EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
Instruction delivery is a critical component for wide-issue processors since its bandwidth and accuracy place an upper limit on performance. The processor front-end accuracy and ba...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
A style for programming problems from matrix algebra is developed with a familiar example and new tools, yielding high performance with a couple of surprising exceptions. The under...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
A lot of recent research on content-based P2P searching for file-sharing applications has focused on exploiting semantic relations between peers to facilitate searching. To the be...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
Most programs are repetitive, meaning that some parts of a program are executed more than once. As a result, a number of phases can be extracted in which each phase exhibits simila...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
— We present the Adaptive Group Notification (AGNO) scheme for efficiently contacting large peer populations in unstructured Peer-to-Peer networks. AGNO defines a novel implic...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
This paper describes a novel approach to performance analysis for parallel and distributed systems that is based on soft computing. We introduce the concept of performance score re...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
Abstract. This article presents a fault tolerant extension for the NaimiTrehel token-based mutual exclusion algorithm. Contrary to the extension proposed by Naimi-Trehel, our appro...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
Dynamic instruction scheduling logic is one of the most critical and cycle-limiting structures in modern superscalar processors, and it is not easily pipelined without significant ...
EUROPAR
2005
Springer
14 years 6 months ago
2005
Springer
There are two important hurdles that restrict the scalability of directory-based shared-memory multiprocessors: the directory memory overhead and the long L2 miss latencies due to ...