New supercomputer to feature 245 teraflops of compute power, 64TB of DRAM, 256TB of flash memory and 4PB of disk storage
Appro has announced a design win for the Appro Xtreme X supercomputer, named ‘Gordon’ by the San Diego Supercomputer Center (SDSC). This win is a result of a grant from the National Science Foundation (NSF) to build and operate a supercomputer to solve critical science and societal problems using HPC technology.
The supercomputer, which is to be built on Appro’s Xtreme X, employs flash memory to speed compute offerings. It employs Intel SSD flash memory to reduce I/O latency and also includes, ‘supernodes’ which exploits virtual shared memory software to create shared memory systems that increase computing speed, Appro said.
When configured and deployed in 2011, Gordon will feature 245 teraflops of total compute power, 64TB of DRAM (digital random access memory), 256TB of flash memory, and four petabytes of disk storage, the company said.
According to Appro, the new supercomputer features 32 ‘supernodes’ based on Intel xeon processors. Using virtual shared-memory software, each of the system’s 32 supernodes has the potential of 7.7TF of power and 10TB of memory. The supernodes will be interconnected via an InfiniBand network, capable of 16 gigabits per second of bi directional bandwidth.
In addition, the system will come pre configured with Appro Cluster Engine(ACE) management software designed to manage large number of physical computers connected together allowing them to function as a single computing system. The remote management suite includes network, server, cluster and storage management.
Richard Dracott, general manager of high performance computing at Intel, said: Appro Xtreme-X Architecture rises to the challenge of the San Diego Supercomputer Center’s computational demands with delivery of a balanced solution which blends the best of breed for compute, network and storage. Platforms with the next generation Intel Xeon processors and next generation Intel SSDs will maximise I/O capabilities and flash memory support to analyse large data sets used by SDSC to answer modern science’s critical problems.