S. Salisbury, Z. Chen, and R. Melhem, "Modeling Communication Locality in Multiprocessors ", J. of Parallel and Distributed Computing, vol 56, no 2, pp. 71-98, 1999. 29 Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
An Empirical Study of Reliable Multicast Protocols over.. - Lane, Daniels, Yuan (2001) (Correct)
....presence of heterogeneous receivers. In a homogeneous cluster, some of the issues are no longer problems. In such a system, all workstations have the same CPU power and network connection, the receivers are homogeneous. In addition, communication patterns in parallel programs change very slowly [18], which indicates that for most communication patterns in parallel programs, multicast groups are static and that the group membership maintenance is not a problem. We will assume in this paper that a multicast group does not change and focus on the performance issue. Some LAN features that may ....
C. Salisbury, Z. Chen and R. Melhem, "Modeling Communication Locality in Multiprocessors", The Journal of Parallel and Distributed Computing, vol 56, no 2, pp. 71-98, 1999.
A Comprehensive Study of Reliable Multicast Protocols over.. - Lane (2000) (Correct)
....receivers. The situation is different when developing reliable multicast protocols in the LAN environment for parallel applications. First, communication patterns in parallel applications exhibit locality, that is, the communication pattern changes very slowly during the execution of a program [22]. In most cases, the communication patterns are predictable at compiler time [11] This indicates that for most communication patterns in parallel programs, multicast groups are static, that is, group members do not join and leave a group dynamically. In this paper, we will ignore the membership ....
C. Salisbury, Z. Chen and R. Melhem, "Modeling Communication Locality in Multiprocessors", The Journal of Parallel and Distributed Computing, vol 56, no 2, pp. 71-98, 1999.
Algorithms for Supporting Compiled Communication - Yuan, Melhem, Gupta Self-citation (Melhem) (Correct)
....sometimes has negative e#ects in the communication performance. As can be seen in Table 6, for the benchmark 0013, the compiled communication performs worse than dynamic communication with d = 1. The performance of the communication phase analysis can be improved when the communication locality [20] of the program is taken into consideration instead of blindly inserting patterns into each phase. benchmarks Compiled Dynamic Communication Communication d = 1 d = 4 d = 14 d = 20 0001 45,624 888,240 357,600 267,360 273,360 0003 752 14,804 5,960 4,456 4,556 0004 1,368 1,920 2,208 3,504 ....
S. Salisbury, Z. Chen, and R. Melhem, "Modeling Communication Locality in Multiprocessors ", J. of Parallel and Distributed Computing, vol 56, no 2, pp. 71-98, 1999. 29
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC