| Issue |
EPJ Web Conf.
Volume 214, 2019
23rd International Conference on Computing in High Energy and Nuclear Physics (CHEP 2018)
|
|
|---|---|---|
| Article Number | 08024 | |
| Number of page(s) | 8 | |
| Section | T8 - Networks & facilities | |
| DOI | https://doi.org/10.1051/epjconf/201921408024 | |
| Published online | 17 September 2019 | |
https://doi.org/10.1051/epjconf/201921408024
Trident : An Automated System Tool for Collecting and Analyzing Performance Counters
CERN,
1211
Geneva,
Switzerland
* e-mail: servesh.muralidharan@cern.ch
** e-mail: david.smith@cern.ch
Published online: 17 September 2019
Trident, a qualitative analysis tool that can look at various low level metrics with respect to the Core, Memory and I/O to highlight performance bottlenecks during the execution of an application. Trident uses a three pronged approach in analysing a node's utilisation of hardware resources and to help a non system expert understand the stress on different parts of the system by a given job. Currently metrics such as memory bandwidth, core utilization, active processor cyles, etc., are being collected. Interpretation of the data in raw form is often non intuitive. Therefore, the tool converts these data into derived metrics that are then represented as a system wide extended Top-Down analysis that helps developers and site managers likewise understand the application behavior without the need for in-depth expertise of architecture details.
© The Authors, published by EDP Sciences, 2019
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.