Cisco Blogs


Cisco Blog > Data Center and Cloud

The IT Organization Relay Race

“Spanning the globe to bring you the constant variety of sport… the thrill of victory… and the agony of defeat… the human drama of athletic competition,” voiced Jim McKay in the Wide World of Sports introduction. In two years, the world will tune in to Rio de Janeiro as the Brazil hosts the human drama of the next Summer Olympics.

The IT Organization Relay RaceI enjoy both the Winter and Summer games, but I’m particularly drawn to the Summer’s 4x100-meter relay race. My connection to the sport harkens back to summers at Fordham University in Bronx, NY. While my dad coached a summer league basketball team, I took to the track and the 4x100. It was the irony of the sport that intrigued me. In 100-meter sprints the fastest runners always lead the pack across the finish line. However, winning the 100-meter relay requires four sprinters to collectively run the fastest 400-meters, 100 meters at a time. Easy enough, one might think, just pick the four fastest sprinters and go, go, go! Of course, it’s not that simple.

The 4x100-meter relay requires acceleration, speed and coordination. A pair of sprinters must exchange a baton within a 20-meter changeover box. Both sprinters are running during the exchange. The sprinter receiving the baton races forward with his hand extended behind him. Upon feeling the baton in hand, he must quickly accelerate to full speed towards another sprinter or the finish line.

Fumble the baton exchange and the race is over. Breach the changeover box before receiving the baton and the race is over. Run out of the lane and the race is over. Having the four fastest individual sprinters does not guarantee victory.

Victory goes to the team with the fastest end-to-end performance: fast out of the blocks, quick hand off, accelerate to the next sprinter, quick hand off, accelerate to the next sprinter, quick hand off, and finally, accelerate to the finish line. Every step matters and every hand off matters. No slips, no stumbles, no drops, no falls, and most of all, no excuses.

While fundamentals of the 4x100-meter relay have remained the same for a hundred years, training and the technology beneath the sprinters have evolved. Track surfaces advanced while performance footwear adopted lighter flexible materials. As a result, today’s fastest relay teams complete the 400-meter race in under 37 seconds!

This relay shares striking similarities to IT organizations. Such organizations are involved in a relay beginning with adding assets to data centers, followed by provisioning the resources for applications and then those applications consume the resources and run as quickly as they can towards the business goal. If any leg in the relay trips, stumbles or falls, the race to transact, analyze or process data takes longer or fails. Too often the initial legs of the IT-organization-relay complete only to pass the baton to an application running too slowly to win or even finish.

IT runs a relay for businesses all day, everyday. The winners are those who are able to quickly pass the baton with agility and accuracy (Tweet This). Similar to the advancements in training and technology supporting 4X100-meter sprinters, the technology around IT must advance in order to increase overall performance.

Let applications bolt as swiftly as Usain

UCS Is the new surface and the lighter flexible material. From the start, computing fabric enables the fast deployment of resources while flash memory lets applications bolt as swiftly as Usain (Tweet This). Today’s fastest sprinters don’t win in yesterday’s rubber soled sneakers. The real-time business must run applications on the fastest, easiest to manger infrastructure. Runners, take your mark!

Tags: , , ,

Steria Group : 5 Years of Cloud Business Expansion with Cisco UCS

April 1, 2014 at 5:30 pm PST

As we have been celebrating over the past weeks the Cisco UCS 5 year anniversary, here is another great example of an enthusiastic customer and partner .

Steria has been amongst the early adopters of Cisco UCS in 2009, as the IT organization was looking for innovative solutions.

5 years after the inception,  Eric Fradet , CTO on Infrastructure Management, reflected on the achievements and shared at Cisco Live Milan how transformative for his IT organization this “bold” move was .
Thanks to the UCS deployment, Steria has been able to develop quickly cloud services, starting with the IaaS and PaaS offers, and moving now into the desktop-as-a-service with an offer called Workplace on Command.
Amongst the qualities brought to the market by the UCS concept,  Eric Fradet was prompt to highlight the performances, the ease of deployment and the security .
And it was with great delight that a very satisfied customer  wished a warm “bon anniversaire” to UCS!

Actually the story of Steria is quite remarkable as the offer evolved recently to embrace the desktop with the deployement of Cisco Prime Service Catalog [ Spoiler alert : Stay tuned for more good news around Cisco Prime Service Catalog - Check Phillip Han's blog on this topic in the following days ]

Already one of the largest providers of IT-enabled business services in Europe, Steria is also becoming a global player with a growing presence in India, North  Africa, and South East Asia. With proven consulting skills, and expertise in IT and business process outsourcing, the company decided that the time was right to extend its offer to the cloud with offers targeting enterprise users .
Read More »

Tags: , , , , , , ,

Maximizing Big Data Performance and Scalability with MapR and Cisco UCS

Huge amounts of information are flooding companies every second, which has led to an increased focus on big data and the ability to capture and analyze this sea of information. Enterprises are turning to big data and Apache Hadoop in order to improve business performance and provide a competitive advantage. But to unlock business value from data quickly, easily and cost-effectively, organizations need to find and deploy a truly reliable Hadoop infrastructure that can perform, scale, and be used safely for mission-critical applications.

As more and more Hadoop projects are being deployed to provide actionable results in real-time or near real-time, low latency has become a key factor that influences a company’s Hadoop distribution choice. Thus, performance and scalability should be evaluated closely before choosing a particular Hadoop solution.

Performance

The raw performance of a Hadoop platform is critical; it refers to how quickly the platform can ingest, process and analyze information. The MapR Distribution for Hadoop in particular provides world-record performance for MapReduce operations on Hadoop. Its advanced architecture harnesses distributed metadata with an optimized shuffle process, delivering consistent high performance.

The graph below compares the MapR M7 Edition with another Hadoop distribution, and it vividly illustrates the vast difference in latency and performance between these Hadoop distributions.

High Performance with Low Latency

 

One particular solution that is optimized for performance is Cisco UCS with MapR. MapR on the Cisco Unified Computing System™ (Cisco UCS®) is a powerful, production-ready Hadoop solution that increases business and IT agility, supports mission-critical workloads, reduces total cost of ownership (TCO), and delivers exceptional return on investment (ROI) at scale.

Read More »

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

Simplifying Cisco UCS Management with Microsoft PowerShell

March 27, 2014 at 12:12 pm PST

Many IT organizations use Microsoft PowerShell to automate and accelerate data center management tasks. The Cisco UCS PowerTool module for PowerShell provides users a comprehensive list of cmdlets to manage all components of Cisco UCS.  Users can use these cmdlets to write PowerShell scripts to simplify, speed up, and error proof UCS management and deployment. With Cisco UCS PowerTool, IT operations teams can tie together the management of storage, compute, and software applications into custom, end-to-end management solutions.

New to scripting on UCS? Check out this Google + Hangout (Tips and tricks for managing Cisco UCS with PowerShell and Microsoft Systems Center) or this video on Microsoft’s Channel 9: Manage Cisco UCS with System Center 2012 and PowerShell.

Everyone is invited to share their scripts and possibly win some prizes by entering the UCS Power Scripting Contest on the Cisco Communities.  The contest will run until May 11th.  A distinguished panel of judges (@jsnover, @rob_willis, @concentrateddon, @halr9000, & @ThomasMaurer) along with other members of the community will select the grand prize winner from a set of five finalists.

Distinguished Judges for the UCS Power Scripting Contest

Distinguished Judges for the UCS Power Scripting Contest

 

We will announce the finalists at Microsoft TechEd 2014 and announce the grand prize winner at Cisco Live 2014 in San Francisco. You don’t have to submit scripts to participate! You are encouraged to review, comment, and share best practices. Use the Twitter hashtag #CiscoScripts to share.

If you are in the Houston or Dallas area, Cisco will be hosting a UCS User Group Meeting (Houston) and Meetup (Dallas) to help you get started.

Are you interested in starting a Cisco UCS User Group in your area or hosting a local meetup? Contact your Cisco representative and let us help!

 

Tags: , , , ,

Cisco UCS Five Years On: The Right Solution at the Right Time

March 25, 2014 at 12:06 pm PST

I knew we were on to something good when a customer told me “This is so easy, it’s CTO proof.”

Early in the business, I was talking to a front-line server admin who had found that Cisco UCS made server deployment so reliable, automated and simple that he was convinced even his CTO could pull it off without breaking anything.  The enthusiasm was real, and infectious, and it changed the face of the data center market.

Thinking back five years to March of 2009, when Cisco introduced UCS, the economy was still spiraling into the worst recession of our lifetime.  IT budgets were being slashed.   Many wondered if it was the right time for Cisco to enter a new market with deeply entrenched competitors.

As it turns out, it was the perfect time.  Because change occurs fastest when times are hard.

looking for change

In the decade leading up to 2009, computing innovation had stalled.  The incumbents still had tunnel vision on the power and cooling challenges that arose out of multi-core processing in the mid-2000’s.  Innovation was essentially focused on mechanical packaging:  blade servers for mainstream IT and  “skinless” boxes for the hyperscale crowd.   Overlooked was the real problem for the vast majority of customers:  operational complexity. Remember that server virtualization was rapidly spreading in nearly every data center.  Again, this was originally a response to a hardware problem: processor utilization; but as everyone recognized the operational benefits, virtualization was taking hold very fast.  As was cloud.  Combine all this with the disaggregation of data storage from the server, which had already moved out onto the network as NAS and SAN many years before, and you had a perfect storm of complexity threatening to outpace the capacity of many IT organizations.  The individual technologies in the data center were not overwhelmingly complex but tying them all together, into a system where you could land and scale an application in a very secure and available way, became the all-consuming job of the customer.   Collectively, the industry had failed.   In 2009, more than ever, customers needed something to help them slash OPEX in the data center and free people up to face the challenges of the day.  This was the innovation vacuum that UCS had been designed to fill.

Think of UCS as the Turducken of the data center:  the sum is much, much greater (and tastier) than the parts.   A lot of true innovation has gone into UCS in the areas of server I/O and in fundamental advancements to server management technology.   The latter is especially critical, because what is often overlooked in virtualization and cloud discussions is the underlying issue of deploying, managing and scaling the physical infrastructure itself (details, details…)   The advent of UCS completed the total abstraction and automation of hardware in crucial ways that hypervisor and cloud technology still can’t acheive on their own.   API-controlled data center hardware is a foundational element of modern IT innovation, and UCS started it all.  This may be Cisco’s greatest contribution to the industry and charted the course for Cisco ACI in the broader data center.

Cisco Unified Computing System Five Years of Data Center Innovation from Cisco Data Center

 

Cisco’s not stopping.   In the intervening five years, new innovation opportunities have appeared.   Most recently, the addition of flash systems to Unified Computing in the form of UCS Invicta, which opens up a whole new chapter for what customers will be able to achieve with the System.  UCS Director is taking on a pivotal role for automation across Cisco solutions and the integrated infrastructures that we construct with our storage partners.  The future is so bright, our partners need sunglasses.

UCS timeline

The team has put together this interactive timeline that commemorates many of the milestones in the first five years of UCS.   Looking back over it, I can only feel proud and humbled to be associated with the team here at Cisco, our technology and channel partners, and most importantly with our customers, who have clearly proven that UCS was (and is) the right solution at the right time.

Tags: , , , ,