Tracing and monitoring large online server clusters for performance analysis, security and maintenance

跟踪和监控大型在线服务器集群以进行性能分析、安全和维护

• 批准号: 411938-2010
• 负责人: Dagenais, Michel
• 金额: $ 1.34万
• 依托单位: École Polytechnique de Montréal
• 依托单位国家: 加拿大
• 项目类别: Strategic Workshops Program
• 财政年份: 2010
• 资助国家: 加拿大
• 起止时间: 2010-01-01 至 2011-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=473172
• 关键词: Tracing; monitoring; large; online; server

In recent years, the Internet and cellular networks have become the center of all interactions from text, voice and video applications to shopping, entertainment, information gathering and government services. This is the case both at home and at work within organizations. The underlying infrastructure has evolved accordingly at an accelerated pace. The newer electronic service centers are built out as clusters of numerous high performance relatively low cost multi-core computers connected through reduntant high speed networks. Furthermore, these clusters are efficiently shared among several different applications which need peak resources at different times. In such a context, where a service is offered "in the cloud" by virtual machines running on different computers at different times, it becomes extremely difficult to monitor and optimize the performance and security of each service.

近年来，互联网和蜂窝网络已成为从文本、语音和视频应用到购物、娱乐、信息收集和政府服务的所有交互的中心。在组织内部，无论是在家里还是在工作中都是如此。相应地，基础设施也在加速发展。较新的电子服务中心是由大量高性能、相对低成本的多核计算机通过新的高速网络连接而成的集群。此外，这些集群在不同时间需要峰值资源的几个不同应用程序之间有效地共享。在这样的环境中，服务是由在不同时间在不同计算机上运行的虚拟机在云中提供的，监控和优化每项服务的性能和安全性变得极其困难。