OverviewThe MCAS Project ( Metrics Correlation and Analysis service ) provides integral solution for system operators or VO users to uniformly access, transform and represent disjoint metrics data generated by distributed middle ware or user services. The infrastructure assists indexing , navigation of existing metrics as well as it supplies tools and service to define and store quantifiable on-demand document data. The project basic interfaces will be developed to satisfy criteria of low cost integrability into existing applications. Simply put, project offers user a simple Web based toolkit for organizing and plotting any kind of system state data. The project is at the stage where prototype software has been developed to showcase project main features to the current stakeholders. Several prototype portal pages have been developed and deployed: Note that some of the graphs may not display for users outside Fermi domain.The experiments which rely on meta computing infsatriucture are invited to provide stakeholder feedback to the project. In particular , MCAS continiues to work with DZero, CDF, CMS group of experiments and FermiGrid. The project started in 2008 to capitalize on experience of running experiments in areas of troubleshooting and monitoring of independently owned and maintained computing infrastructures. The main focus of the project is thus not to replace existing monitoring solutions but rather build on top integrating umbrella service which would bridge and fill gaps in functionalities of metric collection and display products used in the field. |
Architecture Diagram
|
WorkflowThe system workflow is organized between four players - user, content management system, data integration layer, data source(s). User initiates connection to content Management system. Content management system builds composition of independent interface elements using
Last modified by Andrii Baranovski on 10/09/08 |