| Issue 43: | finding the tier for allocation | |
| 1 person starred this issue and may be notified of changes. | Back to list |
There are three parts of this algorithm -> checking the user id of the new file -> checking the group id of the new file -> checking the extention of the new file Checking the used id -> current macro contains the information about the user id -> a simple linear comparison with allocation policy and the user id field of the current macro is done -> which results in the tier on which the file should be placed -> the ohsm_home_tid filed is then initialized to the resulting tier -> the allocation algorithm then refers the sam table for the block group range for that tier and allocates blocks accordingly Checking the used id -> current macro contains the information about the group id -> a simple linear comparison with allocation policy and the group id field of the current macro is done -> results in the tier on which the file should be placed -> the ohsm_home_tid filed is then initialized to the resulting tier -> the allocation algorithm then refers the sam table for the block group range for that tier and allocates blocks accordingly Checking the extention o fthe new file -> the ext2 module gets the dentry of the file during inode creation -> the dentry has the name of the file -> from the name of the file the extention is extracted and compared to the allocation policy -> it results in the tier on which the file should be placed -> the ohsm_home_tid filed is then initialized to the resulting tier -> the allocation algorithm then refers the sam table for the block group range for that tier and allocates blocks accordingly
Jan 28, 2009
Project Member
#1
sandeepksinha
Jan 30, 2009
The code is complete and working fine
Status:
Done
|