Cisco Blogs


Cisco Blog > Data Center and Cloud

Nexus 1000V Responding to Scale Requirements with 10,000+ Ports per Virtual Switch

February 16, 2012 at 10:45 am PST

At Cisco Live London 2012, we announced that the Nexus 1000V distributed virtual switch (DVS) architecture will scale to support 10K+ ports across hundreds of servers.  This is a multi-fold increase over our current support of 2K ports and 64 servers.  What is driving the need to scale?  Two reasons: More VMs and broader VM mobility.

The number of VMs is growing leaps and bounds in data centers and cloud computing environments, which in turn is driving the need to scale virtual switch ports.  Depending on who you ask, we have already reached or are about to reach the tipping point where 50% of enterprise workloads have been virtualized.  In most IT environments today, you get a VM by default for computing needs; to run an app on a bare metal physical server requires special approval.  And needless to say, Moore’s Law continues to drive dense multi-core CPUs with extended memory architectures – thus enabling many more virtual machines to be instantiated on a single physical server.  We have seen UCS customers deploy 10 – 30 VMs per server for production workloads, and 50+ (in some cases 100+) VMs per server for non-production workloads and virtual desktops.  Increased adoption of public cloud computing resources, as well as growing deployments of private clouds in enterprises is also rapidly increasing the VM count.  Also, customers often assign multiple vNICs per VM, e.g. a NIC for data traffic, another for management, a third for backup and so on.  These factors are contributing to increased demand for virtual Ethernet (vEth) ports on the Nexus 1000V DVS. Read More »

Tags: , , , ,

Accelerating Customer and Partner Success with VXI and VMware View

Did you catch the news today?  As Cisco and VMware continue to collaborate in transforming virtual desktops into virtual workspaces, we reached an important milestone – an agreement to sell VMware View 5 software as part of an integrated VXI offer to our customers.  Read More »

Tags: , , , , ,

Cisco and Greenplum Partner to Deliver High-Performance Hadoop Infrastructure

Expanding its Big Data portfolio, Cisco announced a fully integrated end-to-end hardware and software infrastructure for enterprise Hadoop deployments in partnership with Greenplum, a division of EMC, that delivers industry-leading performance, scalability, advanced management capabilities and enterprise-class service and support. This solution consists of Cisco UCS 6200 Series Fabric Interconnects, Cisco UCS C-Series rack mount servers and Greenplum MR. Greeplum MR is based on the MapR M5 distribution, a completely re-engineered implementation of the Apache Hadoop stack with 100 percent compatibility. Cisco UCS is the exclusive integrated platform for Greeplum MR that can significantly reduce time-to-value and the operating expenses associated with Hadoop implementations.

Hadoop implementations can present a number of challenges to enterprise environments, many of these arise from the dichotomy between the introduction of innovative new technology and the enterprise-class performance, reliability, and support demanded by mission-critical systems. The collaboration between Cisco and Greenplum is specifically designed to provide a solution to these challenges. The joint solution delivers radically simplified deployment and management, high availability, excellent performance, exceptional scalability, and world-class service and support from long-time collaborators Cisco and EMC.

This solution can also connect, across the same management plane, to other Cisco UCS deployments running enterprise applications, thereby radically simplifying data center management and connectivity.

The configuration starts in a single rack with the ability to extend into multiple racks.

For more information or deal inquiries, please email us at: gponucs@external.cisco.com. A joint white paper is available at http://www.cisco.com/en/US/solutions/collateral/ns340/ns517/ns224/ns944/wp_greenplum.pdf.

Tags: , , , , ,

Cloud Management with Cisco Intelligent Automation + vCloud Director at VMware Partner Exchange

This week is a busy one for the Intelligent Automation cloud team, with both VMware Partner Exchange in Las Vegas and Cloud Connect in Santa Clara.

Cloud Connect is (as you might expect) all about cloud.  At VMware Partner Exchange, you’ll learn about virtual desktops, virtual workspaces, and VXI with Cisco and VMware View.  But you can also learn about more about Cisco and VMware’s complementary cloud management solutions – to help our customers accelerate their journey to cloud computing.

You may have heard about the Cisco IT internal private cloud (CITEIS – Cisco IT Elastic Infrastructure Services) by now.  At VMworld last fall, our IT team gave a presentation on how they deployed Cisco Intelligent Automation for Cloud together with VMware vSphere and vCloud Director for this successful initiative:

If you’re at VMware Partner Exchange this year, you’ll have an opportunity to learn how we did it, how it works, why it delivered great results, and how you can deploy a similar solution.  Just make sure you visit Cisco in booth 308 to see a demo of cloud management with Intelligent Automation – and attend our sessions below.

Read More »

Tags: , , , , , , , ,

Cisco Unified Computing XML API with Curl and xmlstarlet

February 9, 2012 at 2:04 pm PST

One of my favorite books is The Pillars of the Earth by Ken Follet, I’ve read it and reread it many times and each time I read it I get something new out of it.  With so many good books out there it seems silly to reread a book, especially a very long book. I think what it is, is that the story is so good, the characters so compelling that I don’t want to leave them and when I’m finished with the book I miss them.  Fortunately the book was made into a mini-series that I enjoyed and brought a nice visualization of the story.  I also think the mini-series may have attracted a new set of readers in the viewing audience.

New audiences come with new methods of distribution for the same, similar or different presentation of an already published work.  With the intent to reach a new audience I am republishing a UCS XML API focused blog from another blog site on Cisco Developer Network UCS Section.  I wrote this blog in April 2010, but the methods utilized seemed to flow from my prior entries on this site.The previously published blog has references to other blogs on the on the Cisco Developer Network site in the Cisco UCS section.

The previous blog…

Last time I wrote about using telnet to connect to the UCS Manager XML API as a way to introduce the API and show it’s lack of complexity. Now I don’t expect anyone to write an application that uses telnet to manage a UCS system, I just wanted to get across that if text, XML structured text, can be pushed across an open port to the listening API process on the UCS then it doesn’t matter how the push is done.

However telnet is not very practical, so I thought I would write about curl and xmlstarlet (xmlstarlet referred to as xml in this entry). curl is used to handle the request and response cycle with the UCS and xml is used to process the XML response. In some of my early scripts I used sed and awk to “parse” the output. I say parse but it was more pattern matching; by the way sed and awk are great tools, but maybe I’m partial to them because I’ve been around for a while. The reason I started with curl, sed and awk was not because I lacked XML experience but because I wanted to appeal to the administrators out there and show that XML experience, while beneficial, is not specifically needed.

Read More »

Tags: , , , ,