Weijie Liu
ABSTRACT
To further improve the computing performance of the cloud for complex computing environments, efficient task scheduling strategies suitable for container deployment can improve the computing performance of the system. The current task scheduling algorithm is based on the current situation of the computing environment and the internal logical relationship between tasks to ensure that performance indicators are met. In this article, we first use lstm and attention algorithms to extract features from the historical data of processor operation to obtain better scheduling strategies in different states, and improve the existing processor scheduling strategies based on this. The method proposed in this paper has been experimentally verified in kubernate, which proves that the method in this paper is reasonable and effective.
- Meng X, Bradley J, Yavuz B, , 2015, “MLlib:machine learning in apache spark”, Journal of Machine LearningResearch, 17(1): 1235-1241.Google Scholar
- M. Abadi, P. Barham, , 2016, “TensorFlow:A System for Large-scale Machine Learning”, Proceedings of the 12th conference on Operating Systems Design and Implementation. USENIX, 265-283.Google Scholar
- Chen T, Li M, Li Y, , 2015, “MXNet:A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems”, Neural Information Processing Systems Workshop on Machine Learning Systems, 28-33.Google Scholar
- Frank Seide, Amit Agarwal, 2016, “CNTK:Microsoft's Open-Source Deep-Learning Toolkit”, SIGKDD International Conference on Knowledge Discovery and Data Mining, 2135-2135.Google Scholar
- S. Venkataraman, Z. Yang, M. Franklin, , 2016, “Ernest:Efficient Performance Prediction for Large-Scale Advanced Analytics”, Symposium on Networked SystemsDesign and Implementation. USENIX, 312-321.Google Scholar
- O. Alipourfard, H. H. Liu, , 2017, “CherryPick:AdaptivelyUnearthing the Best Cloud Configurations for Big Data An⁃alytics”, Networked Systems Design and Implementation. USENIX, 1-10.Google Scholar
- A. Ghodsi, M. Zaharia, , 2011, “Dominant Resource Fairness:Fair Allocation of Multiple Resource Types”, Networked Systems Design and Implementation. USENIX, 548-557.Google Scholar
- Jyothi S A, Curino C, Menache I, , 2016, “Morpheus:towards automated SLOs for enterprise clusters”, Conference on Operating Systems Design and Implementation.USENIX, 117-134.Google Scholar
- Ferguson A D, Bodik P, Kandula S, , 2012, “Jockey:guaranteed job latency in data parallel clusters”, European Conference on Computer Systems. EUROSYS, 99-112.Google Scholar
- Liu Z, Squillante M S, Wolf J L, 2001, “On maximizing service-level-agreement profits”, Acm Sigmetrics Performance Evaluation Review, 29(3): 43-44.Google ScholarDigital Library
Recommendations
Optimal Resource Usage in Multi-Cloud Computing Environment
Cloud computing has emerged as a new paradigm for accessing distributed computing resources such as infrastructure, hardware platform, and software applications on-demand over the internet as services. This paper presents an optimal resource management ...
Task scheduling techniques in cloud computing: A literature survey
AbstractCloud computing manages a variety of virtualized resources, which makes scheduling a critical component. In the cloud, a client may utilize several thousand virtualized assets for every task. Consequently, manual scheduling is not a ...
Highlights- Presents a comprehensive survey of task scheduling strategies and the associated metrics suitable for cloud computing environments.
Multi-job Associated Task Scheduling Based on Task Duplication and Insertion for Cloud Computing
Wireless Algorithms, Systems, and ApplicationsAbstractThe jobs processed in cloud computing systems may consist of multiple associated tasks which need to be executed under ordering constraints. The tasks of each job are run on different nodes, and communication is required to transfer data between ...
Comments