Migrating a National Cloud Platform for Urban Analytics: A Performance Assessment Framework

被引:0
作者
Pursultani, Hossein [1 ]
Sinnott, Richard O. [1 ]
机构
[1] Univ Melbourne, Sch Comp & Informat Syst, Melbourne, Vic, Australia
来源
11TH IEEE INTERNATIONAL CONFERENCE ON CLOUD COMPUTING TECHNOLOGY AND SCIENCE (CLOUDCOM 2019) | 2019年
关键词
Cloud Platforms; Urban Research; OpenStack; VMware;
D O I
10.1109/CloudCom.2019.00061
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The Australian Urban Research Infrastructure Network (AURIN wytm.aurin.ora.au) is a national platform (in Australia) for research into the urban and built environment that has been funded as part of the National Collaboration Research Infrastructure Strategy (NCRIS). The system provides secure, seamless access to over 5,000 (typically definitive) data sets from over 100 major organisations crossing government, industry and academia. The system has a rich and diverse user community with over 15,000 users and many actively using the systems on a regular basis with periodic spikes in usage, e.g. when large-scale urban and built environment classes are run. AURIN has been running on dedicated hardware miming VMware with redundancy/back-up systems miming on the OpenStack-based National eResearch Collaboration Tools and Resources (NeCTAR - www.nectar.org.au) Research Cloud. In 2019, AURIN was awarded a further $18.9m for a further 5-years as part of the NCRIS program. As such, there are decisions to be made on the future deployment of AURIN and whether to go fully onto a public Cloud such as NeCTAR or to continue to run on dedicated hardware. This paper explores the technical performance considerations that should underpin such a decision. We identify that even though NeCTAR performs better for most Cloud resources, it is less performant for disk operations which is essential for the majority of AURIN usecases.
引用
收藏
页码:354 / 361
页数:8
相关论文
共 12 条
[1]  
Iosup A, 2014, CLOUD COMPUTING DATA, P83, DOI DOI 10.1007/978-1-4939-1905-5_4
[2]  
Nino-Ruiz M., 2014, INT C E SCI 2014 SAO
[3]   Cloud WorkBench - Infrastructure-as-Code Based Cloud Benchmarking [J].
Scheuner, Joel ;
Leitner, Philipp ;
Cito, Jurgen ;
Gall, Harald .
2014 IEEE 6TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING TECHNOLOGY AND SCIENCE (CLOUDCOM), 2014, :246-253
[4]   A scalable Cloud-based system for data-intensive spatial analysis [J].
Sinnott, R. O. ;
Voorsluys, W. .
INTERNATIONAL JOURNAL ON SOFTWARE TOOLS FOR TECHNOLOGY TRANSFER, 2016, 18 (06) :587-605
[5]  
Sinnott R.O., 2015, IEEE INT C INF REUS
[6]  
Sinnott R.O., 2015, INT WORKSH SCI GAT B
[7]  
Sinnott R.O., 2014, J CONCURRENCY CO APR, DOI [10.1002/cpe.3282, DOI 10.1002/CPE.3282]
[8]  
Sinnott R.O., 2013, 11 AUSTR S PAR DISTR, P39
[9]   Privacy Preserving Geo-Linkage in the Big Urban Data Era [J].
Sinnott, Richard O. ;
Bayliss, Christopher ;
Bromage, Andrew ;
Galang, Gerson ;
Gong, Yikai ;
Greenwood, Philip ;
Jayaputera, Glenn ;
Marques, Davis ;
Morandini, Luca ;
Nogoorani, Ghazal ;
Pursultani, Hossein ;
Sarwar, Muhammad ;
Voorsluys, William ;
Widjaja, Ivo .
JOURNAL OF GRID COMPUTING, 2016, 14 (04) :603-618
[10]   The Australian Data-Driven Urban Research Platform: Systems Paper [J].
Sinnott, Richard O. .
AUSTRALIAN ECONOMIC REVIEW, 2016, 49 (02) :208-223