As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
The AutoTune energy plugin here presented selects and configures the frequency of the compute nodes for program regions based on actually executed patterns, while having only a minimal or acceptable performance impact. This plugin allows to determine the dependence of application performance on processors frequency and the energy consumption. The plugin is based on a self-developed library which allows to know the distribution of the energy consumption among the different components of the compute nodes of a Sandy-Bridge supercomputer system while using the dynamic frequency scaling system feature to give some control in throttling the system to consume less power when not operating at full capacity. We also present an energy model to be used with that plugin, which will enable the detailed analysis of a parallel code as a whole, i.e. per code region and task and, which will serve information for a possible heuristic function.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.