VISION
MISSION
Delight our Partners and Alliances delivering best in class cache coherent node controller technology in Scale-up computing.
JOURNEY OF NUMASCALE
Founded in Oslo in 2008 as a spin-out from Dolphin Interconnect Solutions. Numascale technology has roots back to Norsk Data’s and Dolphin’s Scalable Coherent Interface (SCI) products. The experienced Numascale development team has previously contributed to several important commercial products, such as:
Convex Exemplar (now HP Superdome)
Data General Aviion (Acquired by EMC)
SunCluster (Sun Microsystems)
Past Product
CC Ctrl. Chip For Data General Intel Platform – Acquired by EMC 2002
Low Latency Clustering H/W Used in Sun, Siemens, Philips, Dassault/Thales Rafale Fighter
Convex Exemplar Supercomputer – Acquired by HP 1995
HT ASIC NC
HT FPGA NC
CABLE
HTX CARD
5000 Cores, Single Image OS
108 nodes.
6 x 6 x 3 Torus.
5 184 CPU cores.
58 TFlops.
Single Image OS.
10TB/s Memory BW.
UPI 1.0 NC is Deployed.
UPI 2.0 NC is in Execution.
PRESENT PRODUCTS
BASED ON UPI 2.0 TECHNOLOGY
BASED ON UPI 1.0 TECHNOLOGY
NUMASCALE IN THE NEWS
The STREAM benchmark is a synthetic benchmark program that measures sustainable memory bandwidth and the corresponding computation rate for simple vector kernels
The STREAM benchmark is specifically designed to work with datasets much larger than the available cache on any given system, so that the results are more indicative of the performance of very large, vector style applications
Proven technology: Massive computing power on commercially delivered system
Link: http://www.businesswire.com/news/home/20141120006251/en/Numascale-Supermicro-AMD-Announce-World%E2%80%99s-Largest-Shared
STREAM benchmark records (scale operation)
The A Register
HPC Wire
GEO SITES
NUMASCALE WORLDWIDE PRESENCE
