1. Please help support our sponsors by considering their products and services.
    Our sponsors enable us to maintain high-speed Internet connection and fast webservers.
    They support this free information and knowledge exchange forum service at no cost to you.

    Please welcome our latest sponsor Tectrade . We can show our appreciation by learning more about Tectrade Solutions

HSM Client side Activity

Discussion in 'Hierarchical Storage Management' started by mguery, Apr 7, 2013.

  1. mguery

    mguery ADSM.ORG Member

    Joined:
    Jan 8, 2003
    Messages:
    16
    Likes Received:
    0
    Occupation:
    AIX & TSM Consultant
    Location:
    Ile de La Réunion
    Hi all,
    I am looking for reports on the activity HSM (client side), namely:
    How many files are migrated daily, the size of these files, and the time for migration.
    How many files are daily reminded each day, the size of these files, and the time for the reminder of these files.

    The customer is SunOS 5.10 with TSM / HSM v6.1.5
    The TSM server is AIX v7.1 and TSM 6.2.3

    Does someone has already worked on this subject?
     
  2.  
  3. mguery

    mguery ADSM.ORG Member

    Joined:
    Jan 8, 2003
    Messages:
    16
    Likes Received:
    0
    Occupation:
    AIX & TSM Consultant
    Location:
    Ile de La Réunion
    Hi,
    Is there a tool for this activity? (Server or client side) ?
     
  4. mguery

    mguery ADSM.ORG Member

    Joined:
    Jan 8, 2003
    Messages:
    16
    Likes Received:
    0
    Occupation:
    AIX & TSM Consultant
    Location:
    Ile de La Réunion
    Some statistics from TSM/HSM server...

    Hi, every TSM/HSM administrators....
    Here two sql for statistics about HSM activity :

    The first is about for both MIGRATE and RECALLs activities, where the file's size is not null, for one day.
    The second is just for checking...

    1) Activity HSM for yesterday,
    I suppose that when the file's size is null ( BYTES=0), then the file is premigrated ?

    Here is the sql :

    select cast(start_time as date) as date, entity, activity, sum(affected) as Affected, -
    sum(bytes)/1024/1024 as Vol_Mb, min(mediaw) as Min_WAIT , max(mediaw) as Max_WAIT, -
    avg(mediaw)as Avg_WAIT -
    from summary -
    where (activity='HSM-MIGRATE' or activity='RECALL') -
    and (bytes>0) and (cast(Start_time as date)=(cast(current_timestamp as date) - 1 days)) -
    group by cast(start_time as date), entity, activity

    The result is similar to :
    DATE: 2013-04-10 <<<< = (cast(current_timestamp as date) - 1 days))
    ENTITY: NODE_A
    ACTIVITY: HSM-MIGRATE
    AFFECTED: 119 <<<< Total of migrated files for Date and Entity (node name)
    VOL_MB: 2730 <<<< Size of Total of migrated files for Date in Mb
    MIN_WAIT: 0 <<<< Mini Wait time (in seconds) ( if MIN_WAIT=0 then migrate to DISK)
    MAX_WAIT: 0 <<<< Max Wait time (in seconds) ( if MAX_WAIT>0 then migrate to tape : very bad situation !!! )
    AVG_WAIT: 0 <<<< Average Wait time for HSM-MIGRATE (in seconds)

    DATE: 2013-04-10
    ENTITY: NODE_A
    ACTIVITY: RECALL
    AFFECTED: 538 <<<< Total of recalled files for Date and Entity (node name)
    VOL_MB: 5610 <<<< Size of Total of recalled files for Date in Mb
    MIN_WAIT: 0 <<<< if Min_Wait=0 : at least 1 file was on DISK (HSMpool)
    MAX_WAIT: 211 <<<< Maximum Wait time ( read from tape) (in seconds)
    AVG_WAIT: 13 <<<< Average Wait time for HSM-RECALL (in seconds)

    2) To verify "ACTIVITY: HSM-MIGRATE AFFECTED: 119" :

    select count(file_name) as HSM_Mig_1_days -
    from SPACEMGFILES -
    where (cast(insert_date as date)=(cast(current_timestamp as date) - 1 days)) -
    and node_name="NODE_A"

    The result is similar to :
    HSM_MIG_1_DAYS
    ---------------
    119


    If you need further explanation, let me know.
    I suppose these two queries are consistent. If someone could confirm, that would be nice.
    I still hope to find a solution from the HSM client.
     
    Last edited: Apr 14, 2013
: HSM, reporting, tsm 6

Share This Page