Cisco Blogs

Cisco Blog > Data Center

UCS: Over 10,000 Served

This month we’re marking a special milestone… there are now 10,000 UCS customers worldwide. The natural question becomes: what’s driving this phenomenal growth? How could this possibly have been predicted?

The best explanation of snowballing UCS adoption is found in customer results. Lest we forget, adopting a new platform in the data center is not a decision undertaken lightly in IT, but word has spread in the industry about the real world benefits UCS is delivering. More and more customers are taking a look and liking what they find. It’s an admittedly bold statement to say UCS has changed the economics of the datacenter, but I’m here to tell you that it’s not marketing hype. We’re hearing from customers who are reporting all-in savings in the range of 40% on the cost of computing. Travelport, for example, conducted a deep dive TCO analysis of their pre/post UCS world and here is how they are seeing their data center economics change over the next 5 years:

Cost Category Cisco Environment Prior Environment Dollar Savings Percentage Savings
Hardware Capital $7,406,982 $9,747,500 $2,340,518 24%
Power and Cooling $915,468 $2,947,273 $2,031,805 69%
Server Administration $162,255 $1,151,945 $989,690 86%
Hardware Maintenance $301,194 $585,749 $284,555 49%
TOTAL $8,785,899 $14,432,467 $5,646,568 39%


The savings stem from a variety of sources: lower capex as the platform efficiently scales, dramatically reduced administrator time, density/ power savings and reduced SW licensing costs as more workload lands on fewer servers.  It’s cumulative and powerful.   If you want a firsthand look at the TCO/ROI impact UCS can make in your data center, check out our calculator; with 5 minutes you can get a ballpark estimate.

Economics aside, UCS just seems to make people happy.   I had a customer declare that his infrastructure was now “CTO proof.” He went on to explain that this meant the boss could deploy a server by himself without breaking anything. The infrastructure team let their CTO take a B-series blade straight out of the box, insert it into a chassis slot, and as the system identified and integrated the new resource into the available pool, they congratulated him on his first server deployment.

Beyond economic impact and increasing happiness in the data center, it doesn’t hurt that you can drop the clutch and put serious power to the ground in application performance.   In December Cisco posted TPC benchmark results that surpassed existing records by as much as 32% in raw performance and 26% in price performance.  This brings the total number of UCS world record results to 54 since introduction in 2009.

10,000 customers and growing, and it’s no wonder why.

Tags: , , , , ,

Ice Ice Baby (Word To Your Data Center Manager)

How many ice cubes does it take to cool a server?

This week’s Data Center Deconstructed Q&A item isn’t quite phrased that way, but if you consider the history of refrigeration that’s the riddle we’re pondering.  It’s a good one to figure out; Data Center designers and operators face the issue daily as part of their mission to create efficient server environments.

Read More »

Tags: , , , , , , ,

How Are Large Enterprises Utilizing Collaboration in the Cloud?

Today, we ‘re featuring a guest post from Brian Blatnik, a senior manager within Cisco’s Collaboration Technology Group:

In the month since our CloudVerse announcement the notion of a world of many clouds – public, private, and hybrid – has resonated with our customers, partners, and industry analysts. I’d like to share some perspective on how those types of clouds address different customers in the collaboration cloud services market. Since last month’s announcement highlighted our private cloud model in that market, Hosted Collaboration Solution for Large Enterprises, I’ll focus on that model. As a reminder, the Cisco Hosted Collaboration Solution gives partners, including service providers and integrators, the ability to deploy multiple collaboration applications on one server in a virtualized environment and then host those applications for multiple client organizations. The solution is designed to be run from partner data centers.

I’m often asked, “Haven’t enterprise voice and other UC services always been delivered from what we now call a private cloud?” It’s true that IP PBXs and other UC servers, like their PBX predecessors, provide services to users from a remote room or facility via a network. But there are two ways in which today’s cloud service delivery differs. First, there is the efficiency of pooling computing, network, and storage resources across multiple locations and services. Second, the services can be delivered in an on-demand fashion with elastic scaling.

The financial and strategic benefits deriving from these two factors are leading many businesses to consider consuming collaboration services in a utility model from Cisco’s partners in the Hosted Collaboration Solution (HCS). But the same drivers can result in substantial benefits to businesses that aren’t looking for services from a third party’s public cloud. Read More »

Tags: , , , , , , , , , , , , , , , , , , , , , ,

FHRP – Egress Path Optimization from the Server to the Client

I previously discussed using LISP to optimize your client-server traffic so today I’ll discuss the reverse direction: Egress Path Optimization from the Server to the Client. Let’s go over the need for Path Optimization in the direction from Server-to-Client with some pictures and explanations.

The Virtual Machine (VM) server is configured with a default gateway IP address,, which is the next hop IP address that the VM will forward packets towards as the traffic returns to the client outside the data center.  In this data center environment, we’ve deployed the default gateway using the First Hop Redundancy Protocol (FHRP).  In reality, FHRP is an umbrella technology term that includes Hot Standby Routing Protcol (HSRP) and Virtual Router Redundancy Protocol (VRRP), two main technologies that provide transparent failover and redundancy at the first hop IP router.  Please see info on FHRP here.

Also notice that the VM default gateway is the same as the HSRP Virtual IP Address (VIP). The HSRP VIP binds itself to one of the physical HSRP Routers via an HSRP election process using Layer 2 control packets between the two physical HSRP Routers and this means that the VM default gateway, since it points to a VIP, may move between physical HSRP Routers, and of course which is then intent and design when using any type of FHRP.

In the above picture, the Path is Optimized from Server to Client, so now let’s take a look at what happens when we migrate the VM to the new data center.

Read More »

Tags: , , , , , , , , , , , ,

Location, Location, Data Center Location

I spent a recent weekend helping a friend load his belongings into a moving van, in preparation for relocating his family from California to Texas.  Sometime between lifting my end of a sofa and carrying boxes that apparently contained his collection of concrete blocks, we talked about his search for a new home.

Although there are certain amenities he would like his new house to have, the biggest influence on his choice of residence is the old real estate axiom location, location, location.  He has a child in preschool and a good job, so you can guess his priorities:  something near a highly rated school, not too far from work and where property values are apt to rise over time.

Location is a critical consideration for Data Centers, too, although for different reasons.  Read More »

Tags: , , , , , , , , , ,