Home

CMSSpark provides a python based framework to perform analytics tasks with CMS data-streams on CERN analytix cluster. To gain access to the cluster you need to login to analytix.cern.ch node (please obtain an account before that). In following sections you'll find all information you'll need to setup CMSSpark job.

• Introduction
• Architecture
• Data streams
• Setup and usage of aggregation task
• Data collection from multiple streams script
• Data collection and aggregation script
• Cron job and keytab
• Plot generation script
• Statistics