Difference: DataManagementWorkPlan (15 vs. 16)

Revision 162007-06-11 - SophieLemaitre

Line: 1 to 1
 
META TOPICPARENT name="DataManagementTop"

LFC and DPM Work Plan

Line: 15 to 15
 
3 integrate to build system   Akos   DONE
4 build RPMs on SLC4/ia64   Akos   IN PROGRESS
5 Implement automatic reconnection to database when connection lost   Jean-Philippe   DONE
Changed:
<
<
6 gSOAP 2.6.2 -> 2.7.6        
7 bulk delete function, given a list of GUIDs (ATLAS) End of July Sophie High not started
>
>
6 gSOAP 2.7.6 -> 2.7.9   Akos (to provide recipe)    
7 bulk delete function, given a list of GUIDs (ATLAS) End of June Sophie Critical not started
 
8 limited wildcard support on LFNs for query/delete catalogue operations (ATLAS) End of July Sophie High not started
Changed:
<
<
9 query/delete the catalog content using the combination of list of GUIDs and host field (ATLAS) End of August Sophie Medium not started
>
>
9 query/delete the catalog content using the combination of list of GUIDs and host field (ATLAS) End of July Sophie Medium not started
 
10 possibility to modify the "logically deleted" flag (ATLAS) End of July Sophie High not started
Added:
>
>
11 Python interface for items 7, 8, 9 and 10 End of June/July Robert High not started
12 special thread for special operations (like bulk update of paths oldStorageEndpoint to newStorageEndpoint instead of script run by admin) after items 7, 8, 9, 10, 11 Sophie Medium not started
 

LFC

Line: 31 to 33
 
4 LFC fail over, load balancing       won't do
5 selection of LFC according to the operation (read versus rw)       won't do
6 extension of lcg_util to replace lfc client (lcg_mkdir...)        
Changed:
<
<
7 implement "lfc-ls --replica"        
>
>
7 implement "lfc-ls --replica"     Low  
 
8 keep case for GUIDs   Jean-Philippe High (LHCb) DONE
11 script to remove the port number in all LFC entries End of July Sophie Medium not started
Line: 39 to 41
 

DPM

Item What ? When ? Who ? Priority Status Needs
Changed:
<
<
1 DPM with disk servers on several network domains (distributed Tier2)   Jean-Philippe   DONE  
>
>
1 DPM with disk servers on several network domains (distributed Tier2)   Jean-Philippe   DONE  
 
2 maximum pin time (possibly with different values depending on VOMS role)          
3 pool hierarchy (+ default pool)          
4 relative path (VO HOME, not chdir)          
Changed:
<
<
5 drain          
>
>
5 drain   Jean-Philippe   DONE  
 
6 (automatic) replication inside a pool          
7 global space reservation with max lifetime (possibly with different values depending on VOMS role)          
8 limit number of streams per disk server (may be useful for some applications like bulk replication, so pool dependent)          
Line: 56 to 58
 
15 test of DPM with multiple disk servers 13/08/2005 Sophie   DONE (tests through SC3) - OK  
16 test of compatibility between dCache client utility srmcp and DPM          
17 implementation of a migrator/recaller to either recall/migrate files automatically between Tier1 and Tier2 or interface to a tape/DVD backend          
Changed:
<
<
18 dpm-gsiftp with last Globus version   Yaodong   NEEDS TESTING Item 11
19 Perl/Python interface   Jean-Philippe + Sophie   DONE - NEEDS IMPROVEMENTS !  
20 DPM dynamic info script   Graeme     Item 19
>
>
18 dpm-gsiftp with last Globus version June 2007 Yaodong + Jean-Philippe   DONE Item 11
19 Perl/Python interface   Jean-Philippe + Sophie   DONE - NEEDS IMPROVEMENTS  
20 DPM dynamic info script   Graeme   DONE Item 19
 
21 Classic SE -> DPM migration script   Yvan   High TEST THE LAST VERSION
22 Support for Large File Systems in 1.3.8 release Jean-Philippe DONE High (for UK)  
23 Checking file integrity after copy (file size, checksum)       Critical  
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback