A Comparison of System Performance on a Private OpenStack Cloud and Amazon EC2

被引：7
作者：
Kang, Mikyung [1 ]
Kang, Dong-In [1 ]
Walters, John Paul [1 ]
Crago, Stephen P. [1 ]
机构：
[1] Univ Southern Calif, Informat Sci Inst, Arlington, VA 22203 USA
来源：
2017 IEEE 10TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD) | 2017年
基金：
美国国家科学基金会;
关键词：
Cloud; performance; PerfKitBenchmarker; AWS; OpenStack; Chameleon; benchmarking;
D O I：
10.1109/CLOUD.2017.47
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
Cloud computing is becoming increasingly pervasive and is being adopted even for high performance computing and mission critical applications. As cloud computing extends its usage, understanding of its performance becomes more important. In this paper, we present the system performance using Amazon EC2, representing a large public cloud platform, and OpenStack, representing the most popular open-source infrastructure-as-a-service cloud platform that can be used to implement both private and public clouds. Our study covers the performance of all significant cloud infrastructure resources - CPU, memory, storage, and network. We compare the performance of the public Amazon EC2 and a private OpenStack platform with similar configurations using several benchmarks. We compare and analyze diverse performance aspects based on different hypervisor, storage, and/or network configurations. For the OpenStack platform, we measure virtualization overhead for each resource. We also investigated the performance impact of more advanced features of each cloud, including faster storage options of AWS and the Infiniband network fabric with the Chameleon OpenStack platform.
引用
收藏
页码：310 / 317
页数：8
相关论文
共 12 条
[1]   Heterogeneous Cloud Computing: The Way Forward [J].
Crago, Stephen P. ;
Walters, John Paul .
COMPUTER, 2015, 48 (01) :59-61
[2]   Heterogeneous cloud computing [J].
Crago, Steve ;
Dunn, Kyle ;
Eads, Patrick ;
Hochstein, Lorin ;
Kang, Dong-In ;
Kang, Mikyung ;
Modium, Devendra ;
Singh, Karandeep ;
Suh, Jinwoo ;
Walters, John Paul .
2011 IEEE INTERNATIONAL CONFERENCE ON CLUSTER COMPUTING (CLUSTER), 2011, :378-385
[3]   Study of the KVM CPU performance of Open-Source cloud management platforms [J].
Gomez-Folgar, F. ;
Garcia-Loureiro, A. J. ;
Pena, T. F. ;
Zablah, J. I. ;
Seoane, N. .
2015 15TH IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND GRID COMPUTING, 2015, :1225-1228
[4]  
Hwang JH, 2013, 2013 IFIP/IEEE INTERNATIONAL SYMPOSIUM ON INTEGRATED NETWORK MANAGEMENT (IM 2013), P269
[5]   Next Generation Clouds, The Chameleon Cloud Testbed, and Software Defined Networking (SDN) [J].
Mambretti, Joe ;
Chen, Jim ;
Yeh, Fei .
2015 INTERNATIONAL CONFERENCE ON CLOUD COMPUTING RESEARCH AND INNOVATION (ICCCRI), 2015, :73-79
[6]  
Maraj Arianit, 2013, Proceedings of the 2013 55th International Symposium. ELMAR-2013, P239
[7]   Bridging the Virtualization Performance Gap for HPC Using SR-IOV for InfiniBand [J].
Musleh, Malek ;
Pai, Vijay ;
Walters, John Paul ;
Younge, Andrew ;
Crago, Stephen P. .
2014 IEEE 7TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD), 2014, :627-635
[8]   Performance Analysis of an I/O-Intensive Workflow executing on Google Cloud and Amazon Web Services [J].
Nawaz, Hassan ;
Juve, Gideon ;
da Silva, Rafael Ferreira ;
Deelman, Ewa .
2016 IEEE 30TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2016, :535-544
[9]   A Comparative Study on I/O Performance between Compute and Storage Optimized Instances of Amazon EC2 [J].
Shoeb, Abu Awal Md ;
Hasan, Ragib ;
Hague, Md. ;
Hu, Meng .
2014 IEEE 7TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD), 2014, :970-971
[10]   GPU Passthrough Performance: A Comparison of KVM, Xen, VMWare ESXi, and LXC for CUDA and OpenCL Applications [J].
Walters, John Paul ;
Younge, Andrew J. ;
Kang, Dong-In ;
Yao, Ke-Thia ;
Kang, Mikyung ;
Crago, Stephen P. ;
Fox, Geoffrey C. .
2014 IEEE 7TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD), 2014, :636-643
12