Robust Processor Allocation for Independent Tasks When Dollar Cost for Processors is a Constraint

Sugavanam, Prasanna; Siegel, Howard Jay; Maciejewski, Anthony A.; Zhang, Junxing; Shestak, M.; Raskey, M.; Pippin, Alan; Pichel, Ron; Oltikar, M.; Mehta, Ashish M.; Lee, Panho; Krishnamurthy, Yogish; Horiuchi, A.; Guru, K.; Aydin, M.; Alotaibi, Mohammad T.; Ali, Shoukat

doi:10.1109/clustr.2005.347023

Cited by 5 publications

(5 citation statements)

References 37 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Work presented in [9,14,15,17] assumes a-priori known task execution times. One relaxation is found in [4] where all tasks are assumed to be in the same complexity class and there is a calibration step to determine execution time estimates per machine type.…”

Section: Related Workmentioning

confidence: 99%

Bag-of-Tasks Scheduling under Budget Constraints

Oprescu

Kielmann

2010

2010 IEEE Second International Conference on Cloud Computing Technology and Science

114

View full text Add to dashboard Cite

Commercial cloud offerings, such as Amazon's EC2, let users allocate compute resources on demand, charging based on reserved time intervals. While this gives great flexibility to elastic applications, users lack guidance for choosing between multiple offerings, in order to complete their computations within given budget constraints. In this work, we present BaTS, our budget-constrained scheduler. BaTS can schedule large bags of tasks onto multiple clouds with different CPU performance and cost, minimizing completion time while respecting an upper bound for the budget to be spent. BaTS requires no a-priori information about task completion times, and learns to estimate them at run time. We evaluate BaTS by emulating different cloud environments on the DAS-3 multi-cluster system. Our results show that BaTS is able to schedule within a user-defined budget (if such a schedule is possible at all.) At the expense of extra compute time, signifcant cost savings can be achieved when comparing to a cost-oblivious round-robin scheduler.

show abstract

Section: Related Workmentioning

confidence: 99%

Bag-of-Tasks Scheduling under Budget Constraints

Oprescu

Kielmann

2010

2010 IEEE Second International Conference on Cloud Computing Technology and Science

114

View full text Add to dashboard Cite

show abstract

“…Work presented in [11,12,13,14] assumes a-priori known task execution times. One relaxation is found in [15] where all tasks are assumed to be in the same complexity class and there is a calibration step to determine execution time estimates per machine type.…”

Section: Related Workmentioning

confidence: 99%

Budget Estimation and Control for Bag-of-Tasks Scheduling in Clouds

Oprescu

Kielmann

Leahu

2011

Parallel Process. Lett.

View full text Add to dashboard Cite

Commercial cloud offerings, such as Amazon's EC2, let users allocate compute resources on demand, charging based on reserved time intervals. While this gives great flexibility to elastic applications, users lack guidance for choosing between multiple offerings, in order to complete their computations within given budget constraints. In this work, we present BaTS, our budget-constrained scheduler. Using a small task sample, BaTS can estimate costs and makespan for a given bag on different cloud offerings. It provides the user with a choice of options before execution and then schedules the bag according to the user's preferences. BaTS requires no a-priori information about task completion times. We evaluate BaTS by emulating different cloud environments on the DAS-3 multicluster system. Our results show that BaTS correctly estimates budget and makespan for the scenarios investigated; the user-selected schedule is then executed within the given budget limitations.

show abstract

“…A metric for the robustness of a resource allocation with respect to desired system performance features against perturbations in system and environmental conditions, and the experiments conducted to illustrate the utility of the robustness metric, are sum Heuristics developed to generate mappings of independent applications in parallel systems such that the robustness of the produced mappings is maximized are summarized from [16]. Finally, heuristics for (1) selecting a set of machines and (2) mapping applications to the set of machines, both to maximize robustness, are summarized from [17].…”

Section: Discussionmentioning

confidence: 99%

“…This section summarizes the study described in [17], which was an extension of [16]. The research environment here differs from the previous study with the addition of the cost constraint for the machines and choosing a subset of all the available machines to be used.…”

Section: Problem Statementmentioning

confidence: 99%

“…For sub-problem 2, the machine assignment heuristics described in the previous section are used as components of the heuristics developed in this research. A method used to generate 100 high application and low machine heterogeneity (high-low) ETC matrices for 1024 independent applications was identical to that used in the previous work (see the details of the simulation setup in [17]). Experiments with simple greedy heuristics were used to decide the value of the cost constraint to be 34,800 dollars and the time constraint to be 12,000 seconds.…”

Section: Problem Statementmentioning

confidence: 99%

See 1 more Smart Citation