[EN] As technology advances, the number of cores in Chip MultiProcessor systems and MultiProcessor Systems-on-Chips keeps increasing. The network must provide sustained throughput and ultra-low latencies. In this paper we propose new pipelined switch designs focused in reducing the switch latency. We identify the switch components that limit the switch frequency: the arbiter. Then, we simplify the arbiter logic by using multiple smaller arbiters, but increasing greatly the switch area. To solve this problem, a second design is presented where the routing traversal and arbitrations tasks are mixed. Results demonstrate a switch latency reduction ranging from 10% to 21%. Network latency is reduced in a range from 11% to 15%. © 2011 Elsevier B....
A comparison is made among a large number of designs for the purpose of specifying low-cost yet cost...
The on-chip communication requirements of many systems are best served through the deployment of a r...
The Clos-network is widely recognized as a scalable architecture for high-performance switches and r...
As technology advances, the number of cores in Chip MultiProcessor systems and MultiProcessor System...
Abstract—Large systems-on-chip (SoCs) and chip multiprocessors (CMPs), incorporating tens to hundred...
Power density and cooling issues are limiting the performance of high performance chip multiprocesso...
Abstract — With the increasing complexity of system-on-chip, Networks on Chip (NoC) of multi-hop swi...
High performance computer and data-centers require PetaFlop/s processing speed and Petabyte storage ...
Packet switches are used in the Internet to forward information between a sender and receiver and ar...
A simple distributed, modular architecture for a very large scale ATM switch is proposed in this pap...
[[abstract]]A switch queue structure for one-network parallel processor systems minimizes chip count...
Many of the issues that will be faced by the designers of multi-billion transistor chips may be alle...
We present a high-throughput FPGA design for supporting high-performance network switching. FPGAs ha...
Recently, high performance processor designs have evolved toward Chip-Multiprocessor (CMP) architect...
Abstract. Whereas efcient barrier implementations were once a concern only in high-performance compu...
A comparison is made among a large number of designs for the purpose of specifying low-cost yet cost...
The on-chip communication requirements of many systems are best served through the deployment of a r...
The Clos-network is widely recognized as a scalable architecture for high-performance switches and r...
As technology advances, the number of cores in Chip MultiProcessor systems and MultiProcessor System...
Abstract—Large systems-on-chip (SoCs) and chip multiprocessors (CMPs), incorporating tens to hundred...
Power density and cooling issues are limiting the performance of high performance chip multiprocesso...
Abstract — With the increasing complexity of system-on-chip, Networks on Chip (NoC) of multi-hop swi...
High performance computer and data-centers require PetaFlop/s processing speed and Petabyte storage ...
Packet switches are used in the Internet to forward information between a sender and receiver and ar...
A simple distributed, modular architecture for a very large scale ATM switch is proposed in this pap...
[[abstract]]A switch queue structure for one-network parallel processor systems minimizes chip count...
Many of the issues that will be faced by the designers of multi-billion transistor chips may be alle...
We present a high-throughput FPGA design for supporting high-performance network switching. FPGAs ha...
Recently, high performance processor designs have evolved toward Chip-Multiprocessor (CMP) architect...
Abstract. Whereas efcient barrier implementations were once a concern only in high-performance compu...
A comparison is made among a large number of designs for the purpose of specifying low-cost yet cost...
The on-chip communication requirements of many systems are best served through the deployment of a r...
The Clos-network is widely recognized as a scalable architecture for high-performance switches and r...