The Onyx supercomputer offers true supercomputing power which combines a parallel CPU subsystem configured with up to 12 R8000 64-bit RISC-based microprocessors. The R8000 is undoubtedly the fastest commercially available supercomputing microprocessor. A single R8000 processor provides 300 million double-precision floating operations per seconds (MFLOPS) and 300 million instructions per second (MIPS). The R8000, formerly code-named TFP (for what some referred to in the past as Tremendous Floating Point, and as Tera FLOPS), performance is roughly equivalent to a Cray Y-MP processor unit. The R8000 costs about twice the dollar amount of the R4400, but provides roughly 4 times the computational power of the R4400 with its 75 MFLOPS and 128 MIPS. The Onyx system will give researchers access to a computing platform that is both highly suited for advanced 3D graphics visualization as well as the development of computationally intensive algorithms, which are indispensable in addressing the critical technology fields of computer vision, cognitive science, and integrated sensing technology. This is made possible by its high bandwidth symmetric multiprocessing architecture, which incorporates a 1.2 Gigabyte per second system bus with the power of 12 R8000 RISC processors yielding 3.6 billion floating point operations per second (GFLOPS). It is this type of processing power that will facilitate the modeling of real-time vision systems.
 
 

CATE in VRML

    The Onyx Supercomputer was used to model the CATE center using Virtual Reality Modeling Language (VRML). This is the current layout of our Lab.

Click on the image to view CATE in VRML

The file is 126 kilobytes.

To view CATE in VRML you need a VRML viewer. It can be downloaded from:

http://www.cosmosoftware.com