Data handling in the ALICE O$^2$ event processing

The ALICE experiment at the Large Hadron Collider (LHC) at CERN is planned to be operated in a continuous data-taking mode in Run 3. This will allow to inspect data from all Pb-Pb collisions at a rate of 50 kHz, giving access to rare physics signals embedded in a large background. Based on experience with real-time reconstruction of particle trajectories and event properties in the ALICE High Level Trigger, the ALICE O$^2$ facility is currently designed and developed to support processing of a continuous, triggerless stream of data segmented into entities referred to as timeframes. Both raw data input into the ALICE O$^2$ system and the actual processing of aggregated timeframes are distributed among multiple processes on a many-node cluster. Process communication is based on the asynchronous message passing paradigm. This paper presents the basic concept for identification of data in the distributed system together with prototype implementations and performance measurements