Results 1 - 10 of 29,215

Table 4: P4 1.2: Synchronous, point-to-point communication with TCP sockets.

in Using Parallel Programming Environments on Clusters of Workstations
by Rudnei Dias, Tim Hopkins

Table 6: TCGMSG 4.02: Synchronous, point-to-point communication with TCP sockets.

in Using Parallel Programming Environments on Clusters of Workstations
by Rudnei Dias, Tim Hopkins

Table 4: P4 1.2: Synchronous, point-to-point communication with TCP sockets.

in Using Parallel Programming Environments on Clusters of Workstations
by Rudnei Dias Da, Tim Hopkins

Table 6: TCGMSG 4.02: Synchronous, point-to-point communication with TCP sockets.

in Using Parallel Programming Environments on Clusters of Workstations
by Rudnei Dias Da, Tim Hopkins

TABLE I SYNCHRONIZATION PATTERNS AND CORRESPONDING CONSTELLATION POINTS.

in Detection of Aperiodically Embedded Synchronization Patterns
by Arkady Kopansky, Maja Bystrom

Table 1: Complexity From table 1 we see that the complexity is roughly the same for all methods. The only exception is MMPE that does not require any inner products. The inner products for the other extrapolation methods comes from forming the normal equations and thus they are not needed for MMPE. The di erence is the number of synchronization points needed for a parallel implementation. A synchronization point is a point in the program where all processors must have completed their tasks before the program can continue. This means that at each synchronization point all

in Solving Linear Equations on Parallel Distributed Memory Architectures By Extrapolation
by Christer Andersson

Table 3: Synchronization Statements Optimized in the Simple Synchronization Model

in Comprehensive Synchronization Elimination for Java
by Jonathan Aldrich, Emin Gun Sirer, Craig Chambers, Susan J. Eggers 2003
"... In PAGE 19: ...1 Dynamic Evaluation of the Synchronization Analyses In this section we evaluate the impact of our analyses on the dynamic behavior of the benchmarks. Table3 shows the... In PAGE 23: ... In a second experiment, we modified the Java class libraries and a subset of our applications to add synchronization to all public methods. Table3 shows the static number of synchronization points optimized by our analyses in this experiment. For most programs, thread-local analysis (shown in the fourth column) was able to eliminate virtually all of the synchronization in these programs, effectively eliminating the extraneous overhead that would be imposed by the more natural synchronization model.... ..."
Cited by 11

Table 2: Dynamic Number of Synchronization Operations Eliminated

in Comprehensive Synchronization Elimination for Java
by Jonathan Aldrich, Emin Gün Sirer, Craig Chambers, Susan J. Eggers 2003
"... In PAGE 21: ... The extent to which the reductions in dynamic synchronization operations translated into execution time speedups depended on the frequency of synchronization operations in the programs. For example, Table2 shows that jlex and jgl do far more synchronization operations per second than the other benchmarks, and that translated into a dramatic speedup for these benchmarks. Figure 6 shows the execution speed of our optimized benchmark programs relative to the unoptimized versions.... In PAGE 24: ...Table2 , which lists the reentrant synchronization points that are not also thread-local. In the multithreaded benchmarks, reentrant lock analysis typically eliminates 10% of the static synchronization points in the program, in addition to what thread-local analysis is able to find.... ..."
Cited by 11

Table 2. The availtible amount of work (in cycles) per synchronization event for a l-million grid point zone.

in Using Loop-Level Parallelism to Parallelize Vectorizable Programs
by Daniel M. Pressel, Jubaraj Sahu, Karen R. Heavey
"... In PAGE 9: ...equired for efficient execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Table2 . The available amount of work (in cycles) per synchronization event for a l-million grid point zone .... In PAGE 15: ... (2) Every effort should be made to maximize the amount of work per synchronization event (see Table 2). Table2 clearly demonstrates the advantage of parallelizing primarily outer loops. It also demonstrates the difficulty associated with the efficient parallelization of boundary condition routines.... ..."

Table 3: Synchronization Statements Optimized in the Simple Synchronization Model

in Comprehensive Synchronization Elimination for Java
by Jonathan Aldrich, Emin Gün Sirer, Craig Chambers, Susan J. Eggers 2003
"... In PAGE 22: ...1 Dynamic Evaluation of the Synchronization Analyses In this section we evaluate the impact of our analyses on the dynamic behavior of the benchmarks. Table3 shows the dynamic percentage of synchronization operations eliminated at runtime by our analyses. The first column represents the percentage of runtime synchronization operations removed by all of our analyses combined.... In PAGE 26: ... In a second experiment, we modified the Java class libraries and a subset of our applications to add synchronization to all public methods. Table3 shows the static number of synchronization points optimized by our analyses in this experiment. For most programs, thread-local analysis (shown in the fourth column) was able to eliminate virtually all of the synchronization in these programs, effectively eliminating the extraneous overhead that would be imposed by the more natural synchronization model.... ..."
Cited by 11
Results 1 - 10 of 29,215