2019
DOI: 10.1007/s10586-018-02902-0
|View full text |Cite
|
Sign up to set email alerts
|

Learning budget assignment policies for autoscaling scientific workflows in the cloud

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
14
0
8

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
2

Relationship

1
5

Authors

Journals

citations
Cited by 8 publications
(22 citation statements)
references
References 13 publications
0
14
0
8
Order By: Relevance
“…Thus, although most of the proposed solutions for the autoscaling problem of workflows in Cloud are based on heuristics, several recent research aims to apply reinforcement learning approaches to solve any of the subproblems involved, the scaling [10,11,12,13,14,15,16,8] or scheduling [17,18,19,20,21,22,23], which are characterized in Cloud by the conditions of uncertainty.…”
Section: Scalingmentioning
confidence: 99%
See 3 more Smart Citations
“…Thus, although most of the proposed solutions for the autoscaling problem of workflows in Cloud are based on heuristics, several recent research aims to apply reinforcement learning approaches to solve any of the subproblems involved, the scaling [10,11,12,13,14,15,16,8] or scheduling [17,18,19,20,21,22,23], which are characterized in Cloud by the conditions of uncertainty.…”
Section: Scalingmentioning
confidence: 99%
“…The uncertainty in these problems is associated with the variability in the performance of the Cloud infrastructure. In this sense, proposals that model the autoscaling problem as an MDP and use different RL techniques to learn adequate scaling policies [10,11,12,13,14,15,16,8] or scheduling [17,18,19,20,21,22,23] have appeared. These policies allow an autoscaler to determine which action is more convenient at any time, in order to optimize in the long-term one or more objectives from the point of view of the execution of one, or a set of applications/workflows.…”
Section: Review Of Cloud Autoscaling Based On Rl Techniquesmentioning
confidence: 99%
See 2 more Smart Citations
“…In a single-tier application, the response time prediction is sufficient to scale the resources [10]. However, the multi-tier applications workload is complex in nature due to the multi-tier architecture and flash traffic [26]. Thus, the operational cost and SLA penalty can be minimized with the autonomic scaling process.…”
Section: Introductionmentioning
confidence: 99%