The Data Stack

2 Posts authored by: ajayc47

Its easy to understand why most of the media attention and news coming from IDF was  surrounding Android smart phones and Win8 tablets and ultrabooks and even the  potential for a new solar-powered “postage stamp processor”.  But in my humble opinion, that was just the proverbial tip of the iceberg.  Below the surface was the massive innovation that Intel is bringing to the data center and cloud.

 

Kirk Skaugen said in his data center update at IDF this week that Xeon E5 will ship later this year and we should start seeing them in servers at the beginning of 2012. The E5 is based on the Sandy Bridge microarchitecture and will come in 4, 6 or 8 cores, run up to 16 threads per second, and are aimed at mainstream cloud and HPC servers with 2 or 4 sockets.  It is the first chip that has the PCI-Express bus integrated into the microprocessor which will improves data throughput while saving power.  And Intel  noted more than 400 design wins for the E5.


Beyond the impressive new performance the mainstream Xeon E5 brings to the data center, Intel announced at IDF that the new 710 SSD series, ranging from 100GB to 300GB, are poised to replace hard drives in enterprise servers.   And then there’s the new 10GbE and Xeon based storage solutions that are coming to market. This “converged IA” in the data center will tear down silos and make it much simpler for developers to program solutions across computing resources, storage, and networking in the data center.

 

And last but certainly not least, I learned a lot and had fun networking with some of the industry’s most influential media and bloggers.  It was great to connect personally with: Matt   Weinberger, Greg Pfister, Reuven Cohen, Alex Williams, Chris Evans, and   John Furrier among others.

 

You can find me on Twitter @ajayc47.

Intel IT began our cloud roadmap implementation back in 2006, which started with our Grid computing environment in Silicon Design. Since then, we expanded that early roll-out to our Office and Enterprise environments. As we plan forward, we anticipate new bottle necks in overall system performance, along with a shift away from compute to storage and networking I/O.  This is primarily driven by the increased capabilities of the latest generation Xeon processors that enabled a greater number of denser VMs.

 

cloud roadmap.JPG

 

After speaking to my peers who manage similar IT enterprises, my observations are that they seem focused on capacity management in terms of CPU, memory and storage capacity without adequate focus on IO capacity.  Properly monitoring and controlling network and storage IO bottlenecks is paramount to maintaining overall system performance.

 

In a recent podcast on technlogy for tomorrow's cloud, I address the technologies Intel IT has on its cloud roadmap, and outline building a cloud for an organization. What challenges and solutions do you have for maintaining overall system performance?

Filter Blog

By date:
By tag: