### Table 1. Dynamic Source Routing (DSR) Algorithm with Non-Blocking Property

in Non-Blocking, Localized Routing Algorithm for Balanced Energy Consumption in Mobile Ad Hoc Networks

"... In PAGE 5: ... This way an intermediate node needs to wait only for a deterministic time and thus it does not affect the non-blocking property of the DSR. Table1 summarizes the DSR algorithm. 3 Energy-Balancing Ad Hoc Routing Algorithms In this section, we introduce power-aware routing algorithms.... ..."

### Table 5: Performance of the different algorithms on stochastic quiz problems as the probability of non-blocking increases. The numbers reported are percentage of the performance of the optimal, averaged across 30 independent problems.

"... In PAGE 21: ...nswer a question at any one time is not blocked over 3 conditions: 0.3, 0.6, and 1.0. The last condition corresponds to the deterministic quiz problems of Section 3. Table5 contains the results of these experiments. As the blocking probability increases, there is increased randomness as to whether questions may be available in the future.... ..."

### Table 1: Timings for blocked and non- blocked variants of the Hessenberg QR algo- rithm on one processor of the Intel Paragon.

"... In PAGE 4: ... Then, the transformations of the set are accumulated in a matrix M of appropriate size, and the remaining portions of H (as well as U) are updated by multiplying with M. Table1 gives timings for the standard double shift algorithm, the non-blocked multishift algorithm, and the variant just described. 0.... ..."

### Table 1: Two communication algorithms P2P A and P2P B are implemented to perform a collective MPI communication operation. The algorithms use non-blocking point-to-point MPI operation.

"... In PAGE 3: ... As communication methods the vendor-specific collective MPI communication operation, orthogonal realizations with all possible two- and three-dimensional group layouts and two different communication algorithms based on virtual topologies are considered. Table1 shows which communication topology is used for which collective MPI operations. For each collective MPI operation two algorithms P2P A and P2P B are implemented using non-blocking point-to-point (P2P) MPI communication operations.... ..."

### Table 5: Speedups with Non-Blocking Loads

1993

Cited by 3

### Table 5: Performance of the different algorithms on stochastic quiz problems as the probability of non-blocking inrieases. The average percentage of admissible questions at a single stage and the lower bound on the probability of successfully answering a question are fixed 10% and 0.2, respectively. The numbers reported are percentage of the performance of the optimal dynamic programming solution achieved, averaged across 30 independent problems.

1998

"... In PAGE 22: ...t any one time is not blocked over 3 conditions: 0.3, 0.6 and 1.0. The last condition corresponds to the deterministic quiz problems of Section 3. Table5 contains the results of these experiments. As the blocking probability increases, there is increased randomness as to whether questions may be available in the future.... In PAGE 22: ...o the deterministic quiz problems of Section 3. Table 5 contains the results of these experiments. As the blocking probability increases, there is increased randomness as to whether questions may be available in the future. This increased randomness leads to improved performance of myopic strategies, as shown in Table5 . Again, the advantages of the rollout strategies are evident even in this favorable case.... ..."

Cited by 39

### Table 5: Performance increase, processing time and range of code size ratios for our algorithm (POLI) and the polyvariant ow-directed inlining of Jagannathan and Wright (PCFA). The columns labelled \non-block quot; indicate results relative to a baseline that is not block compiled. All programs were compiled and run using Chez Scheme. All data was collected on the 150 MHz SGI MIPS R4400.

1997

"... In PAGE 14: ... His inlining algorithm is based on that of Jagannathan and Wright but exposes more opportunities for inlining. Table5 provides a rough comparison of the best speedups obtained by the ow-directed inlining algorithm of Jagannathan and Wright with the best speedups obtained by our algorithm (labeled POLI, for polyvariant online linear inliner). The table also compares processing time and the ranges of object code size for the... ..."

Cited by 8

### Table 5: Performance increase, processing time and range of code size ratios for our algorithm (POLI) and the polyvariant flow-directed inlining of Jagannathan and Wright (PCFA). The columns labelled non-block indicate results relative to a baseline that is not block compiled. All programs were compiled and run using Chez Scheme. All data was collected on the 150 MHz SGI MIPS R4400.

1997

"... In PAGE 14: ... His inlining algorithm is based on that of Jagannathan and Wright but exposes more opportunities for inlining. Table5 provides a rough comparison of the best speedups obtained by the flow-directed inlining algorithm of Jagannathan and Wright with the best speedups obtained by our algorithm (labeled POLI, for polyvariant online linear inliner). The table also compares processing time and the ranges of object code size for the... ..."

Cited by 8

### Table 3. Non-Blocking MPI Calls Apps Isend Irecv

2003

Cited by 2

### Table 3. Non-Blocking MPI Calls Apps Isend Irecv

2003

Cited by 2