Accelerating Graph Processing on Large-scale Multicores
Digital Document
Handle |
Handle
http://hdl.handle.net/11134/20002:860654665
|
||||||
---|---|---|---|---|---|---|---|
Persons |
Persons
Creator (cre): Ahmad, Masab
Major Advisor (mja): Khan, Omer
Associate Advisor (asa): Van Dijk, Marten
Associate Advisor (asa): Chandy, John
|
||||||
Title |
Title
Title
Accelerating Graph Processing on Large-scale Multicores
|
||||||
Origin Information |
Origin Information
|
||||||
Parent Item |
Parent Item
|
||||||
Resource Type |
Resource Type
|
||||||
Digital Origin |
Digital Origin
born digital
|
||||||
Description |
Description
With the ever-increasing amount of data and input variations, portable performance is becoming harder to exploit on today’s architectures. Computational setups utilize single-chip processors, such as GPUs or large-scale multicores for graph analytics. Some algorithm-input combinations perform more efficiently when utilizing a GPU’s higher concurrency and bandwidth, while others perform better with a multicore’s stronger data caching capabilities. Architectural choices also occur within selected accelerators, where variables such as threading and thread placement need to be decided for optimal performance. This paper proposes a performance predictor paradigm for a heterogeneous parallel architecture where multiple disparate accelerators are integrated in an operational high performance computing setup. The predictor aims to improve graph processing efficiency by exploiting the underlying concurrency variations within and across the heterogeneous integrated accelerators using graph benchmark and input characteristics. The evaluation shows that intelligent and real-time selection of near-optimal concurrency choices provides performance benefits ranging from 5% to 3.8x, and an energy benefit averaging around 2.4x over the traditional single-accelerator setup.
|
||||||
Genre |
Genre
|
||||||
Organizations |
Organizations
Degree granting institution (dgg): University of Connecticut
|
||||||
Held By | |||||||
Rights Statement |
Rights Statement
|
||||||
Use and Reproduction |
Use and Reproduction
These materials are provided for educational and research purposes only.
|
||||||
Note |
Note
|
||||||
Local Identifier |
Local Identifier
OC_d_2329
|