With the advent of new routing methods, the distance to which a message is sent is becoming relatively less and less important. Thus assuming no link contention, permutation seems to be an efficient collective communication primitive. All-to-many communication is required for solving a large class of irregular and loosely synchronous problems on distributed memory MIMD machines. In this paper we present several algorithms for decomposing all-to-many personalized communication into a set of disjoint partial permutations. These partial permutations avoid node contention and/or link contention. We discuss several algorithms and study their effectiveness both from the view of static scheduling as well as runtime scheduling. Experimental results...
Hypercube algorithms may be developed for a variety of communication-intensive tasks such as sending...
In this paper, we consider static scheduling techniques for heterogeneous systems, such as clusters ...
In irregular all-to-all communication, messages are exchanged between every pair of processors. The ...
With the advent of new routing methods, the distance to which a message is sent is becoming relative...
In this paper we present several algorithms for decomposing all-to-many personalized communication i...
In this paper we present several algorithms for all-too-many personalized communications which avoid...
In this paper we present several algorithms for performing all-to-many personalized communication on...
Parallelization of many irregular applications results in unstructured collective communication. In ...
This paper presents solutions for the problem of many-to-many personalized communication, with bound...
Hypercube algorithms are developed for a variety of communication-intensive tasks such as transposin...
Abstract. In the context of generating efficient, contention free schedules for inter-node communica...
In applications requiring very high throughput or which have real-time deadlines, the use of paralle...
Hypercube algorithms are developed for a variety of commun-ication-intensive tasks such as transposi...
This paper presents algorithms for implementing the transportation primitive on a distributed memory...
Whole array operations and array section operations are important features of many data-parallel lan...
Hypercube algorithms may be developed for a variety of communication-intensive tasks such as sending...
In this paper, we consider static scheduling techniques for heterogeneous systems, such as clusters ...
In irregular all-to-all communication, messages are exchanged between every pair of processors. The ...
With the advent of new routing methods, the distance to which a message is sent is becoming relative...
In this paper we present several algorithms for decomposing all-to-many personalized communication i...
In this paper we present several algorithms for all-too-many personalized communications which avoid...
In this paper we present several algorithms for performing all-to-many personalized communication on...
Parallelization of many irregular applications results in unstructured collective communication. In ...
This paper presents solutions for the problem of many-to-many personalized communication, with bound...
Hypercube algorithms are developed for a variety of communication-intensive tasks such as transposin...
Abstract. In the context of generating efficient, contention free schedules for inter-node communica...
In applications requiring very high throughput or which have real-time deadlines, the use of paralle...
Hypercube algorithms are developed for a variety of commun-ication-intensive tasks such as transposi...
This paper presents algorithms for implementing the transportation primitive on a distributed memory...
Whole array operations and array section operations are important features of many data-parallel lan...
Hypercube algorithms may be developed for a variety of communication-intensive tasks such as sending...
In this paper, we consider static scheduling techniques for heterogeneous systems, such as clusters ...
In irregular all-to-all communication, messages are exchanged between every pair of processors. The ...