SGI has released an update and expansion for its UV line of high-performance computing systems.
The company said that the UV2 line would offer higher performance as well as a higher capacity for memory and server cores. Based on the Intel Xeon E5-4600 chips, the UV2 servers will range from a four processor two-unit rack mount box to full tower systems scalable as high as 4096 cores and 64TB of memory.
Basing the platform on Linux and running industry standard hardware platforms, SGI is touting its new platform for its flexibility and high scalability when handling big data deployments.
Jill Matzke, director of server marketing for SGI, told V3 that by using open and familiar platforms rather than proprietary architectures, the company is able to offer customers an environment with which they are already familiar.
She noted that engineers often begin by developing their applications on the smallest UV2 box in a familiar Linux environment and then scale up to the full-sized UV2000 systems.
"It has the ability to support any kind of programming and any kind of large data set," Matzke explained.
"You have that big brain, all you need to do is develop your application and then you literally just take that and scale it up."
SGI is targeting the UV2 line at the traditional high-performance markets such as physics simulations and geographic modelling as well as the growing crop of enterprises looking to adopt big data analytics options.
While other vendors have opted for bundled all-in-one big data appliances and platforms, SGI is banking on the scalability of its in-memory hardware approach and the familiarity of Linux environments to set the UV2 platform apart from competition.
"As big data becomes more of a problem, we have the ability to look at that problem in one whole piece," said Matzke.
"We don't have to break it apart, we can operate on that data set or view it all at once in memory."
Bluehole confirms rumours that Playstation 4 port is coming on 7 December
Atmospheric iodine works as a significant sink of tropospheric ozone, nullifying the harmful pollutant
A temperature rise of just 1.8° C would melt major ice sheets
The new framework could enable supercomputers that reach exascale levels