Fermilab Computing Division

CS Document 3926-v4

CHEP 2010 Abstract - Evolution and deployments of MCAS, a Metrics Correlation and Analysis Service

Document #:
Document type:
Submitted by:
Andrew Baranovski
Updated by:
Andrew Baranovski
Document Created:
19 May 2010, 14:01
Contents Revised:
29 Nov 2010, 09:59
Metadata Revised:
29 Nov 2010, 09:59
Viewable by:
  • Public document
Modifiable by:

Quick Links:
Latest Version

Other Versions:
19 Nov 2010, 16:56
15 Nov 2010, 16:14
19 May 2010, 14:01
The Metrics Correlation and Analysis Service (MCAS) is a web-based REST service that provides an integral solution to warehousing, querying, transforming, and displaying of disjoint metrics. These metrics can be generated by common distributed middle-ware or user services. They can also be manually entered. The solution implements a well defined model for organizing information processing and display. The core of the software infrastructure is the MuleESB data transformation engine, which implements a graph-based model for
connecting data analysis, summarization, and reduction rules. The model is completed by a python API which maps the results of the data transformation workflow into a dashboard presentation, using the layout of a few well defined UI window elements.

The paper describes the MCAS service, the options for its deployment, and the performance characteristics tested at the deployment of the pilot project. Finally, we present several applications that are using
the MCAS infrastructure. These are a time series visualization toolkit for the MINOS experiment at Fermilab, a status analysis and monitoring workflow for the dCache storage system, and a portal solution for
metrics integration for the CMS experiment at CERN.

Files in Document:
Other Files:
Associated with Events:
CHEP 2010 held from 18 Oct 2010 to 22 Oct 2010 in Taipei, Taiwan
DocDB Home ]  [ Search ] [ Authors ] [ Events ] [ Topics ]

DocDB Version 8.8.9, contact Document Database Administrators