Title :
Distributed Scheduling of Parallel I/O in the Presence of Data Replication
Author :
Wu, Jan-Jan ; Liu, Pangfeng
Author_Institution :
Inst. of Inf. Sci., Acad. Sinica, Taipei, Taiwan
Abstract :
This paper studies distributed scheduling of parallel I/O data transfers on systems that provide data replication. In our previous work, we proposed a centralized algorithm for solving this problem in systems where data transfer information is centrally available. This algorithm finds the optimal scheduling by constructing augmenting paths in the data transfer bipartite graph, requiring O(nmlog n + {text{n}}^{text{2}} {text{log}}^{frac{3}{2}} n) time, with n nodes and m edges in the bipartite graph. In this paper, we investigate this scheduling problem in distributed systems where data transfer information may not be centrally available. We propose a distributed scheduling algorithm, Highest Degree Lowest Workload First (HDLWF), which approximates the augmenting path algorithm in distributed environments. HDLWF is based on a distributed, two-step scheme that determines appropriate execution order of data requests through a small number of rounds of bidding between clients and I/O servers. Our experimental results indicate that HDLWF yields schedules close to the centralized optimal solution, and in some cases within 3% of the optimal solution.
Keywords :
client-server systems; distributed algorithms; input-output programs; parallel processing; resource allocation; scheduling; HDLWF; augmenting path algorithm; data replication; distributed scheduling algorithm; distributed systems; highest degree lowest workload first; parallel I/O data transfer; Bipartite graph; Computer science; Concurrent computing; Data engineering; Information science; Multimedia databases; Optimal scheduling; Processor scheduling; Read-write memory; Scheduling algorithm;
Conference_Titel :
Parallel and Distributed Processing Symposium, 2005. Proceedings. 19th IEEE International
Print_ISBN :
0-7695-2312-9
DOI :
10.1109/IPDPS.2005.171