Over the last few days, I’ve been listening to some interesting conversations on the topic of “fabric” in the data center. To be honest, one of the common questions I get is if there is anything materially different about “Fabric” (our Data Center Fabric or anyone else’s), or is it merely the latest buzzword from bored marketing geeks. From what I have seen, many of the companies throwing around the term “fabric” are referring only to transport and are usually tying it to a specific product or technology. On these two points, Cisco’s view of fabric markedly differs. Read More »
With all the news over the last few days regarding the continuing growth of Cisco UCS, sometimes it worth taking a step back to look at how we got here. For me, I took a look at a blog post I wrote in March 2009 (pre-FCS), and it’s interesting to see how much mindset shifting has happened in such a short period of time.
A couple of important things should jump out at you:
- Cisco UCS is a simpler, more powerful way of building Data Center (or Cloud) infrastructure.
- While change can be hard, a change to Cisco UCS doesn’t have to be difficult for your organization or your IT staff.
- The short, medium and long-term vision for Cisco UCS (even from an outsider’s point of view) was clear back in 2009, well before we laid out Cisco’s strategy to evolve the Data Center of the future.
- Even as server technology has evolved over the past two years, the core UCS architecture focus on automation has continued to differentiate the product.
- No company has greater experience in helping customers transition through technology and business shifts, as is evident by the diagram above. In today’s confusing IT environment, businesses look to technology partners they can trust to help them through transitions and deliver solutions that are ahead of the curve.
If you are managing an Itanium or SPARC based server architecture, you may be experiencing increasing maintenance costs, scarcity of administration resources, shrinking ISV support and unclear roadmaps from vendors like HP and Oracle/Sun. You may not have thought there was a viable alternative, but imagine if you could unite computing, networking, and storage access resources with a management system designed to deliver the flexibility and agility needed to scale business operations.
Cisco’s Unified Computing System was designed for the modern data center and has just expanded to include a portfolio of two-socket and four-socket blade and rack-optimized servers based on the new Intel® Xeon® processor E7 family. Check out the nine new world record performance benchmarks here. One of the new offerings is a two-socket Cisco UCS C260 server which can support up to 20 cores, 1 TB of memory and 16 SFF disk drives in a dense, 2U form factor. This is big-time compute, memory, and storage capability in an extremely small footprint. Up to 50 GB of memory per core for demanding enterprise applications, such as large-data-set and transaction-intensive databases, enterprise resource planning (ERP) applications, and decision-support systems.
Cisco UCS C260 M2 Rack-Mount Server Video Data Sheet
It was a busy week here at Cisco! Check out what you might have missed and what’s happening next week.
On Wednesday, Cisco announced technology innovations across its entire Data Center Business Advantage portfolio. With these innovations, Cisco continues to tightly integrate its Unified Fabric, Unified Computing, and Unified Network Services into a holistic data center fabric designed to be simple, scalable and highly secure, delivering any application across any location, within the data center, across data centers, or to the cloud.
Among all the IT domains, perhaps the most action is in the data center, and by extension, in the cloud. Virtualization has taken root, and delivered a lot of operational efficiency. It has provided some interesting challenges as well. Virtual Machine (VM) mobility is one. Tracking workloads as they move between servers, within and across data centers is more fun than most people imagined. So, how does one take this dynamic environment, and leverage it to fulfill requirements such as:
- Delivering anything as a service – handling heterogeneous workloads for any application
- Dealing with VM mobility – optimizing resource allocation across any location
- Offering dynamic response – responding to real-time requirements at any scale
How does one solve these emerging challenges to achieve the next levels of productivity and efficiency?
For quite some time, Cisco has believed in the promise of “going beyond silos” (Yeah, that’s the campaign we launched as well, for those of you who saw the recent ads). But awareness campaigns apart, the concept is pretty simple – how do we take some of the traditional silos in the data center like the network, compute, storage and application services and bring them together – holistically – to deliver better efficiency, resource utilization, simplicity and cost benefits.
Fundamentally, this is the promise of Cisco’s data center fabric approach – it delivers on the vision of a high-performance, shared infrastructure, that brings together the network, compute, storage access elements, and L4-7 application services into a tightly integrated resource. It is open, integrated, flexible, scalable, resilient and secure. And it is built off a vision that Cisco has been executing for 3+ years now on the foundation of Unified Fabric, Unified Network Services and Unified Computing. This foundation will form the bedrock for customers looking to move towards cloud-based models exploring application independence, location freedom and massive scale.