Monitoring High Performance Computing Systems for the End User
Monitoring High Performance Computing Systems for the End User
复制标题
为最终用户监控高性能计算系统
DOI：
10.1109/cluster.2015.124
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
M. Mason
中科院分区：
文献类型：
--
作者：
C. Moore;P. Khalsa;Todd Alan Yilk;M. Mason
监控高性能计算集群目前旨在为系统管理员提供他们需要的信息，以便他们就集群中使用的资源做出明智的决策。然而，这种强调忽略了最终用户，即那些将集群资源用于项目和方案的用户，因为他们没有被给予他们的工作流程如何影响集群的信息。通过以终端用户可以轻松解释和使用的格式提供监控数据子集，他们可以帮助更好地利用提供给他们的计算资源。
Monitoring High Performance Computing clusters is currently geared towards providing system administrators the information they need to make informed decisions on the resources used in the cluster. However, this emphasis leaves out the End User, those who utilize the cluster resources towards projects and programs, as they are not given the information of how their workflow is impacting the cluster. By providing a subset of monitoring data in a format End Users can easily interpret and utilize, they can help make better use of the computing resources provided to them.