ExoApp: Performance Evaluation of Data-Intensive Applications on ExoGENI

被引:1
作者
Yu, Ze [1 ]
Liu, Xinxin [1 ]
Li, Min [1 ]
Liu, Kaikai [1 ]
Li, Xiaolin [1 ]
机构
[1] Univ Florida, Scalable Software Syst Lab, Gainesville, FL 32611 USA
来源
2013 SECOND GENI RESEARCH AND EDUCATIONAL EXPERIMENT WORKSHOP (GREE) | 2013年
关键词
MapReduce; Experiment; Data-intensive; ExoGENI;
D O I
10.1109/GREE.2013.14
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
ExoGENI is a new GENI-federated Infrastructure-as-a-Service (IaaS) framework. In this paper, we evaluate the performance of data-intensive applications on ExoGENI's resources. To simplify experiments, we design an automatic provisioning system called ExoApp. This paper focuses on MapReduce-based applications. Users can easily deploy applications in ExoGENI using ExoApp, without having to manually configure cluster runtime environments. We then conduct a series of experiments using real-world data sets and standard benchmarks through ExoApp. Our result shows that ExoGENI demonstrates similar resource quality when hosting data-intensive applications and its Network-as-a-Service (NaaS) model maintains stable network performance. We finally identify the pros and cons of the ExoGENI's NaaS model in supporting data-intensive applications.
引用
收藏
页码:25 / 28
页数:4
相关论文
共 7 条
[1]  
[Anonymous], 2011, TR20112 U CAL SAN DI
[2]  
[Anonymous], 2003, SOSP
[3]  
Baldine I., 2012, TRIDENTCOM INT C TES
[4]   Managing Data Transfers in Computer Clusters with Orchestra [J].
Chowdhury, Mosharaf ;
Zaharia, Matei ;
Ma, Justin ;
Jordan, Michael I. ;
Stoica, Ion .
ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2011, 41 (04) :98-109
[5]  
Dean J, 2004, USENIX ASSOCIATION PROCEEDINGS OF THE SIXTH SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION (OSDE '04), P137
[6]  
Dean Jeffrey, 2006, PACT P 15 INT C PAR, V16
[7]  
Thusoo Ashish, 2010, DAT ENG ICDE 2010 IE