On September 10th, 2013 Cisco introduced support for the Intel® Xeon® Processor E5-2600 v2 product family on the Cisco UCS B200 M3, C240 M3, and C220 M3 servers. On the same day as the Intel announcement Cisco captured seven world records on industry benchmarks on Cisco UCS to highlight the breadth of Cisco’s product line and the way in which Cisco can accelerate performance across the data center— on enterprise applications, Java application servers, desktop virtualization, and raw CPU power.
As we know, there is no better way to compare performance than by using industry-standard benchmarks, and with seven new world record performance benchmark results Cisco has demonstrated Cisco Unified Computing System’s outstanding performance and IT productivity across key data center workloads. Check out the Performance Brief for additional information on the seven new Cisco UCS world record benchmarks. The detailed benchmark disclosure reports are available here.
Cisco UCS delivers versatility with performance leadership across a wide range of workloads, enabling customers to eliminate infrastructure silos historically driven by unique application needs. The performance leadership across a wide range of workloads provided by Cisco UCS is validated by the seven World records announced this week which include:
- Oracle E-Business Suite Applications R12 Benchmark
- SPECjbb2013 Benchmark (MultiJVM x86/x64 Java server performance)
- SPECjbb2005 (Java business logic performance)
- SPECint_base2006 (general CPU performance)
- SPECfp_rate_base2006 (general CPU performance)
- SPECompG_base2012
- VMware View Planner Benchmark (desktop virtualization performance)
It is interesting to note that although all vendors have access to same Intel processors, only Cisco UCS unleashes their power to deliver high performance to applications through the power of unification. The unique, fabric-centric architecture of Cisco UCS, with exclusive Cisco SingleConnect Technology, maximizes Intel innovations and as a result Cisco UCS with the Intel Xeon processor E5-2600 v2 family delivered up to 48 percent better performance over the prior generation of Intel Xeon processors as shown in the graph below:
So the momentum continues…In four short years, the Cisco UCS has captured over 81 world records for performance and IT productivity taking its place among the most trusted server vendors on the market. Check out the Cisco Unified Computing System™ 81 World-Record Performance Results.
The architectural advantages of a single cohesive system optimized for virtualized environments coupled with the industry leading performance validated by World Record Industry-Standard Benchmarks makes the Cisco Unified Computing System an “infrastructure platform of choice” to provide industry-leading performance in your data center. Don’t forget to check the excellent blog by Todd Brannon at http://blogs.cisco.com/datacenter/tick-tock-goes-the-server-clock
These world-record performance results further reinforce the fact that Cisco is not just selling servers—it is re-inventing the server market. For additional information on Cisco UCS and Cisco UCS solutions please visit Cisco Unified Computing & Servers web page.
Girish Kulkarni
Sr. Product Marketing Manager
Unified Computing System
You guys need to correct your statements!
The Cisco UCS B200 M3 Blade Server is NOT #1 in 2-socket server SPECjbb2013 MultiJVM as the 62,393 max-jOPS does NOT beat the SPARC T5-2 which achieves 21% higher throughput with 75,658 max-jOPS!
http://www.spec.org/jbb2013/results/jbb2013multijvm.html
http://www.oracle.com/us/solutions/performance-scalability/sparc-t5-2-specjbb2013-1925099.html
No correction needs to be published. The world record was for dual socket x86 servers. While the T5 from Oracle is impressive, that server also takes up 3U, draws 2,000w of power, emits 5000 BTU’s/hr, runs at almost 70dB and requires a commitment to Solaris 11 rather than Red Hat Linux. All specs from Oracle’s product sheet.
http://www.oracle.com/us/products/servers-storage/servers/sparc/oracle-sparc/t5-2/sparc-t5-2-ds-1922871.pdf
The B200M3 is significantly smaller and far more efficient in every other category, beyond the fact that it fits into a much more flexible stateless computing model. The price point is also lower for equivalent configs.
Well then Cisco should be stating that these are x86 world records and not *server* world records compared against *all* servers as is implied. And as for your comparison, probably better/more fair to compare rack server to rack server than blade to rack. Of course there will be differences but you don’t take into account the blade chassis rack space, power and noise thats generated from the required blade chassis.
So why not compare Cisco’s closest comparable rack system like the C240 M3? Whats interesting here is that the SPARC T5-2, while only 1RU larger, supports 33% higher RAM capacity, supports 3 more PCI Gen3 express slots, 2 more 10Gig Ethernet ports (vs only Gig E on theC240) and draws 400W less Power at maximum configuration. And without C240 benchmarks available with E5-2600 v2, we have to assume its similar or slower than the B200 M3 so the SPARC T5-2 is still ~20% faster.And finally, SPARC T5-2 will significantly outperform any E5-2600 v2 based system on database or any heavy I/O bandwidth workloads, considering that the SPARC T5-2 is showing an 81% advantage in STREAM benchmark compared to Oracles own E5-2600 v2 based X4-2L. I guess we’ll have to wait till Cisco publishes a TPC-C or TPC-H result to compare to SPARC T5…
It does say “New Intel Processors – Seven New Cisco UCS Performance Records” as the title – Intel doesn’t make the Spark chip. Also Forgive my ignorance but does a T5 from Oracle support VMWare / Linux or Windows OS native?
nice to know about this new intel processor