Abstract | ||
---|---|---|
Hardware accelerators are known to be performance and power efficient. This article focuses on accelerator design for graph analytics applications, which are commonly used kernels for cognitive systems. The authors propose a templatized architecture that is specifically optimized for vertex-centric graph applications with irregular memory access patterns, asynchronous execution, and asymmetric convergence. The proposed architecture addresses the limitations of existing CPU and GPU systems while providing a customizable template. The authors' experiments show that the generated accelerators can outperform a high-end CPU system with up to 3 times better performance and 65 times better power efficiency. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/MM.2017.7 | IEEE Micro |
Keywords | Field | DocType |
Graph analytics,Data structures,Computational modeling,Synchronization,Random access memory,Scheduling,Accelerators,Parallel processing | Convergence (routing),Electrical efficiency,Asynchronous communication,Data structure,Synchronization,Architecture,Computer science,Scheduling (computing),Parallel computing,Real-time computing,Memory architecture | Journal |
Volume | Issue | ISSN |
37 | 1 | 0272-1732 |
Citations | PageRank | References |
1 | 0.39 | 3 |
Authors | ||
7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Muhammet Mustafa Ozdal | 1 | 313 | 23.18 |
Serif Yesil | 2 | 6 | 1.86 |
Taemn Kim | 3 | 382 | 28.18 |
Andrey Ayupov | 4 | 112 | 7.12 |
John Greth | 5 | 1 | 0.39 |
Steven M. Burns | 6 | 563 | 104.03 |
Ozcan Ozturk | 7 | 112 | 15.25 |