Making big ones out of small ones: ScaleMP
As we talked about in our last post, a handful of vendors are pushing technology that lets users tie bunches of smaller systems into large, shared-everything SMP servers.
ScaleMP was founded in 2003 and has been shipping products since 2006. It is pitching a software-based solution that allows customers to combine up to 16 x86 nodes into a large, single-image system. This can give customers up to 128 cores and 4TB of RAM in a single o/s instance to use with apps that need huge memory or can’t be easily parallelized. However, as the company points out, even highly parallelized message-passing apps perform faster on shared memory systems.
What caught our eye at SC09 is that ScaleMP were being shown off in a number of partner booths, as well as its own booth. ScaleMP’s partner list includes every Tier 1 server vendor and a number of Tier 2 OEMs and, according to the company, its seeing increasing interest as word spreads about ScaleMP’s unique approach to scaling.
The company discusses its offerings in virtualization terms, turning the VMware ‘many virtual machines from one’ concept 180 degrees to ScaleMP’s ‘one virtual machine from many’ value proposition. This is an effective approach, and I found their ‘big picture’ pitch to be among the best that I’ve seen from a company of their age and size.
We spent some time talking to Benjamin Baer, ScaleMP’s VP of Marketing, while visiting their booth at SC09. In the video, Benjamin talks about the show and how their products work…despite the fumbling, inept, “um-” and “ah-” spewing interviewer. Click here to see the video, while keeping your expectations low in terms of interviewer prowess and production values.