I recently was asked what the demand for NVMe drives was, so I pinged my peers to see where they are seeing the adoption of NVMe (in rack or blade servers forms) and got a pretty interesting response. This got me wondering what the options were for NVMe, as well as other acceleration technologies, in the blade server world. Here is what I found.
When I go to San Francisco, I head over to the west side of town to peer through the closed gates of the hidden campus of Lucasfilm’s Industrial Light and Magic (ILM) dreaming of catching a glimpse of the magic. Recently Arik Hesseldahl, from All Things Digital, accomplished my dream and had a peek behind the curtains of what makes ILM run. Arik interviewed ILM’s CIO, Kevin Clark and then toured the ILM data center. Continue reading
IBM recently announced a new addition to their BladeCenter family – the IBM BladeCenter GPU Expansion blade. This new offering provides a single HS22 with the capability of hosting up to 4 x NVIDIA Tesla M2070 or Tesla M2070Q GPUs each running 448 processing cores each. Doing the math, this equals the possibility of having 4,928 processing cores in a single 9u IBM BladeCenter H chassis. That means you could have 19,712 processing cores PER RACK. With such astonishing numbers, let’s take a deeper look at the IBM BladeCenter GPU Expansion blade.