LSM tree-based key-value stores face significant I/O bandwidth consumption and performance bottlenecks due to frequent data rewrites and migrations during compaction. To address this issue, near-data processing technology has emerged as a promising solution and is gaining increasing attention. Near-data processing reduces the data transfer distance between storage and processing resources by placing computational resources closer to storage devices or integrating them into memory, thereby effectively alleviating performance bottlenecks. However, existing multi-near-data processing (NDP) key-value stores still face synchronization problems, leading to long wait times and underutilization of resources. To address these issues, TSYS School computer scientist Yi Zhou and his colleagues from Anhui University, Zhongguancun Laboratory and Auburn University propose an asynchronous parallel compaction for multi-NDP-enabled key-value store named A+Store. In A+Store, to optimize data layout, they implement an MLSM-tree on each NDP device, an asynchronous execution queue for dynamic task management, and an independent metadata management method. This asynchronous mechanism allows each NDP device to update its metadata immediately after completing a compaction task rather than waiting for other devices, thereby eliminating synchronization waiting time among NDP devices. Additionally, as each NDP stores SSTables within specific key ranges, the device can perform sub-compaction tasks in parallel according to its key range, significantly enhancing the execution speed of tasks within each NDP device. This approach can improve the system’s parallel processing capability and resource utilization, addressing the bottlenecks in existing multi-NDP key-value stores in applications with the requirements of large-scale data processing and low latency. To evaluate the performance of A+Store, the study, which appears in the latest issue of the Journal of Systems Architecture, compares A+Store to state-of-the-art key-value stores. Zhou and his colleagues also develop a tested toolkit using the real-world dataset OpenAlex, and study the performance of A+Store under realistic workloads. Experimental results show that A+Store demonstrates superior performance across all tests.
Seven Turner College Management and Marketing Faculty Have Combined to Produce Eight A-Level Journal Publications Between 2021 and the Present
A number of faculty in the Turner College's Department of Management and Marketing, which includes faculty in management information systems, have produced A-level journal publications in the last few years. This report covers that activity, starting with John Finley , the chairperson of the department. Professor Finley published a paper in the Journal of Computer Information Systems in 2022. Finley is joined by Kirk Heriot , the Crowley Distinguished Chair in Entrepreneurship. Heriot, who earned a PhD in management from Clemson University, published in a 2021 issue of Small Business Economics . One of the study's co-authors, Andres Jauregui of Fresno State University, was previously a member of the Turner College's economics faculty. Next is Johnny Ho , a professor of management, who has a 2022 publication in the Journal of Computer Information Systems . Ho has won CSU's Excellence in Research Award on multiple occasions, while he has compiled 2...

Comments
Post a Comment