A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Filters
Sublogarithmic deterministic selection on arrays with a reconfigurable optical bus
2002
IEEE transactions on computers
Preserved Fulltext
AbstractÐThe Linear Array with a Reconfigurable Pipelined Bus System (LARPBS) is a newly introduced parallel computational model, where processors are connected by a reconfigurable optical bus. ...
To our best knowledge, this is the best deterministic selection algorithm on any model with a reconfigurable optical bus. ...
buses using switches (APPBS) [8] , the array with synchronous optical switches (ASOS) [28] , the reconfigurable array with spanning optical buses (ROSOB) [29] . ...
doi:10.1109/tc.2002.1009153
fatcat:msi3h7kunffl3ht5wtam35ysv4
Parallel optical interconnects may reduce the communication bottleneck in symmetric multiprocessors
2001
Applied Optics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
and a short memory access latency. ...
We start from a detailed analysis of the communication issues in today's symmetric multiprocessor (SMP) architectures to study the benefits of implementing optical interconnects (OI) in these machines. ...
Optical Interconnects for a Symmetric Multiprocessor Cache-Coherent Switch The main limitation of shared electric busses comes from the capacitance of connected transceivers that slows down the propagation ...
doi:10.1364/ao.40.003371
pmid:18360362
fatcat:f53vxtdnybblxhpl3gulluizha
Pipelined communications in optically interconnected arrays
1991
Journal of Parallel and Distributed Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The second is a two-dimensional architecture in which processors are placed in a square grid and interconnected to one another through horizontal and vertical pipelined optical buses. ...
Further, they permit all processors to have simultaneous access to the buses using slots within a pipelined cycle. ...
In this paper, we present two multiprocessor architectures, called Array Processors with Pipelined Buses ( APPB ) , which employ optical bus interconnections in processor arrays. ...
doi:10.1016/0743-7315(91)90130-2
fatcat:hsevn53zlbaxhmopgp6nlcg2bi
SYMNET: an optical interconnection network for scalable high-performance symmetric multiprocessors
2003
Applied Optics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
As a solution, we propose a scalable address subnetwork called symmetric multiprocessor network ͑SYMNET͒ in which address requests and snoop responses of SMPs are implemented optically. ...
SYMNET can scale up to hundreds of processors while still using fast snooping-based cache-coherence protocols, and additional performance gains may be attained with further improvement in optical device ...
Litaize and J. H. Collet for their comments on this work. ...
doi:10.1364/ao.42.003407
pmid:12816328
fatcat:fea2igxnkzetnbdxzb3ltktvwi
Bi-directional optical backplane bus for general purpose multi-processor board-to-board optoelectronic interconnects
1995
Journal of Lightwave Technology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2010; you can also visit the original URL.
The file type is application/pdf.
The backplane bus demonstrated, is for general-purpose and therefore compatible with such IEEE standardized buses as VMEbus, Futurebus and FASTBUS, and can function as a backplane bus in existing computing ...
The backplane bus reported here employs arrays of multiplexed polymer-based waveguide holograms in conjunction with a waveguiding plate, within which 16 substrate guided waves for 72 (8 x 9) cascaded fanouts ...
The surface-normal fan-out is provided by a linear hologram array located between the backplane and the processor/memory boards. ...
doi:10.1109/50.390219
fatcat:zb53d7yx3vcnff3j7ry2elzhfu
An optical interconnection network and a modified snooping protocol for the design of large-scale symmetric multiprocessors (SMPs)
2004
IEEE Transactions on Parallel and Distributed Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
In Symmetric Multiprocessors (SMPs), the cache coherence overhead and the speed of the shared buses limit the address/ snoop bandwidth needed to broadcast transactions to all processors. ...
As a solution, a scalable address subnetwork called Symmetric Multiprocessor Network (SYMNET) is proposed in which address requests and snoop responses of SMPs are implemented optically. ...
Litaize for pointing out the limitations of SMPs in a quantitative manner. They would also like to thank Professor J. Collet for many fruitful discussions and suggestions. ...
doi:10.1109/tpds.2004.75
fatcat:uor76io2vjbg7ooep4776dgjk4
Architectural approach to the role of optics in monoprocessor and multiprocessor machines
2000
Applied Optics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The bottlenecks resulting from and the benefits of implementing OI's are discussed with respect to symmetric multiprocessors, rings, and distributed shared-memory supercomputers. ...
Therefore the higher the connectivity ͑possibly with optics͒, the shorter the path to another node, but the more expensive the network and the more complex the structure of electronic nodes. ...
This study partly summarizes the conclusions of the Workshop on Optical Communications and Computer Sciences ͑WOCCS͒ that was held in Toulouse, France, in March 1999. ...
doi:10.1364/ao.39.000671
pmid:18337941
fatcat:mowm54a6b5bidk3v5epnutx2a4
The application of threshold logic to the design of sequential machines
1966
Annual Symposium on Switching and Automata Theory
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Digital Communication through fading multi-path channels: Characterization of fading multi-path channels, the effect of signal characteristics on the choice of a channel model, frequency-Nonselective, ...
slowly fading channel, diversity techniques for fading multi-path channels, Digital signal over a frequency-selective, slowly fading channel, coded wave forms for fading channels, multiple antenna systems ...
Processor and memory organization: Structural unit in as processor, processor selection for an embedded systems. ...
doi:10.1109/swat.1966.29
dblp:conf/focs/MastersM66
fatcat:ierxfghv4vcpbbh2iaktwfdzlm
The Blue Gene/Q Compute chip
2011
2011 IEEE Hot Chips 23 Symposium (HCS)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
External IO
-PCIe Gen2 interface
System-on-a-Chip design : integrates processors,
memory and networking logic into a single chip
BG/Q Processor Unit
A2 processor core
-Mostly same design ...
and buses ECC protected • Minor buses, GPRs, FPRs: parity protected, with recovery • Stacked / DICE latches Single MPI
Task
User defined parallelism
User defined
transaction start
User defined ...
Design Challenges --continued And the enemy is us… Methodology Complexity -Processor cores originated in a high-speed custom design methodology -Rest of the chip implemented as ASIC Required merging ...
doi:10.1109/hotchips.2011.7477488
fatcat:nhwlkj3x2zfwjngllgct55plta
Star: A Local Network System for Real-Time Management of Imagery Data
1982
IEEE transactions on computers
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
A model for comparing cost-effectiveness among Starnet, crossbar, and multiple buses is included. ...
The objective is to accomplish a cost-effective system which provides multiple users a real-time service of manipulating very large volume imagery information and data. ...
Synchronous Operation: The synchronous operation of baseline network and crossbar switch are modeled by assuming each input link generate a request with probability Po in every cycle. ...
doi:10.1109/tc.1982.1675901
fatcat:bab67cqjnjgmxe5h53d6xwdmpe
The Scalable Coherent Interface and related standards projects
1992
IEEE Micro
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
The SCI protocols support cache coherence in a distributed-shared-memory multiprocessor model, message passing, I/O, and local-area-network-like communication over fiber optic or wire links. ...
P1596.1 defines the architecture of a bridge between SCI and VME; P1596.2 compatibly extends the cache coherence mechanism for efficient operation with kiloprocessor systems; P1596.3 defines new low-voltage ...
Therefore, buses can't support a large number of processors, especially not fast ones. ...
doi:10.1109/40.124376
fatcat:3a4opujvrvdllm3aag3cgo6o6a
First- and second-level packaging of the z990 processor cage
2004
IBM Journal of Research and Development
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
The processor frequencies are increased to 1.2 GHz, with a 0.6-GHz nest with synchronous double-data-rate interchip and interblade communication. ...
Each blade is plugged into a common center board, and a blade contains the node with up to sixteen processor cores on the multichip module (MCM), up to 64 GB of memory on two memory cards, and up to twelve ...
Acknowledgment The authors would like to thank Roland Frech for performing the VHDM connector measurements and for providing the coupling data in Table 4 . ...
doi:10.1147/rd.483.0379
fatcat:hr27o73sjjas7ii7km6q3suk7m
QsNetII: Defining High-Performance Network Design
2005
IEEE Micro
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
Scientific codes often use nonblocking communication, and ideally the network and the network interface can make progress with as little processor overhead as possible. ...
On the one hand, this component interfaces with standard I/O buses, such as peripheral component interconnect (PCI), its extended version (PCI-X), and PCI-Express, thus leveraging commodity computing nodes ...
Figure 4 shows the Elite4's layout, and Figure 5 shows the optical interface with a 16way switch. ...
doi:10.1109/mm.2005.75
fatcat:3bl7i55zfrb5jhjpsahy7564ae
High-Performance Energy-Efficient Multicore Embedded Computing
2012
IEEE Transactions on Parallel and Distributed Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
With Moore's law supplying billions of transistors on-chip, embedded systems are undergoing a transition from single-core to multicore to exploit this high-transistor density for high performance. ...
We also discuss modern multicore processors that leverage these HPEEC techniques to deliver high performance per watt. ...
ACKNOWLEDGMENTS This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the US National Science Foundation (NSF) (CNS-0953447 and CNS-0905308). ...
doi:10.1109/tpds.2011.214
fatcat:vagqmojdsjevvc2u2ewqrcjjpq
High-speed interconnect and packaging design of the IBM System z9 processor cage
2007
IBM Journal of Research and Development
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
High frequencies and massively parallel connectivity lead to a raw packaging bandwidth of up to 1,764 GB/s between processors and cache within a single frame for a fully configured four-node z9 system. ...
speed on all high-speed buses. ...
Because they are source-synchronous, z9 high-speed interfaces required a sampling clock to be sent along with data. For the z9 off-MCM buses, this meant a differential clock for each data group. ...
doi:10.1147/rd.511.0037
fatcat:o6cvkk75rbfebbfiznd2s7cp7m
« Previous
Showing results 1 — 15 out of 421 results