Journal article
Exploring PIM Architecture for High-Performance Graph Pattern Mining
IEEE computer architecture letters, Vol.20(2), pp.114-117
07/2021
DOI: 10.1109/LCA.2021.3103665
Abstract
Graph mining applications, such as subgraph pattern matching and mining, are widely used in real-world domains such as bioinformatics, social network analysis, and computer vision. Such applications are considered as a new class of data-intensive applications that generate massive irregular computation workloads and memory accesses, which are different from many well-studied graph applications such as BFS and page rank. In this letter, we use the emerging process-in-memory architecture to accelerate data-intensive operations in graph mining tasks. We first identify the code blocks that are best suitable for PIM execution. Then, we observe a significant load imbalance on PIM architecture and analyze the root cause for such imbalance in graph mining applications. Lastly, we evaluate several scheduling schemes that help reduce the load imbalance and discuss potential optimizations to enhance performance further.
Details
- Title: Subtitle
- Exploring PIM Architecture for High-Performance Graph Pattern Mining
- Creators
- Jiya Su - Illinois Institute of TechnologyLinfeng He - University of IowaPeng Jiang - University of IowaRujia Wang - Illinois Institute of Technology
- Resource Type
- Journal article
- Publication Details
- IEEE computer architecture letters, Vol.20(2), pp.114-117
- Publisher
- IEEE
- DOI
- 10.1109/LCA.2021.3103665
- ISSN
- 1556-6056
- eISSN
- 1556-6064
- Grant note
- CCF-2029014; CCF-2028825 / National Science Foundation (10.13039/100000001)
- Language
- English
- Date published
- 07/2021
- Academic Unit
- Computer Science
- Record Identifier
- 9984259435102771
Metrics
14 Record Views