Task admission control policy in cloud server pools based on task arrival dynamics
Haleh Khojasteh, Jelena Mišić
Abstract
Haleh Khojasteh, Jelena Mišić
Abstract
Abstract In this paper, we propose two task admission control algorithms that utilize random task filtering: a lightweight algorithm based on long‐term estimates of average utilization and offered load, and a more complex algorithm based on instantaneous utilization. Detailed performance evaluation confirms that both algorithms are able to ensure that the system remains in the stable operating region. We have also found that more aggressive filtering tends to decrease task blocking rate and delay. Copyright © 2016 John Wiley & Sons, Ltd.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract In this paper, we propose two task admission control algorithms that utilize random task filtering: a lightweight algorithm based on long‐term estimates of average utilization and offered load, and a more complex algorithm based on instantaneous utilization. Detailed performance evaluation confirms that both algorithms are able to ensure that the system remains in the stable operating region. We have also found that more aggressive filtering tends to decrease task blocking rate and delay. Copyright © 2016 John Wiley & Sons, Ltd.
Key concepts: Computer science, Task (project management), Cloud server, Blocking (statistics), Control (management), Term (time), Admission control, Dynamics (music)