Utilization, Predictability, Workloads, and User Runtime Estimates in Scheduling the IBM SP2 with Backfilling

By Ahuva Mu'alem, Dror G. Feitelson. IEEE Trans. Parallel Distributed Syst., 2001

827 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Backfilling Using System-Generated Predictions Rather than User Runtime Estimates (2007, methodology): “We argue that in any case, explicit SJBF is more sensible than " tricking " EASY into SJFness by doubling [39], [26] or randomizing [28] estimates (see Section 6).”
GreenSlot: Scheduling energy consumption in green datacenters (2011, methodology): “Then, we present our model for solar energy prediction and discuss how GreenSlot adjusts the predictions when it ﬁnds inaccuracies.”
Evaluating the cost-benefit of using cloud computing to extend the capacity of clusters (2009, methodology): “This data structure is based on the idea of availability profile used in some implementations of conservative backfilling [24].”
Environment-conscious scheduling of HPC applications on distributed Cloud-oriented data centers (2011, methodology): “To increase the utilization of data centers and reduce the fragmentation in the scheduling of parallel applications, the local scheduler at each data center uses Conservative Backfilling with advance reservation support as proposed by Mu’alem and Feitelson [43].”