CMSSW I/O with the 2-cache schema

This page discusses CMSSW I/O under the following conditions:

Below is a few graphs showing the I/O activity using the BlockTrace utility:

  • I/O activity on analyzed file over the whole run of the application:
    There are four main bands of activity - first, when ROOT opens the file. Second, when ROOT starts opening and processing the TTree itself (figuring out what it needs to deserialize, event metadata). Third is the initial fill of the "raw tree", which caches all branches. After this, there's a long pause in I/O when the conditions data is loaded. Finally, there is the run through the file itself. The first 20 events are read using the raw tree. Then, the TTreeCache buffer is then filled for the specialized tree. Afterward, you see (almost) no more stalls for the rest of the file. The partial stalls later on in the job are due to a branch being used which is not cached.
  • Notice that the disk-level parallelism is high. However, the job does stall in the posix_fadvise call as in the other asynchronous I/O like the other storage cache jobs. We will (in the long run) be working to avoid the disk-level blocking.
Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r2 - 2010-05-13 - BrianBockelman
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback