ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
The elimination of induction variables and the parallelization of reductions in FORTRAN programs have been shown to be integral to performance improvement on parallel computers 7,...
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
In this paper, we present a GSA-based technique that performs more e cient and more precise symbolic analysis of predicated assignments, recurrences and index arrays. The e ciency...
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
Current data cache organizations fail to deliver high performance in scalar processors for many vector applications. There are two main reasons for this loss of performance: the u...
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
Threads provide a useful programming model for asynchronous behavior because of their ability to encapsulate units of work that can then be scheduled for execution at runtime, bas...
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
Modulo scheduling is an e cient technique for exploiting instruction level parallelism in a variety of loops, resulting in high performance code but increased register requirement...
ICS
1995
Tsinghua U.
13 years 11 months ago
1995
Tsinghua U.
ICPP
1995
IEEE
13 years 11 months ago
1995
IEEE
ICPP
1995
IEEE
13 years 11 months ago
1995
IEEE
Interactive program steering permits researchers to monitor and guide their applications during runtime. Interactive steering can help make end users more effective in addressing ...
ICPP
1995
IEEE
13 years 11 months ago
1995
IEEE
—We propose a unicasting algorithm for faulty hypercubes (including disconnected hypercubes) using the safety level concept. A faulty hypercube is a hypercube with faulty nodes a...