skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: APHiD: Hierarchical Task Placement to Enable a Tapered Fat Tree Topology for Lower Power and Cost in HPC Networks

Abstract

The power and procurement cost of bandwidth in system-wide networks has forced a steady drop in the byte/flop ratio. This trend of computation becoming faster relative to the network is expected to hold. In this paper, we explore how cost-oriented task placement enables reducing the cost of system-wide networks by enabling high performance even on tapered topologies where more bandwidth is provisioned at lower levels. We describe APHiD, an efficient hierarchical placement algorithm that uses new techniques to improve the quality of heuristic solutions and reduces the demand on high-level, expensive bandwidth in hierarchical topologies. We apply APHiD to a tapered fat-tree, demonstrating that APHiD maintains application scalability even for severely tapered network configurations. Using simulation, we show that for tapered networks APHiD improves performance by more than 50% over random placement and even 15% in some cases over costlier, state-of-the-art placement algorithms.

Authors:
; ; ; ; ;
Publication Date:
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
Computational Research Division
OSTI Identifier:
1344101
Report Number(s):
LBNL-1007126
ir:1007126
Resource Type:
Conference
Resource Relation:
Conference: 2017 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, Melia Los Galgos Hotel: Calle de Claudio Coello, 139 Madrid 28006 Spain, 05/14/2017
Country of Publication:
United States
Language:
English

Citation Formats

Michelogiannakis, George, Ibrahim, Khaled Z., Shalf, John, Wilke, Jeremiah J., Knight, Samuel, and Kenny, Joseph P. APHiD: Hierarchical Task Placement to Enable a Tapered Fat Tree Topology for Lower Power and Cost in HPC Networks. United States: N. p., 2017. Web.
Michelogiannakis, George, Ibrahim, Khaled Z., Shalf, John, Wilke, Jeremiah J., Knight, Samuel, & Kenny, Joseph P. APHiD: Hierarchical Task Placement to Enable a Tapered Fat Tree Topology for Lower Power and Cost in HPC Networks. United States.
Michelogiannakis, George, Ibrahim, Khaled Z., Shalf, John, Wilke, Jeremiah J., Knight, Samuel, and Kenny, Joseph P. Sun . "APHiD: Hierarchical Task Placement to Enable a Tapered Fat Tree Topology for Lower Power and Cost in HPC Networks". United States. doi:. https://www.osti.gov/servlets/purl/1344101.
@article{osti_1344101,
title = {APHiD: Hierarchical Task Placement to Enable a Tapered Fat Tree Topology for Lower Power and Cost in HPC Networks},
author = {Michelogiannakis, George and Ibrahim, Khaled Z. and Shalf, John and Wilke, Jeremiah J. and Knight, Samuel and Kenny, Joseph P.},
abstractNote = {The power and procurement cost of bandwidth in system-wide networks has forced a steady drop in the byte/flop ratio. This trend of computation becoming faster relative to the network is expected to hold. In this paper, we explore how cost-oriented task placement enables reducing the cost of system-wide networks by enabling high performance even on tapered topologies where more bandwidth is provisioned at lower levels. We describe APHiD, an efficient hierarchical placement algorithm that uses new techniques to improve the quality of heuristic solutions and reduces the demand on high-level, expensive bandwidth in hierarchical topologies. We apply APHiD to a tapered fat-tree, demonstrating that APHiD maintains application scalability even for severely tapered network configurations. Using simulation, we show that for tapered networks APHiD improves performance by more than 50% over random placement and even 15% in some cases over costlier, state-of-the-art placement algorithms.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sun May 14 00:00:00 EDT 2017},
month = {Sun May 14 00:00:00 EDT 2017}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: