Application-Driven Bandwidth Guarantees in Datacenters

被引:30
作者
Lee, Jeongkeun [1 ]
Turner, Yoshio [1 ]
Lee, Myungjin [2 ]
Popa, Lucian [3 ]
Banerjee, Sujata [1 ]
Kang, Joon-Myung [1 ]
Sharma, Puneet [1 ]
机构
[1] HP Labs, Edinburgh, Midlothian, Scotland
[2] Univ Edinburgh, Edinburgh, Midlothian, Scotland
[3] Databricks, Edinburgh, Midlothian, Scotland
来源
SIGCOMM'14: PROCEEDINGS OF THE 2014 ACM CONFERENCE ON SPECIAL INTEREST GROUP ON DATA COMMUNICATION | 2014年
关键词
Datacenter; Bandwidth; Availability; Cloud; Virtual Network; Application;
D O I
10.1145/2619239.2626326
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Providing bandwidth guarantees to specific applications is becoming increasingly important as applications compete for shared cloud network resources. We present CloudMirror, a solution that provides bandwidth guarantees to cloud applications based on a new network abstraction and workload placement algorithm. An effective network abstraction would enable applications to easily and accurately specify their requirements, while simultaneously enabling the infrastructure to provision resources efficiently for deployed applications. Prior research has approached the bandwidth guarantee specification by using abstractions that resemble physical network topologies. We present a contrasting approach of deriving a network abstraction based on application communication structure, called Tenant Application Graph or TAG. CloudMirror also incorporates a new workload placement algorithm that efficiently meets bandwidth requirements specified by TAGs while factoring in high availability considerations. Extensive simulations using real application traces and datacenter topologies show that CloudMirror can handle 40% more bandwidth demand than the state of the art (e.g., the Oktopus system), while improving high availability from 20% to 70%.
引用
收藏
页码:467 / 478
页数:12
相关论文
共 34 条
[1]  
Adler Brian., 2010, LOAD BALANCING CLOUD
[2]  
[Anonymous], 2014, USENIX HOTCLOUD
[3]  
Asanovic K, USENIX FAST 2014 KEY
[4]  
Ballani H., USENIX NSDI 2013
[5]   Towards Predictable Datacenter Networks [J].
Ballani, Hitesh ;
Costa, Paolo ;
Karagiannis, Thomas ;
Rowstron, Ant .
ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2011, 41 (04) :242-253
[6]  
Benson T., 2011, ACM SOCC
[7]   Guaranteeing High Availability Goals for Virtual Machine Placement [J].
Bin, Eyal ;
Biran, Ofer ;
Boni, Odellia ;
Hadad, Erez ;
Kolodner, Eliot K. ;
Moatti, Yosef ;
Lorenz, Dean H. .
31ST INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS 2011), 2011, :700-709
[8]   Fast unfolding of communities in large networks [J].
Blondel, Vincent D. ;
Guillaume, Jean-Loup ;
Lambiotte, Renaud ;
Lefebvre, Etienne .
JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT, 2008,
[9]   Surviving Failures in Bandwidth-Constrained Datacenters [J].
Bodik, Peter ;
Menache, Ishai ;
Chowdhury, Mosharaf ;
Mani, Pradeepkumar ;
Maltz, David A. ;
Stoica, Ion .
ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2012, 42 (04) :431-442
[10]   The Tail at Scale [J].
Dean, Jeffrey ;
Barroso, Luiz Andre .
COMMUNICATIONS OF THE ACM, 2013, 56 (02) :74-80