Real-time conditions data distribution for the Online data processing of the ALICE experiment

Description

The new ALICE [1] synchronous data reconstruction facility for Run 3 [2] needs a real-time conditions and calibration data distribution mechanism. New calibration objects are produced at up to 50Hz and have to be propagated to about 2000 servers. For efficient data distribution in this environment a network multicast delivery mechanism has to be used. There will be two sides to be implemented for this project: a library to send the newly produced objects and a caching service to run on each of the 2000 servers to receive and keep in memory the objects, making them available to the localhost running processes via a REST API (already defined for the central repository implementation [3, 4]).

Task ideas

Expected results

Two multicast sending libraries and a receiving service that also implements a REST API for serving the objects from memory.

Desirable Skills

Mentors

  1. ALICE experiment at CERN
  2. ALICE Run 3 upgrade Technical Design Report
  3. ALICE CCDB status
  4. ALICE CCDB Git repository

Corresponding Project

Participating Organizations