A collection of powerful workstations interconnected by a local area network can be utilized as compute servers when left idle by their owners. DAWGS allows users to submit jobs for execution on an idle workstation somewhere on a local area network. DAWGS uses a distributed scheduler and a bidding scheme to determine on which machine to run a process. DAWGS can properly redirect all the I/O of a remotely executing process and can checkpoint and then subsequently restart the process, even if the restart is on a different machine than the checkpoint. Our method is different from other work in that it is fault-tolerant and fully distributed.
H. Clark and B. M. McMillin, "DAWGS - A Distributed Compute Server Utilizing Idle Workstations," Proceedings of the 5th Distributed Memory Computing Conference, DMCC 1990, vol. 2, pp. 732 - 741, article no. 556276, Institute of Electrical and Electronics Engineers, Jan 1990.
The definitive version is available at https://doi.org/10.1109/DMCC.1990.556276
International Standard Book Number (ISBN)
Article - Conference proceedings
© 2023 Institute of Electrical and Electronics Engineers, All rights reserved.
01 Jan 1990