Fermilab Computing Division

CS Document 3866-v1

ROOT I/O: The Fast and Furious.

Document #:
Document type:
Submitted by:
Philippe Canal
Updated by:
Philippe Canal
Document Created:
30 Apr 2010, 14:39
Contents Revised:
30 Apr 2010, 14:39
Metadata Revised:
22 Sep 2011, 18:50
Viewable by:
  • Public document
Modifiable by:

Quick Links:
Latest Version

The increase of data size and the increase in the geographical distribution of analysis intensifies the demand on the I/O subsystem. Over the last year, we greatly improved the I/O throughput, in some case by several factors, when reading ROOT files. These improvements leverage existing ROOT techniques for prefetching (the "TTreeCache"), the automatic flushing of data buffers at regular intervals or streaming objects member-wise. These advances reduces the number of transactions with the local disk or the network. We worked in closely collaboration with the LHC experiments to optimize the I/O access to their use cases and to help adapt their framework to take full advantage of these advances. This presentation will describe in details these improvements and how to benefit from them.
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