This page is devoted to keep track of the activity about comparative tests over all SRMv2 Instances. The aim of this activity is to have a preliminary overview of the behavior of all main Storage technologies available on WLCG and equipped with SRM v2 under critical conditions (CASTOR, d-Cache,StoRM and DPM). The persons mainly involved on this activity are:
  • Luca Dell'Agnello (CNAF site manager, CASTOR support)
  • Vincenzo Vagnoni (CASTOR and StoRM support)
  • Angelo Carbone (CNAF site manager, CASTOR support)
  • Luca Magnoni and Riccardo Zappi (StoRM developers)
  • Giacinto Donvito and Vincenzo Spinoso (Bari site managers, dCache support)
  • Elisa Lanciotti and Roberto Santinelli(main testers from LHCb at CERN)

Partecipant storage instances are summarized on this table (that all responsibles have to keep-up-to-dated):

SITE Flavour Harware and Software desription
CNAF Castor
<missing description for CNAF CASTOR>,
under first round of test
2 Frontend ( Opteron 2GHz - 4GB ) - 1 Backend ( Dual Xeon 2.4 GHz - 2 GB),
under first round of tests
Admin Node: PIII con 1GB di RAM; DB server: Dual Xeon PIV 3.0GHz con 2 GB di RAM disco EIDE 100GB; SRM servers: (da 1 a 2/3 macchine) Dual Opteron Dual Core con 4 GB di RAM e disco SATA da 80GB; dCache 1.8.0-x with Chimera NameServer.
under first round of test
to be requested to the responsible

Initially the tests going to be performed entirely consist on the Elisa's TestSuite used previously to study StoRM storage. The plans for this activity consist to extend the scope of these stress tests for more deepy studies on the perfomaces of various storages solutions, their optimizations and their readiness for substain a HEP production like use case. The tests planned differ by any other scheduled SRM2 testing activity. They differ from the GSSD working group activity because they are more oriented to a specific VO use case : we aim to study the feasibility of the LHCb analysis and reprocessing activity on the T0D1 storage system once put under extreme conditions and by extrapolating from the Computing Model the usage planned and reproducing the conditions with custom tools.
These tests also differ by the internal LHCb SRM2 testing activity because the latter is mainly oriented to verifying the integration of DIRAC with the underlying production infrastructure and to test (with official production tools) the readiness of SRM2 endpoints. We rather study the response of the system and we aim to provide useful recomendations and feedbacks to site admins and developers to allow LHCb to run its activities on its Storage endpoints as soon as the grid will speak the SRMv2 slang too.

Minutes available of previous meetings are available at: Minutes

Edit | Attach | Watch | Print version | History: r13 < r12 < r11 < r10 < r9 | Backlinks | Raw View | Raw edit | More topic actions...
Topic revision: r10 - 2007-09-25 - RobertoSantinel
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    LCG All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2021 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