A paper titled 'Crocus: Enabling Computing Resource Orchestration for Inline Cluster-wide Deduplication on Scalable Storage Systems', written by Awais Khan and Lee Chang-gyu (combined master's–doctoral students, supervisor: Professor Kim Young-jae) of the Department of Computer Science & Engineering together with Prince Hamadawana (combined master's–doctoral student at Ajou University, first author), has been accepted for publication in IEEE Transactions on Parallel and Distributed Systems (TPDS), the world's most prestigious international journal in parallel and distributed systems. The research proposes a scheduling framework that can dynamically allocate the fingerprinting computation for data deduplication to available computing resources (CPU, GPU) within a cluster in the cloud storage system Ceph (an open-source object storage platform). The GPU-based computation offloading technology in this research in particular was carried out with the help of Professor Park Sung-yong of the Department of Computer Science & Engineering, an expert in GPU virtualization.

▲ Awais Khan (combined master's–doctoral student), Lee Chang-gyu (combined master's–doctoral student), Professor Park Sung-yong, Professor Kim Young-jae (corresponding author)

▲ IEEE Transactions on Parallel & Distributed Systems (left); first page of the Crocus paper (right)

IEEE Transactions on Parallel and Distributed Systems (TPDS) is a distinguished international journal covering research in parallel and distributed systems. The specific research areas it covers include parallel algorithms, big data applications, large-scale social network analysis, big data management, cloud and grid computing, parallel and distributed architectures, the design, analysis, implementation, fault recovery and performance measurement of multiprocessor systems, multicore processors, heterogeneous many-core systems, parallel and distributed software, parallel programming paradigms, and programming environments and tools.

(Source: https://www.computer.org/csdl/journal/td)