Cisco Blogs


Cisco Blog > Data Center

Cisco publishes results of the New SAP Concurrent Benchmark

Cisco is the first organization to publish a result for the new SAP Concurrent Benchmark

Cisco is the first organization to publish a result for the new SAP Concurrent Benchmark and have it certified by SAP on behalf of the SAP Benchmark Council. The benchmark allows vendors to demonstrate how well their SAP environments work side by side in a shared environment. Getting a new benchmark running and tuned can be difficult for some vendors, but because the Cisco Unified Computing SystemTM (Cisco UCS®) is a platform built for virtualization we were the first to demonstrate results—and we did it all using virtualization with Microsoft software: the operating system, the database management system, and the hypervisor.

Recently, the SAP Benchmark Council created a new category of concurrent benchmarks that allows benchmarking of multiple SAP dialog applications
running concurrently using shared resources—in our scenario, on a single server. The benchmark rules allow the use of any supported partitioning and isolation technologies, including hypervisors, hardware partitioning, and operating system containers. With a benchmark designed by SAP to measure the performance of these environments, we now can make objective comparisons between the same SAP applications running on bare metal or in concurrent environments with results certified by SAP.

Benchmark Results

Not only did Cisco publish the first- ever results on this new concurrent benchmark, but the results are remarkable. Comparison of the results with the results for the same software configuration running on a bare-metal server shows that the penalty for running in a virtualized environment was only 6.6 percent in terms of benchmark users, and only 6.7 percent in terms of SAP Application Performance Standard (SAPS) score.

With the Cisco UCS C240 M4 Rack Server powered by the Intel Xeon processor E5-2600 v3 product family, Cisco supports a total of 14,975 SAP SD users or a total SAPS score of 81,827. This result is excellent for virtualized environments and is further evidence that when you choose Cisco® servers and a complete Microsoft software stack, you have access to outstanding SAP performance.

Conclusion

Many organizations and SAP administrators prefer to run their landscapes on Microsoft software stacks. This first-ever SAP Concurrent Benchmark result shows just how easily you can incorporate virtualization software from Microsoft to add
more flexibility to SAP application deployments with little performance impact.
Now you can use our SAPS score certified by SAP on behalf of the
SAP Benchmark Council to estimate your capacity on Cisco UCS running Microsoft software and run all your SAP landscapes in a shared environment with higher utilization rates and with less infrastructure.

 

Tags: , , , , , ,

How a Traffic Jam in Hong Kong Gave Me Hope

A couple weeks ago, I had the opportunity to travel to China and South Korea to meet with Cisco customers and partners. The meetings went well, but it was clear that these countries share what seems like a universal condition afflicting so many cities all over the world: traffic.

I know what you’re thinking, “Traffic? Really?” Fair enough, but bear with me on this one.

Admittedly, the traffic may have been top of mind for me because of a recent advertising campaign Cisco unveiled foreshadowing the last traffic jam. The irony is that sitting in bumper-to-bumper traffic in Hong Kong gave me time to think about this in a more critical way.

Why, in today’s modern, technology-advanced era, have we not yet discovered a way to avoid traffic or at least control it? Sitting idle in traffic for many is an accepted daily annoyance, but it can also present serious consequences to the welfare and economy of many people and organizations. In the U.S. alone, it’s estimated that traffic costs $124B in lost productivity, fuel waste and higher prices for goods as a result of higher transportation costs. Multiply this by a global factor, and you begin to get the enormity of this so called “annoyance.”

At Cisco, we’re focused on creating solutions that deliver business outcomes for our customers: faster decision-making, lowering costs, increasing productivity, etc. Being close to Cisco’s data center solutions and the company’s Internet of Everything vision, I got to thinking how we’re not that far off from leaving the traffic jam in the dust.

Read More »

Tags: , , , , ,

Your Design Engineers Need Support and ‘Expertise on Tap’ Too!

If you are involved in designing, supporting or managing a data center, you will undoubtedly rely on technical support services from one or more vendors.  Running your data center, there is always the risk of a hardware failure or being impacted by a software defect.  While relatively rare, hardware does occasionally fail unfortunately.  However you undoubtedly have technical support in place to deal with such problems.  You may have invested in a few extra switches as backup, you may also have failover mechanisms in place.  Almost certainly you will have a support contract in place with your Cisco partner or with Cisco, so you have break/fix expertise on tap for when something goes wrong.   This is critical support for your business, no debate from me.

Engineer Under Stress!

Engineer Under Stress!

Now, arguably the most important resource you have in your data center is not so much individual switches, routers or servers.  It’s your engineers, those who design and support your data center.  If they have a problem, where and how do they get help?  Who helps them when they are stretched?  When business pressures are telling?  Of course, their colleagues and managers can and will help.  Where, however, can they tap into additional sources of expertise so that they can become even more productive for you?  This is where Cisco Optimization Services come in – including our award-winning Cisco Network Optimization Service (or “NOS” for short), Collaboration Optimization Service, and the one I’m involved with, Cisco Data Center Optimization Services.

 

Read More »

Tags: , , , , , , , ,

Cisco UCS Mini – a deep dive

In September 2014, Cisco launched the UCS Mini for what they dubbed “Edge-Scale Computing”. The UCS Mini offering is a stripped down version of the classic UCS system with Fabric Interconnects. To learn about classic UCS, click here.

In this post I will go into some details about the UCS Mini that I’ve learned over the past few months. The information here will assume you have the basic knowledge about what the UCS Mini is and what components are involved. If you need a refresher, I’ve written up a small introduction here.

Let’s dive in.

Fabric Interconnect

The UCS FI 6324 is the same size as the UCS I/O Modules (IOM 2208XP, 2204XP, etc) and fits in the back IOM slot of the UCS Chassis 5108 (version 2).

ucs-fi-6324-breakdown Read More »

Tags: , ,

Announcing Cisco Nexus 1000V for VMware vSphere 6 Release

The Cisco Nexus 1000V has been supported in VMware vSphere hypervisor since 4.0 release (August 2009) up to the current vSphere release 5.5 update 2.  We are happy to announce that the Nexus 1000V will continue to be supported in the latest vSphere 6 release which VMware recently announced. Customers who are currently running Nexus 1000V will be able to upgrade to the vSphere 6 release and the new vSphere 6 customers will have the Nexus 1000V as part of their choices for virtual networking.

Cisco is fully committed to support the Nexus 1000V product for our 10,000+ Advanced Edition customers and the thousands more using the Essential Edition software in all future releases of VMware vSphere. Cisco has a significant virtual switching R&D investment with hundreds of engineers dedicated to the Nexus 1000V platform.  The Nexus 1000V has been the industry’s leading virtual switching platform with innovations on VXLAN (industry’s first shipping VXLAN platform), and distributed zone firewall (via Virtual Security Gateway released in Jan 2011).

The Nexus 1000V also continues to be the industry’s only multi-hypervisor virtual switching solution that delivers enterprise class functionality and features across vSphere, Hyper-V and KVM.

In the last major release of the Nexus 1000V for vSphere, version 3.1 (August 2014) we added significant scaling and security features and we continue to provide subsequent updates (December 2014) with the next release planned for March 2015. The recently released capabilities include:

  • Increased scale per Nexus 1000V:
    • 250 hosts
    • 10,000 virtual ports
    • 1,000 virtual ports per host
    • 6,000 VXLAN segments with ability to scale out via BGP
  • Increased security and visibility
    • Seamless security policy from campus and WAN to datacenter with Cisco TrustSec tagging/enforcement capabilities
    • Distributed port-security for scalable anti-spoofing deployment
    • Enhanced L2 security and loop prevention with BPDU Guard
    • Protection against broadcast storms and or attacks with Storm control
    • Scalable flow accounting and statistics with Distributed Netflow
  • Ease of management via Virtual Switch Update Manager (VSUM) – a vSphere web-client plug-in

One of the common questions coming from our customers is whether VMware is still re-selling and supporting the Nexus 1000V via VMware support?

VMware has decided to no longer offer Nexus 1000V through VMware sales or sell support for the Nexus 1000V through the VMware support organization as of Feb 2nd 2015.  We want to reiterate that this has NO IMPACT on the availability and associated support from Cisco for the Nexus 1000V running in a vSphere environment.  Cisco will continue to sell Nexus 1000V and offer support contracts. Cisco encourages customers who are currently using VMware support for the Nexus 1000V to migrate their support contracts to Cisco by contacting their local Cisco Sales team to aide in this transition.

For questions or help, please reach out nexus1000vinfo@cisco.com

Tags: , , , , , , , ,