Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

ATCOM: Automatically Tuned Collective Communication System for SMP Clusters

Thesis/Dissertation ·
DOI:https://doi.org/10.2172/861637· OSTI ID:861637
 [1]
  1. Iowa State Univ., Ames, IA (United States)

Conventional implementations of collective communications are based on point-to-point communications, and their optimizations have been focused on efficiency of those communication algorithms. However, point-to-point communications are not the optimal choice for modern computing clusters of SMPs due to their two-level communication structure. In recent years, a few research efforts have investigated efficient collective communications for SMP clusters. This dissertation is focused on platform-independent algorithms and implementations in this area. There are two main approaches to implementing efficient collective communications for clusters of SMPs: using shared memory operations for intra-node communications, and overlapping inter-node/intra-node communications. The former fully utilizes the hardware based shared memory of an SMP, and the latter takes advantage of the inherent hierarchy of the communications within a cluster of SMPs. Previous studies focused on clusters of SMP from certain vendors. However, the previously proposed methods are not portable to other systems. Because the performance optimization issue is very complicated and the developing process is very time consuming, it is highly desired to have self-tuning, platform-independent implementations. As proven in this dissertation, such an implementation can significantly out-perform the other point-to-point based portable implementations and some platform-specific implementations. The dissertation describes in detail the architecture of the platform-independent implementation. There are four system components: shared memory-based collective communications, overlapping mechanisms for inter-node and intra-node communications, a prediction-based tuning module and a micro-benchmark based tuning module. Each component is carefully designed with the goal of automatic tuning in mind.

Research Organization:
Ames Lab., Ames, IA (United States)
Sponsoring Organization:
USDOE
DOE Contract Number:
W-7405-ENG-82
OSTI ID:
861637
Report Number(s):
IS--T 1987
Country of Publication:
United States
Language:
English

Similar Records

One-Sided Communication on the Myrinet-Based SMP Clusters Using the GM Message-Passing Library
Conference · Thu Dec 06 23:00:00 EST 2001 · OSTI ID:951046

Exploration of SMP-Aware DAO Memory Performance Issues-Final Report 2002
Technical Report · Mon Feb 03 23:00:00 EST 2003 · OSTI ID:15003253

Distributed out-of-memory NMF on CPU/GPU architectures
Journal Article · Fri Sep 08 00:00:00 EDT 2023 · Journal of Supercomputing · OSTI ID:2246858