Accord: Accelerating Big Data Applications on Heterogeneous Processors

Many applications require fast access, processing, and analytics of their big data, for example, millions of image files, tables containing hundreds of millions of rows, and billion-edge graphs. With the rapid advances of hardware technologies, these applications run on powerful servers with heterogeneous processors, computer clusters, and cloud computing platforms. However, the data handling and processing is often not in accordance with the underlying hardware resources. Therefore, we proposed to develop software techniques that accelerate these applications by utilizing their hardware resources effectively. In particular, we work on data-parallel primitives as well as parallel algorithms on graph analytics and relational queries. The processors we consider include multicore CPUs, Intel Knights Landing (KNL) processors, and GPUs.

Publications

Yulin Che, Zhuohang Lai, Shixuan Sun, Yue Wang, Qiong Luo: Accelerating Truss Decomposition on Heterogeneous Processors. Proc. VLDB Endow. 13(10): 1751-1764 (2020)
Lipeng Wang, Songgao Ye, Baichen Yang, Youyou Lu, Hequan Zhang, Shengen Yan, Qiong Luo: DIESEL: A Dataset-Based Distributed Storage and Caching System for Large-Scale Deep Learning Training. ICPP 2020: 20:1-20:11
Shixuan Sun, Qiong Luo: In-Memory Subgraph Matching: An In-depth Study. SIGMOD Conference 2020: 1083-1098
Shixuan Sun, Yulin Che, Lipeng Wang, Qiong Luo: Efficient Parallel Subgraph Enumeration on a Single Machine. ICDE 2019: 232-243
Shixuan Sun, Qiong Luo: Scaling Up Subgraph Query Processing with Efficient Subgraph Matching. ICDE 2019: 220-231
Yulin Che, Zhuohang Lai, Shixuan Sun, Qiong Luo, Yue Wang: Accelerating All-Edge Common Neighbor Counting on Three Processors. ICPP 2019: 42:1-42:10
Zhuohang Lai, Qiong Luo, Xiaolong Xie: Efficient Data-Parallel Primitives on Heterogeneous Systems. ICPP 2019: 74:1-74:10
Shixuan Sun, Qiong Luo: Parallelizing Recursive Backtracking Based Subgraph Matching on a Single Machine. ICPADS 2018: 42-50
Zhuohang Lai, Qiong Luo, Xiaoying Jia: Revisiting Multi-pass Scatter and Gather on GPUs. ICPP 2018: 25:1-25:11
Yulin Che, Shixuan Sun, Qiong Luo: Parallelizing Pruning-based Graph Structural Clustering. ICPP 2018: 77:1-77:10
Lipeng Wang, Xiaoying Jia, Qiong Luo: Betweenness Centrality Revisited on Four Processors. ICPADS 2017: 614-623

Source Code Repository

Rapids at HKUST on GitHub

Acknowledgement

We thank our industrial sponsors Alibaba and SenseTime for their generous support.