24–28 May 2021
Via telematica
Europe/Rome timezone

IDDLS - A prototype for the Italian Distributed Data Lake for Science

26 May 2021, 11:15
30m
Via telematica

Via telematica

Infrastrutture ICT e calcolo distribuito Infrastrutture ICT e calcolo distribuito

Speaker

Daniele Cesini (CNAF)

Description

The IDDLS project is configured as an INFN-GARR collaboration aimed at investigating and testing high bandwidth programmable networking technologies for the direct interconnection of INFN datacenters. The project is jointly funded by the INFN-CSN5 and GARR. The final goal is to implement a prototype of a single entry-point federated storage infrastructure, or Data Lake, to serve the High Energy Physics experiments and other collaborations accessing the INFN computing infrastructures. The CNAF Tier1 and some of the INFN Tier2s are involved as resource providers. The middleware/software layer to create the Data Lake is based on scalable storage federations technologies such as RUCIO, DYNAEED, XROOTD. Once the prototype will be created, real-life use cases for performance testing will be provided by CMS, ATLAS, BELLE-II and possibly some LNGS experiments in a second phase. In this talk we present the status of the prototype and the technologies that have been selected for its implementation.

Primary author

Daniele Cesini (CNAF)

Presentation materials