of virtual machines (VMs) their hardware can support. But those numbers are hardly the result of scientific method.

For instance, San Diego, Calif.-based Verari Systems recently announced that its VMware ESX 3.5-certified VB1257 for BladeRack 2 XL supports up to twice as many VMs as competitive offerings (16). After speaking with Verari, I asked the competition — Sun Microsystems, Hewlett-Packard and Dell — how many VMs their blades can hypothetically support, and was given some big numbers.

But are these server vendors asking the right question? According to Anne Skamarock, a research director at Focus Consulting, the answer is no. Although vendors boast about the number of VMs their hardware supports, “it really is a silly way to look at it,” she said.

“The number of VMs supported depends on the workload. For CPU-intensive workloads, memory will also be a significant factor in performance,” Skamarock said. ”I have spoken with customers who are running 30 VMs per 8-core system and expect to increase that to 50 VMs per system.”

Skamarock said Virtual Desktop Infrastructure adds another twist. “The rule of thumb is six to eight virtual desktops per core, but again, memory will be a big issue here depending on the OS.”

According to preliminary data from SearchDataCenter.com’s 2008 Purchasing Intentions Survey, 61% of the respondents run less than 10 VMs per server, though 33% run 10 to 25, and a mere 5% run more than 25 VMs on a server.

Vendors make big VM support claims

According to VMware Inc.’s website, server consolidation ratios commonly exceed 10 virtual machines per physical processor; so presumably, a blade server with two CPUs, like Verari’s VB1257, should be able to support at least 20 VMs.

Within HP’s ProLiant blade server line, the ProLiant BL460c/465c and BL680c/BL685c would be a good choice for a virtual server platform, primarily because they offer a large memory footprint, which means more than 16 VMs per blade in both cases, plus more network expansion and storage performance, HP spokesman Eric Krueger said.

“Keep in mind of course the number of VMs always vary – the number could be higher or lower depending on the needs the application/VM — but based on the rule of thumb … the BL460c can support up to 16 VMs and the BL680c up to 32,” Krueger said.

Sun Microsystems Inc. claims its Sun Blade servers pack two and three times that many VMs. The Sun Blade X6250, which has up to eight cores with Intel Xeon processors, 64 GB RAM, 110 Gbps I/O and 800GB of internal storage, supports 36 VMs; the Sun Blade X6450, with two or four dual-core or quad-core Intel Xeon processors and up to 96 GB of memory, can support up to 42 VMs and the Sun Blade X8450 with 16 cores per module and 128 GB Memory, supports up to 48 VMs, according to Sun.

Dell was hesitant to name a number of VMs that its PowerEdge blade servers can support, because the number is dependent on a number of factors, like workload, memory, I/O. A spokesperson did say that “Dell has blades that support up to 66 loaded VMs. This is based on VMware’s VMmark benchmark test,” a spokesperson said. “This is an area where we are doing quite a bit of work, so stay tuned.”

So I’m wondering: Are VM support numbers a consideration when buying server hardware, or is it too subjective? Let us know what you think.

Add PCI Express I/O connectivity without adding PCI Express

Recently, Tucson, Ariz.-based NextIO announced its ExpressConnect I/O virtualization product, which adds additional PCI Express (PCIe) I/O connectivity to any server in a data center.

“PCI Express is cost-effective, has a lot of bandwidth and a wide range of standard based I/O devices are available on PCIe, but usually there is only one [PCIe] device per server,” said Chris Pettey, the CTO and co-founder of NextIO. “With [ExpressConnect], you can have many PCIe devices for many servers.”

ExpressConnect works by virtualizing PCIe. It’s a 3U high box with slots into which you can plug I/O devices and is coupled with the N1400-PCM High-Speed Switch Module which enables blade servers to expand their PCIe signals outside the chassis. Doing so creates a pool of I/O resources that is separate from the server itself and can be accessed by any server.

Pettey compared ExpressConnect to Hewlett-Packard’s Virtual Connect, which virtualizes the connection between HP BladeSystem servers and a network but is proprietary to HP BladeSystem. “[ExpressConnect] can do everything HP Virtual Connect can do, only across many platforms and blades and racks. You can run any virtualization platform, any OS, and mix and match servers.”

Egenera’s Processing Area Network PAN Manager software also virtualizes I/O resources and is available on Egenera’s servers and Dell PowerEdge servers. Dell released its own version recently, called FlexAddress, for its PowerEdge M-series servers.

David G. Hill, a principal analyst at the Mesabi Group in Westwood, Mass, ranks NextIO’s product highly for data centers with high I/O throughput demands. “NextIO has the greatest impact in processing environments where the bottleneck is I/O performance, at a reasonable price,” Hill said. “The initial benefits are in I/O performance-demanding environments, such as high bandwidth, high-definition video processing, financial modeling and Web 2.0 data center virtualization.”

A few months ago, I spoke with NextI/O, then waited weeks and weeks for the company to come up with a user reference and some product pricing to no avail. In a case like this, I generally move product information from my My Documents folder to the Recycle Bin, but at face value it appears to be a pretty good technology, and Hill gave it high marks, so I (begrudgingly) decided to post this in case anyone is looking for such a technology.

Just don’t ask me what users think about ExpressConnect, because I don’t know that there are any. As for pricing, the company suggests contacting marketing@NextIO.com