I was getting caught up with my reading this weekend and was reminded of the John F Kennedy quote about a rising tide lifting all boats. Intel has once again done their magic and released their Xeon 5600 CPU (aka Westmere) sporting a couple of more cores, a larger L3 cache and some other neat tidbits. The newest Intel processor becomes a platform upon which we build some interesting things by layering on our own innovations. However, as I worked my through another article, I began to wonder if we were the only ones thinking that way. In this interesting InfoWorld article, Paul Venezia reviews three of the latest blade server offerings based on the Westmere processor (Cisco UCS was not one of the systems reviewed). The interesting thing was the at the end of the day, Paul concluded “[t]here is no significant difference in blade performance in similarly equipped blades from any of these vendors.” So, while it seems Intel has done their part, it also seems very little was added by the vendors in the test.
Paul continues that meaningful differentiation came down to features such as options, management, etc. As I mentioned, our UCS was not part of this test, Paul recently completed a thorough evaluation of the system, where he concluded that the UCS was “a more manageable, more scalable, and essentially superior blade server system.” So, all things being equal, I think UCS had a good chance at taking top honors in a direct comparison (the UCS out-scored the systems in the recent test, although I am not sure its appropriate to compare scores across articles). That being said..all things are not equal…we have some significant innovations to bring to the party beyond Intel’s efforts, including our extended memory technology and our FEX technology, which I think translate to meaningful and tangible benefits for our customers. As examples, we just released results on our latest app testing including a new record on the VMware VMmark (with a 42% improvement over the previous 2-socket record), and Oracle just announced a new record on the SPECjAppServer2004 benchmark, which measures application server performance.
Cisco continues to prove it’s Unified Computing System is not only an architectural game changer, but it’s wide adoption and consistent performance are impressing folks like Boyd Davis, General Manager, Data Center Marketing, Intel who said “It’s quite an achievement for Cisco to have scored so high on so many of the industry benchmarking tests”
Cisco’s Unified Computing System B250 M2 Blade Server, based on the newly announced Xeon 5600 Processor, achieved record results with a score of 35.83 @ 26 tiles in the VMware VMmark™ benchmark, which measures virtualization performance using a server consolidation workload. This represents a 42 percent improvement over the previous highest two-socket published result.
In the SPECjAppServer2004 benchmark, the UCS C250 M2 Server set a new record. In this test, the Cisco Unified Computing System increased performance by 30% compared to previously published two-socket single-node server results.
Based on a deluge of questions in my inbox, I guess its time to re-visit the perilous topic of equipment testing again. As I recently noted, you can have rigorously designed tests executed in a conscientious manner that may still only provide an incomplete picture at best. And sometimes you get tests that reveal more about the tester than the actual product that was purportedly being tested.
In case you missed it, a competitor recently released a report they commissioned that shows that their blade system offered better throughput than the Cisco UCS because of what they proposed was an inherent design flaw—over-subscription of the blade uplinks from each server. Unfortunately for them, it was their test plan that was flawed, not our UCS architecture or blade server innovation.
The fundamental fault with the test was it was conducted with a mis-configured Cisco UCS. Essentially, the testing firm matched a normally running system from our competitor against a Cisco UCS in its failover state.
Our favorite geeks are coming back ! Discover how Vblock delivers pre-engineered, integrated, and tested IT infrastructure that combines storage, compute, hypervisor, management, and security into a single, highly virtualized and standardized platform. This 60-minute TechWiseTV episode on air tomorrow Thursday March 4 explores this latest advancement in unified computing and shows how it will make it easier for organizations to build out data centers, based on Cisco innovative systems such as our blade server
So I love being an evangelist, which is probably why its good that I am doing what I am. Back when we initially launched Data Center 3.0 we spent a lot of time preaching the virtues of “consolidation, virtualization and automation” and there are times I was certainly felt like I was wandering in the wilderness, but regardless, it was a lot of fun.
Fast forward a couple of years…I was working on an AR/PR deck and I was stuck by how much of what we talked about has turned into shipping product and implementable solutions. While that is cool, what I find even cooler is that that these solutions are not just Cisco-only efforts, but encompass technologies from a broad cross-section of partners. Here are a couple of examples of the kinds of things we are doing with folks that are probably already in your data center. First off, here is a recap of what we are dong with VMware and NetApp on workload mobility--for a more in-depth look, check out this post by Brian Gracely.