SHMEMGraph: Efficient and Balanced Graph Processing Using One-Sided Communication
- Florida State University, Tallahassee
- ORNL
State-of-the-art synchronous graph processing frameworks face both inefficiency and imbalance issues that cause their performance to be suboptimal. These issues include the inefficiency of communication and the imbalanced graph computation/communication costs in an iteration. We propose to replace their conventional two-sided communication model with the one-sided counterpart. Accordingly, we design SHMEMGraph, an efficient and balanced graph processing framework that is formulated across a global memory space and takes advantage of the flexibility and efficiency of one-sided communication for graph processing. Through an efficient one-sided communication channel, SHMEMGraph utilizes the high-performance operations with RDMA while minimizing the resource contention within a computer node. In addition, SHMEMGraph synthesizes a number of optimizations to address both computation imbalance and communication imbalance. By using a graph of 1 billion edges, our evaluation shows that compared to the state-of-the-art Gemini framework, SHMEMGraph achieves an average improvement of 35.5% in terms of job completion time for five representative graph algorithms.
- Research Organization:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1468157
- Resource Relation:
- Conference: IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing - Washington D.C, District of Columbia, United States of America - 5/1/2018 12:00:00 PM-5/4/2018 12:00:00 PM
- Country of Publication:
- United States
- Language:
- English
One trillion edges: graph processing at Facebook-scale
|
journal | August 2015 |
Mizan-RMA: Accelerating Mizan Graph Processing Framework with MPI RMA
|
conference | December 2016 |
Trinity: a distributed graph engine on a memory cloud
|
conference | January 2013 |
Scalable Graph500 design with MPI-3 RMA
|
conference | September 2014 |
Mizan
|
conference | April 2013 |
Layered label propagation: a multiresolution coordinate-free ordering for compressing social networks
|
conference | January 2011 |
Pregel: a system for large-scale graph processing
|
conference | January 2010 |
SYNC or ASYNC: time to fuse for distributed graph-parallel computation
|
journal | December 2015 |
High-Performance Key-Value Store on OpenSHMEM
|
conference | May 2017 |
X-Stream: edge-centric graph processing using streaming partitions
|
conference | January 2013 |
Designing Scalable Out-of-core Sorting with Hybrid MPI+PGAS Programming Models
|
conference | October 2014 |
To Push or To Pull
|
conference | June 2017 |
Ligra
|
journal | February 2013 |
An experimental comparison of pregel-like graph processing systems
|
journal | August 2014 |
G
|
conference | August 2015 |
Distributed GraphLab: a framework for machine learning and data mining in the cloud
|
journal | April 2012 |
The webgraph framework I: compression techniques
|
conference | January 2004 |
Introducing OpenSHMEM: SHMEM for the PGAS community
|
conference | January 2010 |
Similar Records
Final Report for Project DE-FC02-06ER25755 [Pmodels2]
Decomposition of Large Scale Semantic Graphsvia an Efficient Communities Algorithm