We'd love to hear from you. Let us know how we're doing and how we can make your experience here more enjoyable! We are always looking for ways to improve.
So, lets take a closer at OTV and how it works. As a reminder, OTV is an NX-OS feature that allows us to extend Ethernet LANs between data centers. One of the nice things about OTV is that it is transport agnostic—the connectivity between data centers can be L2 based, L3 based, IP switched or label switch—pretty much anything that can transport IP.
OTV works by creating an OTV control plane through authenticated links between the Nexus 7000 switches at each of your data centers (called edge nodes in OTV parlance). You can then “route” your LAN traffic by encapsulating it and routing it through this IP infrastructure. Routing of the traffic is determined by associating a MAC address with a next-hop IP address. The process is fully dynamic, so there is no need to establish and manage tunnels and virtual wires. This approach certainly simplifies management and administration over existing approaches, but it also allows you to take full advantage of your IP core such as optimal routing and features such as load balancing, multicast traffic replication, and fast failover.
If you look at history, cabling infrastructure has generally been in place ahead of network adapters and switches. That was certainly the case when the market transitioned from 10/100 to 1Gigabit Ethernet. And BASE-T technologies using twisted-pair cables have been the volume leader during each transition.
Same thing is happening here today with 10GBASE-T, short for 10Gigabit Ethernet over twisted-pair cables. 10GBASE-T can utilize the widely deployed Cat6, 6A or 7 cable plants.
But wait, why would anyone need 10G in the data center?
One of the central tenets for Data Center 3.0 is the migration from GbE to 10GbE. Whether its to support unified fabric (FCoE or iSCSI) or to support the type of I/O consumption server virtualization is driving, we feel 10GbE is a fundamental building block of the next generation of data centers.
To that end, we have spent a good deal of effort providing our customers a granular and cost effective path from their current GbE infrastructure to 10GbE. We support fibre based connections across our switching portfolio with a wide variety of optics. With the advent of the Nexus 5000, we also added twin-ax to the mix with significantly lowered costs. A little while later, we introduced the Nexus 2000 fabric extenders at yet another option for customer migrating their data centers to 10GbE that both lowered costs and simplified management. We also recently added the Cisco Nexus 4000 blade switch for the IBM BladeCenter to the mix. The latest option we are offering is 10GBase-T (IEEE 802.3an-2006) support which allows customers to take advantage of their existing copper cabling as they navigate the transition to 10GbE. In keeping with the extend-your-investment theme, 10GBase-T will initially be available for the Cisco Catalyst family first, then the Cisco Nexus family.
With the addition of the 10GBase-T options, we continue to offer the broadest, most flexible portfolio of 10GbE options. Here are some more details:
One of the ongoing challenges for our customers is finding ways to easily interconnect their data centers. Traditional drivers for this have been business continuance and the desire to load balance and make better use of underutilized resources. While these continue to be important, because of the spread of server virtualization, we also see emerging drivers around supporting inter-data center workload mobility and cloud import/export of workloads.
At this point, you may be thinking “Omar, there are already ways to do this, some even offered by Cisco—some you have have blogged about!” Yes, dear reader, its true, but with the release of a new NX-OS feature called Overlay Transport Virtualization (OTV) on our Nexus 7000, we expect to make connection your data centers simpler while at the same time, making those connections more intelligent and better suited to the emerging demands on a data center interconnect solutions.
The cool, unique thing about our OTV solution is that it works with your existing transport. Essentially, OTV provides Ethernet LAN extension and “MAC routing” on top of the existing layer 3 (i.e. IP) infrastructure. Here is a quick (~ 3 min) video that goes over the basics of the OTV solution.
The adoption of cloud-based computing promises to improve the agility, efficiency, and cost effectiveness of IT operations required to provision, scale, and deliver applications to the enterprise. As with other technology trends, delivering applications from the cloud, to remote sites, creates challenges with application performance, availability, and security.
Enterprise IT departments are continuing to invest in technologies that generate cost savings while making their business applications more agile and available. These initiatives, such as consolidation of branch-office servers and virtualization of data center servers, are increasingly being adopted by the enterprise; however, they have not been without consequences. For example, branch-office server consolidation projects, while reducing the server footprint, can result in a poor end-user experience and increased bandwidth utilization because applications traverse a WAN link with higher latency and packet loss and lower bandwidth than they traverse a LAN link. WAN optimization solutions, such as Cisco® Wide Area Application Services (WAAS), are implemented to deliver LAN-like application response times for end users and to defer a WAN bandwidth upgrade.
We have seen some incredible uptake on the Cisco Nexus 1000V as customers deploy the switch to help them scale their sever virtualization efforts. In support of that, starting next week, we are rolling out a seminar series to help customers better understand the solution and the applicability in their own environments. The sessions are free and are delivered by folks that are experts in their areas—OK, disclaimer, I am leading one session, but that’s as a host for a customer panel, so I think I shouldn’t screw that up too much. There is a business track and a technical track and include topics like ROI and business impact, basic and advanced design and configuration topics, and broader architectural considerations.
Anyway, check out the complete list of sessions and and register here. BTW, if there is a topic you’d like to see us address, let me know in the comments section.
The importance of reviewing data for Virtualization by Harris Sussman, Cisco Data Center Solutions - Unified Computing System
When buying a car, you have a choice to do research, ranging from perusing the mfg’s brochure, Consumer Report or actually driving a vehicle. Choosing a server vendor is similar, in that an IT Mgr. needs to assure the server meets the criteria for their business objectives.
For most IT buyers, purchasing decisions are not trivial, and each organization applies their own philosophies. As IT staffs embark on new virtualization projects, the aim is to reduce cost, increase business agility, and reduce complexity. There are a plethora of tools and industry benchmarks available, but when it comes to virtualized environments, it’s critical organizations get this decision right.
While most Hypervisor vendors have adequate benchmarks for their respective products, VMware’s VMmark is still perceived as the gold standard, The which combines 6 of the most common DC workloads running within a unit of work referred to as a tile. This methodology is still the most sought after.
VMware’s VMMark benchmark is one of the most active benchmark sites where vendors are constantly trying to improve their results. Just recently, Cisco published their latest UCS results http://www.vmware.com/products/vmmark/results.html to regain the number one position using the latest Intel Xeon processor for 8 core 2 socket systems. While bragging rights are important for the vendor, customers rely in this info for buying decisions.
Performance benchmarks are an important data point, but in the absence of a standard virtualization benchmark businesses must do the due diligence necessary to ensure they choose the right Hypervisor. Chris Wolf, A Burton Group analyst posted a nice blog about the need for SPECvirt now http://www.chriswolf.com/?p=303.
Here is another of my continuing series of UCS walkthoughs with Brian Schwarz. This time, we are taking a look at the Cisco Integrated Management Controller for our Cisco UCS C-Series rack servers (you can click though to our Facebook page to see a bigger version of the video).
I just wrapped up with a couple of customer briefings this morning that reinforced a trend I have been seeing over the last few months. For the last couple of years, most of the customers I talked to were in sponge mode—they were absorbing info and trying to wrap their minds around things like unified fabric and virtualization—all while battling things like immediate challenges such as regulatory compliance and infrastructure sprawl.
These days, I find our customers much more opinionated—they are asking very pointed questions, pushing back, and generally expressing a more fully formed vision of where they want to go with their data centers. To be honest, this makes for much more interesting conversations and is generally a positive development.
Steve Chambers embarked these days on a series of posts designed to open our eyes to a new approach to IT infrastructure. Sounds ambitious or arrogant? Well, Steve is a well known and respected blogger, who capitalizes on his deep knowledge of Unified Computing systems as a Cisco Data Center Architect. As Steve Kaplan another well respected IT blogger wrote on this blog “Excellent post Steve. I write and talk extensively about UCS, but didn’t really think about the benefits described in your Last Thought.”,
We have been talking extensively about the extended memory technology provided by the UCS. While Steve does not discuss in this blog, it’s definitely a key feature of the UCS that further increases the OS flexibility and ROI he describes.
In his post , Steve is bringing on another key differentiator of the UCS “This post explains how Cisco UCS offers IT infrastructure a new way of working with stateless computing. You could manage UCS the way you manage any other compute platform with the minimum of change, just like you can drive a Ferrari the same way you’d drive a pick up truck. But should you? No, you shouldn’t, you’d be wasting a fantastic opportunity”
“So how does UCS change how you drive your infrastructure?” To answer this question Steve startswith a brief description of four types of workloads (Oracle RAC database, IBM Websphere, Microsoft Windows 2008, VMware vSphere 4) and how these workloads can all run on the same blade type without any human having to reconfigure the compute, network or storage. He then reviews six UCS features (WOW-A, Unified Fabric, Over-commitment, Service Profile, Virtualized CNA and Stateless Computing) which contribute to what Steve describes is “about increasing ROI and reducing complexity, and therefore reducing OpEx”.
This first pass is just the beginning , as Steve will provide in six follow-on posts practical examples of each of these six UCS features
Delivery of hosted virtualized desktops is seeing increasing up take by organizations that want to reduce the cost of deployment and management of end user systems. While desktop virtualization offers many advantages, there are benefits to be gained in performance, scaling and security that can be addressed by choosing the right hosting platform and networking components. By optimizing the network architecture organizations can host and deliver the highest number of view images and users sessions while keeping down the total cost of ownership.
For example by hosting virtual desktop images on the Cisco Unified Computing System organizations can greatly increase the number of images per server while reducing complexity, by taking advantage of capabilities such as memory capacity, virtualized adaptors, Unified Fabric and centralized management that is integrated with solutions from VMware.
Knowledge is Power. The Cisco Live 2010 tagline here in Barcelona seems to have been well understood. Despite a sunny day, the audience was still very eager to participate and learn. Large flows of visitors, and attentive attendees characterized the spirit of this day on the show floor as well as in the conference rooms. Last time I checked on our data center partners , Panduit ,APC, Intel, EMC, VMware, NetApp were enjoying a significant traffic on their booth, making the participation to this event obviously worthwhile.
Another type of partners were eager to participate and learn – At the Partner Symposium, Cisco, VMware and EMC shared with our channel partners the Virtual Computing Environment (VCE) coalition strategy and efforts to expand the Vblock partner ecosystem. Amongst our partners, the VMware, and/or EMC partners are ready , but EMC as well as VMware are seriously committed to make their best efforts to invite Cisco partners to their network in order to build a robust group of partners able to support a growing demand for Vblocks . This symposium was certainly a great opportunity for the members of the coalition to communicate on rich set of partners programs.
The meeting with Groupe Agrica was another highlight of the day . This Cisco VMware EMC French customer is an insurance conglomerate, which manages retirement, injury, health care and saving plans for employees in the agricultural sector. Groupe Agrica is also well known, as it was one of the fist company to roll out a VDI project with VMware View. Today, Groupe Agrica manage 200 Virtual Servers running on 20 ESX hosts, and 550 Virtual Desktops running on 8 ESX hosts. Almost 99% of servers are virtualized yet. Groupe Agrica is also one of the first user of Nexus 1000V, as well as an early adopter of unified fabric and FCoE on Nexus 5000. The presentation of Julien Mousqueton , Groupe Agrica’s Data Center Architect, generated a ton of interest and a lot of questions from an audience eager to collect customer details on the challenges , best practices and benefits of these migrations.The deployment of a suite of Cisco Nexus virtual switches, combined with VMware vSphere management tools, facilitated cooperation between server and networking teams and resolved challenges arising from this infrastructure consolidation project . According to Julien, “The combination of these solutions allows our IT teams to discover new benefits to virtualized infrastructure because we can now administer a network in avirtual environment the same way as we have done in the physical world, and it is simplified and intuitive.” I will come back very soon with video interviews of Julien Mousqueton. Stay tuned.
Tonight, before an outstanding appreciation show with good music, nice food and wonderful artist , I invited bloggers from VMware and NetApp to share with us the experience of the day. It was an opportunity to talk about the Cisco VMware NetApp January 26th announcement around secure multi-tenancy design architecture
Tomorrow is the last day of Barcelona Networkers at Cisco Live 2010, with a new series of break-out sessions and the well expected super session with David Lawler on “Data Centre Virtualisation Architecture”
Today we open the “World of solutions” at 10:00 am - As experienced in San Francisco at Cisco Live last July, the Data Center of the Future demo was quickly packed with very interested visitors - The Data Center of the Future demo is a 15 racks simulated data center showcasing a complete suite of Data Center partner solutions in a data center environment - As expected we got a lot of questions on the Vblocks infrastructure packages which were presented for the first time by VMware, Cisco and EMC on a show floor.
Amongst the different data center break-out sessions, the panel around “Application Mobility across Data Centres” with Cisco Thomas Scheibe, VMware Ravi Neelakant and NetApp Joel McKelvey attracted quite a lot of attendees. Not a big suprise, as this session was dedicated to the architectural evolution in virtual machine (VM) and application mobility across separate data center sites or across different pods in the same data center. Topic we started to address last August On the networking front, this session highlighted a new Cisco solution as a way of extending layer 2 across data centers using an IP core. Additionally, Nexus 1000V Virtual switches provide granular visibility and integration at the VM level. This session also covered Routing optimizations, where optimal and service aware routing is provided to client accessing services running on Virtual Machines which have been VMotioned from one data Center or pod to another. On the Application and Storage front, using VMware’s VMotion and NetApp’s Flexcache Storage technology, customers can achieve Active-Active VMotion with no application downtime while the virtual machine moves. Stay tuned as we will post very soon some interesting video interviews that we captured today on this topic.
After a long day of videotaping and meeting customers and partners, I invited bloggers from VMware, EMC and Cisco to join me for a new Daily Blogger Techminute - It was an opportunity to talk about some of the major data center events of the day including Nexus 1000 V, the Virtual Computing Environment (VCE) panel and Vblocks.
Networkers at Cisco Live 2010 started today in Barcelona with a series of one day technical seminars and break out sessions .There was a lot of activities on the show floor as well, to be ready for the tomorrow opening of the “World of the Solutions”
The 12 racks of the Data Center of the Future demo are now ready, as well as the 7 demo pods covering Data Center Interconnect, Secure cloud, Unified Fabric, Application Delivery, Advanced Services, Virtual Computing Environment (VCE) coalition Vblocks and the data center experience of Cisco IT.
In between two meetings with customers and partners , I met briefly David Lawler, VP product Management UCS, Cisco Server Access & virtualization BU, who will be the key notes speaker of the Data Centre Power Session on Thursday 28.
David, who just came back from a trip to China, shared with me some insights on his Thursday presentation, highlighting the fresh approach, the strong foundation and the real -world impact of Cisco vision and solutions for the virtualized data center and the journey to the private cloud.
In addition of the break-out sessions previously identified by our experts, I’d like to bring your attention to several sessions recently added to the agenda
Cisco has been—and continues to—actively deploy its technology to support its own business initiatives, as well as develop best practices and target results for its customers. Called “Cisco on Cisco”, it has encompassed a wide range of technologies.
This blog is a cross-link to a posting from a Cisco IT team on their own company-wide deployment of Cisco WAAS . WAAS, or Wide Area Application Services, is a core technology in Cisco Data Center and Borderless Networks architectures.
You can see Cisco IT’s own descriptions of its 200+ site deployment, and some of the goals and projected results (e.g. $850K 3 yr. savings). Stay tuned for a more detailed case study in the weeks to come.
Posted by Omar Sultan at 12:45AM PST
The importance of reviewing data for Virtualization 
Cisco Around the Web