▲ The first authors of the four IEEE/ACM CCGRID '25 regular papers

(From top left) Jung Yeon-woo, doctoral student; Byun Hong-su, combined master's–doctoral student

(From bottom left) Kim Ji-seop, master's student; Oh Se-jung, master's student; Kim Yu-chan, master's student

The Data-Centric Computing and Systems Laboratory (DISCOS), jointly supervised by Professor Park Sung-yong and Professor Kim Young-jae, has had a total of four papers published at the IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing. IEEE/ACM CCGRID covers a wide range of topics in clusters, clouds and distributed systems, chiefly for machine learning, and is a prestigious international conference rated outstanding in the Korean Institute of Information Scientists and Engineers' 2024 (revised) list of outstanding conferences.

This year 219 papers were submitted, of which 55 were finally accepted for publication. Four papers accepted from a single laboratory is a highly unusual achievement and is regarded as evidence of the laboratory's outstanding research capability.

MemoryBridge (first author: Kim Yu-chan) is the result of international collaborative research between Professor Kim Young-jae and Professor Weikuan Yu (Florida State University), while ECO-KVS (first authors: Kim Ji-seop and Byun Hong-su) is the outcome of international collaborative research carried out by Professor Park Sung-yong with Professor Kim Young-jae and Professor Zaipeng Xie (Hohai University). The EQULIBRIA paper (first author: Oh Se-jung) is the result of collaborative research between Professor Park Sung-yong and Professor Moon Eui-hyun. The DISCOS laboratory is thus producing outstanding research results through active collaboration among faculty within the Department of Computer Science & Engineering as well as through international collaboration.

Details of the four regular papers published at IEEE/ACM CCGRID '25 are given below.

1. StreamRAG: A High-Performance Lock-Aware and Traffic-Aware Query Coordinator in RAG System, Yeonwoo Jeong, Kyuli Park and Sungyong Park, In Proceedings of the IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2025), Tromsø, Norway, May, 2025.

2. ECO-KVS: Energy-aware Compaction Offloading Mechanism for LSM-tree based Key-Value Stores in Edge Federation, Jeeseob Kim*, Hongsu Byun*, Seungjae Lee, Myoungjoon Kim, Youngjae Kim, Zaipeng Xie and Sungyong Park (*joint first authors), In Proceedings of the IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2025), Tromsø, Norway, May, 2025.

3. EQUILIBRIA : Co-Optimizing Energy and Latency in Online ML-based Stream Processing Systems, Sejeong Oh, Gordon Euhyun Moon, Sungyong Park, In Proceedings of the IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2025), Tromsø, Norway, May, 2025.

4. MemoryBridge: Leveraging Cloud Resource Characteristics for Cost-Efficient Disk-based GNN Training via Two-Level Architecture

Yoo-Chan Kim, Hong-Yeon Kim, Weikuan Yu, Youngjae Kim
In Proceedings of the IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2025), Tromsø, Norway, May, 2025.