High-performance parallel processors based on star-coupled wavelength division multiplexing optical interconnects
- Pleasanton, CA
- Castro Valley, CA
- Arvada, CO
As the performance of individual elements within parallel processing systems increases, increased communication capability between distributed processor and memory elements is required. There is great interest in using fiber optics to improve interconnect communication beyond that attainable using electronic technology. Several groups have considered WDM, star-coupled optical interconnects. The invention uses a fiber optic transceiver to provide low latency, high bandwidth channels for such interconnects using a robust multimode fiber technology. Instruction-level simulation is used to quantify the bandwidth, latency, and concurrency required for such interconnects to scale to 256 nodes, each operating at 1 GFLOPS performance. Performance scales have been shown to .apprxeq.100 GFLOPS for scientific application kernels using a small number of wavelengths (8 to 32), only one wavelength received per node, and achievable optoelectronic bandwidth and latency.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA
- DOE Contract Number:
- W-7405-ENG-48
- Assignee:
- The Regents of the University of California (Oakland, CA)
- Patent Number(s):
- US 6411418
- OSTI ID:
- 874546
- Country of Publication:
- United States
- Language:
- English
Express cubes: improving the performance of k-ary n-cube interconnection networks
|
journal | January 1991 |
Selective-broadcast optical passive star comer design for dense WDM networks
|
journal | May 1991 |
Star coupler based WDM switch employing tunable devices with reduced tunability range
|
journal | January 1992 |
Similar Records
Development of Components for Wavelength Division Multiplexing Over Parallel Optical Interconnects
Cascaded wavelength division multiplexing for byte-wide optical interconnects
Related Subjects
256
32
achievable
application
apprxeq100
attainable
bandwidth
based
capability
channels
communication
concurrency
considered
distributed
division
electronic
elements
fiber
fiber optic
gflops
high-performance
improve
increased
increases
individual
individual elements
instruction-level
interconnect
interconnects
kernels
latency
memory
memory element
multimode
multiplexing
node
nodes
operating
optic
optical
optics
optoelectronic
parallel
parallel processor
performance
processing
processing systems
processor
processors
provide
quantify
received
required
robust
scale
scales
scientific
shown
simulation
star-coupled
systems
technology
transceiver
wavelength
wavelengths
wdm