Paper Info
Title
TOD-Tree: Task-Overlapped Direct send Tree Image Compositing for Hybrid MPI Parallelism.
Abstract
Modern supercomputers have very powerful multi-core CPUs. The programming model on these supercomputer is switching from pure MPI to MPI for inter-node communication, and shared memory and threads for intra-node communication. Consequently the bottleneck in most systems is no longer computation but communication between nodes. In this paper, we present a new compositing algorithm for hybrid MPI parallelism that focuses on communication avoidance and overlapping communication with computation at the expense of evenly balancing the workload. The algorithm has three stages: a direct send stage where nodes are arranged in groups and exchange regions of an image, followed by a tree compositing stage and a gather stage. We compare our algorithm with radix-k and binary-swap from the IceT library in a hybrid OpenMP/MPI setting, show strong scaling results and explain how we generally achieve better performance than these two algorithms.
Year
DOI
Venue
2015
10.2312/pgv.20151157
EGPGV
Field
DocType
Citations 
Bottleneck,Shared memory,Supercomputer,Programming paradigm,Computer science,Workload,Parallel computing,Theoretical computer science,Thread (computing),Compositing,Computation
Conference
0
PageRank 
References 
Authors
0.34
13
5
Name
Order
Citations
PageRank
A. V. Pascal Grosset1222.52
Manasa Prasad242.72
Cameron Christensen3193.50
Aaron Knoll430021.33
Charles D. Hansen58112.77