评估云计算规模自动调整应用
|
摘要
根据工作量和数据存储量,云计算能够动态扩充系统资源。除了用于Web服务器的传统规模扩充机制之外,用于数据库服务器的基于KVS的规模扩充机制正在引起关注。然而,关于Web服务器和数据库服务器自动规模扩充方法,以及在工作量和数据存量较少的数据库服务器上进行规模缩减方法的研究并不是很多。该文介绍了他们的系统及其在云计算平台上进行Web服务器和数据库服务器规模扩充与缩减的测试结果。测试结果表明云计算平台上的工作量影响实际应用。他们还成功地完成了数据库服务器在线规模缩减实验。
Cloud computing dynamically enables to scale out of system resources in terms of workloads and data volume.In addition to the conventional scaling-out mechanisms of Web servers,KVS based scaling-out mechanisms of database servers are drawing increasing attention from the circle of research.However,there are not many researches about the approaches to the automatically scaling out of Web servers and database servers and the approaches to the scale reduction of database servers with less workloads and smaller data volume.In this paper,the two approaches are presented and the test results of the scaling out and the reduction of Web servers and database servers on a cloud-computing platform are also introduced.The test results of the two approaches show that the workloads on a cloud-computing platform affect their applications.The testing of scale reduction of live database servers is successfully completed.
Cloud computing dynamically enables to scale out of system resources in terms of workloads and data volume.In addition to the conventional scaling-out mechanisms of Web servers,KVS based scaling-out mechanisms of database servers are drawing increasing attention from the circle of research.However,there are not many researches about the approaches to the automatically scaling out of Web servers and database servers and the approaches to the scale reduction of database servers with less workloads and smaller data volume.In this paper,the two approaches are presented and the test results of the scaling out and the reduction of Web servers and database servers on a cloud-computing platform are also introduced.The test results of the two approaches show that the workloads on a cloud-computing platform affect their applications.The testing of scale reduction of live database servers is successfully completed.
引文
[1]G.Ito.从日本东部大地震中回顾灾难管理.2011.http://www.ascii.jp/elem/000/000/619/619868/.
[2]S.Malkowski,M.Hedwig,and C.Pu.“N-级系统实验评估:多瓶颈观察与分析,”工作量特性[A].IISWC.IEEE国际论文集[C].2009:118-127.
[3]S.Pullara,E.M.Halpern,P.Peddada,A.Messinger,and D.B.Jacobs.用于会话复制与故障倒换的方法与装置:美国,7409420[P].2008.
[4]OpenStack.用于构建私有和公共云端的开源软件[EB/OL].http://www.openstack.org/.
[5]H.A.Proxy.可靠的高性能TCP/HTTP负载均衡器[EB/OL].http://www.haproxy.1wt.eu/.
[6]PlayFamework.Play Framework使得利用Java和Scala构建Web应用变得更简单[EB/OL].http://www.playframework.org/.
[7]G.DeCandia,D.Hastorun,M.Jampani,G.Kakulapati,A.Lakshman,A.Pilchin,S.Sivasub-ramanian,P.Vosshall,and W.Vogels.Dynamo.亚马逊的高可用性关键价值储存[A].第21届ACJMSIGOPS操作系统原理会议论文集[C].2007:205-220.
[8]R.Geambasu,A.A.Levy,T.Kohno,A.Krishnamurthy,and H.M.Levy.Comet:一个活动的分布式关键价值存储[A].第9届USENIX操作系统设计与实现会议论文[C].2010:1-13.
[2]S.Malkowski,M.Hedwig,and C.Pu.“N-级系统实验评估:多瓶颈观察与分析,”工作量特性[A].IISWC.IEEE国际论文集[C].2009:118-127.
[3]S.Pullara,E.M.Halpern,P.Peddada,A.Messinger,and D.B.Jacobs.用于会话复制与故障倒换的方法与装置:美国,7409420[P].2008.
[4]OpenStack.用于构建私有和公共云端的开源软件[EB/OL].http://www.openstack.org/.
[5]H.A.Proxy.可靠的高性能TCP/HTTP负载均衡器[EB/OL].http://www.haproxy.1wt.eu/.
[6]PlayFamework.Play Framework使得利用Java和Scala构建Web应用变得更简单[EB/OL].http://www.playframework.org/.
[7]G.DeCandia,D.Hastorun,M.Jampani,G.Kakulapati,A.Lakshman,A.Pilchin,S.Sivasub-ramanian,P.Vosshall,and W.Vogels.Dynamo.亚马逊的高可用性关键价值储存[A].第21届ACJMSIGOPS操作系统原理会议论文集[C].2007:205-220.
[8]R.Geambasu,A.A.Levy,T.Kohno,A.Krishnamurthy,and H.M.Levy.Comet:一个活动的分布式关键价值存储[A].第9届USENIX操作系统设计与实现会议论文[C].2010:1-13.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心