Cisco Blogs


Cisco Blog > Data Center

Delivering Network Function Virtualization (NFV) on OpenStack Hangout

We had our first OpenStack Hangout, “OpenStack as a NFV Platform” on Tuesday Oct 21st. If you are excited about the potential of NFV, this hangout is for you. Our expert panel discussed NFV fundamentals, NFV momentum, the Neutron NFV subteam, and the new Linux Foundation OPNFV project. What I liked the most about this hangout is that we had influencers, from different companies with different backgrounds, joining us and sharing their use cases and views on future trends.

Network virtualization is poised to become a key enabler for technology and business. This innovation is driven by the OpenStack Community that has come together to solve customers’ business problems. For example, NFV promises to not only deliver high performance but also deterministic performance. This theme was highlighted throughout the discussion.

I want to take a moment and thank our moderator and panelists for doing an amazing job!

  • Mark Voelker, Technical Leader, Cisco
  • Russell Bryant, Senior Principal Software Engineer, Red Hat
  • Toby Ford, AVP of Cloud Infrastructure and Platform Architecture & Strategy, AT&T
  • Mark McLoughlin, Consulting Software Engineer, Red Hat
  • Mark McClain, Yahoo!
  • Ian Wells, Technical Leader, Cisco
  • Chris Wright, Technical Director of SDN and NFV, Red Hat

Thank you!


Read More »

Tags: , , , , ,

OpenStack Juno: The Basics

Guest Blog by Mark Voelker, Technical Lead, Cisco http://blogs.cisco.com/author/MarkVoelker/

Today, the OpenStack@Cisco team is in a celebratory mood: OpenStack 2014.2 (“Juno”) has been released!  The 10th release of OpenStack contains hundreds of new features and thousands of bugfixes and is the result of contributions from over 1400 developers.  You can find out more about Cisco’s contributions to Juno here.  What’s more, in just a few short weeks we’ll be joining the rest of the OpenStack Community in Paris for the OpenStack Summit, where plans for the next release (“Kilo“) will be laid.  We think that OpenStack’s appeal has never been higher, and are excited to see continued growth forecast for the OpenStack market.  Since OpenStack continues to see new growth, we thought this would be a good time to take a step back and review a few basics for those of you that are just beginning to get acquainted with today’s dominant open source cloud platform.

First, a bit of history.  OpenStack was founded in the summer of 2010 as an open source project driven primarily by Rackspace Hosting (who contributed a scalable object storage system that is today known as OpenStack Swift) and NASA (who contributed a compute controller that is today known as OpenStack Nova).  The announcement quickly attracted attention, and in September of 2012 the OpenStack Foundation was created as an independent body to promote the development, distribution, and adoption of the OpenStack platform.  Since then, the Foundation has grown to over 18,800 members spanning over 140 countries and representing over 400 supporting companies.

Simply put, OpenStack is “Open source software for creating private and public clouds.”  Not only is it developed by a wide variety of corporate and individual contributors, it is also used by hundreds of companies (including Cisco!) for a variety of purposes.  You can find a sampling at the OpenStack User Stories and OpenStack SuperUser websites.  The software itself is a set of loosely coupled distributed systems comprised of several discrete pieces of software with a focus on supporting multi-tenancy and scalability for on-demand resources.  Whereas OpenStack originally contained just two major components, today’s integrated Juno release contains 11:

Read More »

Tags: , , , , ,

A New Partnership for OpenStack

What happens when the market leader in cloud infrastructure partners with an open source expert? We witness a very unique opportunity for disruption.

Recently, Cisco and Red Hat announced an integrated infrastructure for OpenStack-based cloud deployments. OpenStack is a collaborationf developers and cloud computing technologists aimed at creating new, scalable solutions. This partnership will target enterprise and mid-market customers by streamlining deployment and operation of OpenStack-based private clouds while providing access to the InterCloud. The integrated infrastructure will combine Cisco Unified Computing (UCS), Cisco Application Centric Infrastructure (ACI), and Red Hat Enterprise Linux OpenStack Platform.

Together, Cisco and Red Hat will deliver a set of InterCloud-ready solutions designed to bring OpenStack to enterprise and service provider customers. By combining complementary strengths, Cisco and Red Hat are in the position to capitalize on this disruption and strengthen their positions as market leaders. Read More »

Tags: , , , ,

#EngineersUnplugged S6|Ep13: #OpenStack and Cinder

In this week’s episode of Engineers Unplugged, John Griffith (@jdg_8) and Kenneth Hui (@hui_kenneth) discuss Cinder–a way to abstract and give you block storage services inside of OpenStack. Great info with practical applications in this second of our series on OpenStack leading up to OpenStack Summit in Paris.

And let there be whiteboards and unicorns!

Cinder + OpenStack + Unicorns (courtesy of John Griffith and Kenneth Hui!)

Cinder + OpenStack + Unicorns (courtesy of John Griffith and Kenneth Hui!)

**Want to be Internet Famous? Act now! Join us for our next shoot: NetApp Insight. Tweet me @CommsNinja!**

This is Engineers Unplugged, where technologists talk to each other the way they know best, with a whiteboard. The rules are simple:

  1. Episodes will publish weekly (or as close to it as we can manage)
  2. Subscribe to the podcast here: engineersunplugged.com
  3. Follow the #engineersunplugged conversation on Twitter
  4. Submit ideas for episodes or volunteer to appear by Tweeting to @CommsNinja
  5. Practice drawing unicorns

Join the behind the scenes by liking Engineers Unplugged on Facebook.

Tags: , , ,

The Top Five Myths about Cisco ACI

It’s been almost a year since Cisco publicly unveiled its Application Centric Infrastructure (ACI). As we’ve noted in the past, ACI had to overcome a number of preconceived notions about Software Defined Networking (SDN), and without some detailed explanation, it was hard to get your head around how ACI worked and how it related to SDN. As we continue to clarify the message, there are still a number of ACI myths running around out there that we have to spend a good amount of time dispelling, so I thought I’d summarize them here. (Like Centralized Policy Management, Centralized Myth Handling can lead to greater efficiency and increased compliance. :-)).

1.      MYTH:  Cisco has limited software expertise and can’t deliver a true SDN solution because ACI requires Cisco switches (hardware) as well as the APIC controller (software).
REALITY: Cisco believes data centers require a solution that combines the flexibility of software with the performance and scalability of hardware. ACI is the first data center and cloud solution to offer full visibility and integrated management of both physical and virtual networked IT resources, all built around the policy requirements of the application. ACI delivers SDN, but goes well beyond it to also deliver policy-based automation.

2.      MYTH: ACI requires an expensive “forklift upgrade” Cisco customers must replace their existing Nexus switches with new ACI-capable switches.
REALITY: ACI is actually quite affordable due to the licensing model we use and because customers can extend ACI policy management to their entire data center by implementing a “pod” with a cost-effective ACI starter kit. On July 29, Cisco announced four ACI starter kits which are cost effective bundles that are ideal for proof-of-concept and lab deployments, and to create an ACI central policy “appliance” for existing Cisco Nexus 2000-7000 infrastructure to scale out private clouds using ACI. Customers who compare ACI to SDN software-only solutions discover that operational costs, roughly 75 percent of overall IT costs, are substantially lower with ACI — so the total cost of ownership is compelling. Along with the fact that the existing network infrastructure can still be leveraged.

3.      MYTH:  The ACI solution is not open; Cisco doesn’t do enough with the open community. 
REALITY: Openness is a core tenet in ACI design. We see openness in three dimensions: open source, open standards, and open APIs. This naturally fosters an open ecosystem as well. Several partners like F5 and Citrix already are shipping device packs for joint deployments. Customers experience tremendous benefits when vendors come together to provide tightly integrated solutions engineered to work together out of the box.

ACI is designed to operate in heterogeneous data center environments with multiple vendors and multiple hypervisors. ACI supports an open ecosystem covering a broad range of Layer 4-7 services, orchestration platforms, and automation tools. One of the key drivers behind this ecosystem is OpFlex, an open standards initiative that helps customers achieve an intelligent, multivendor, policy-enabled infrastructure. Additionally, through contributions to OpenStack Neutron with our Group-Based Policy model, we are offering a fully open source policy API available to any OpenStack user. Cisco is also working with open source Linux vendors like Red Hat and Canonical to distribute an ACI Opflex agent for OVS, and contributing the Group-based Policy model to Open Daylight.

4.      MYTH: Customers want SDN solutions for their data center networks, but ACI is not an SDN solution.
REALITY:  We believe that SDN or even software defined data centers are not the sole results customers are looking for – it is the policy-based management and automation provided by ACI that delivers tremendous benefits to application deployment and troubleshooting– and provides a compelling TCO by cutting operational costs.  Channel partners agree with us:  a recent study by Baird Equity Research surveyed 60 channel solution providers and found that they would recommend the Nexus 9000 portfolio and ACI to their customers.

5.      MYTH:  Cisco can’t compete against cheap commodity “white box” switches – they are the future of data center networks.  
REALITY:  The truth is that only a handful of companies can effectively deploy white boxes because they require a great deal of operational management and troubleshooting, which is more expensive than the upfront costs of non-commodity hardware. Deutsche Bank published a report last year titled “Whitebox Switches Are Not Exactly a Bargain” which explains how the total cost equation changes when you take into account operational costs. In addition, white boxes don’t include the rich features and capabilities that most companies want. Channel solution providers know this very well. The same Baird Equity Research study of 60 channel solution providers cited above indicated that only 2% would recommend NSX running on white-box or non-Cisco networking gear.

In the data center, “one size does not fit all”, so Cisco offers a variety of switch configurations to match customer needs. For example, customers can start with merchant silicon-based line cards and migrate to an ACI environment with ACI-capable line cards and APIC, if and when they wish.

BOTTOM LINE:  We believe that Cisco will continue to win with our partners in the data center by delivering innovation through a highly secure and application centric infrastructure.  Through training, support, and new certifications, we are empowering over two million networking engineers and thousands of channel partners worldwide to succeed with ACI in the data center and cloud.

Tags: , , , , , , ,