juvenile Posted June 9, 2013 Report Posted June 9, 2013 impdp chala slow ga nadustundi mavayyalu .. status custe ..Worker Parallelism: 1 ane chuistundi eventhough call chesetappudu parallel=16 icha .. 80 million rows .. chaala time taking .. oracle 11.2.0.3 .. AIX .. impdp system/'ABCDEFGH' tables='AGGI' directory=impdir parallel=16 dumpfile=expdp%U.dmp table_exists_action=replace logfile=impdptblog.log
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 Available slaves lev emo os la?? At times import aiyye time lo many parallels ichina konta time of execution lo single slave ee work cheste migata anta wait status lo untai... Db level lo resource oriented jobs emanna nadustunnaya??
juvenile Posted June 9, 2013 Author Report Posted June 9, 2013 Nothing running in DB mayya .. Process Name: DW00 State: EXECUTING Object Schema: PRD Object Name: DBTABLOG Object Type: TABLE_EXPORT/TABLE/TABLE_DATA Completed Objects: 1 Completed Rows: 8,842,421 Completed Bytes: 96,162,811,144 Percent Done: 10 Worker Parallelism: 1 idi current status
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 hmm.. good to know: [font=Garamond][size=2] [size=4]The first worker begins to load all the metadata – the tablespaces, schemas, etc., until all the tables are created. [/size][/size][/font][font=Garamond][size=2] [size=4]• Once the tables are created, the first worker starts loading data instead of metadata and the rest of the workers start loading data too. [/size][/size][/font][font=Garamond][size=2] [size=4]• Once the table data is loaded, the first worker returns to loading metadata again. The rest of the workers are idle until the first worker loads all the metadata up to package bodies. [/size][/size][/font][font=Garamond][size=2] [size=4]• Multiple workers load package bodies in parallel. [/size][/size][/font][font=Garamond][size=2] [size=4]• One worker loads metadata up to and including secondary tables. [/size][/size][/font][font=Garamond][size=2] [size=4]• Multiple workers load secondary table data. [/size][/size][/font][font=Garamond][size=2] [size=4]• One worker loads the remaining metadata. [/size][/size][/font][font=Garamond][size=2] [size=4]Note: One worker creates all the indexes but uses PX processes up to the PARALLEL value so indexes get created faster. [/size][/size][/font][font=Garamond][size=2] [size=4]Thus, an import job can be started with a PARALLEL = 10, and the user will only see one worker being utilized at certain points during job execution. No other workers or Parallel Execution Processes will be working until all the tables are created. When the tables are created, a burst of workers and possibly PX processes will execute in parallel until the data is loaded, then the worker processes will become idle.[/size][/size][/font] [font=Garamond][size=2] [size=4]adi bhayya matter[/size][/size][/font]
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 [font=Garamond][size=2] [size=5]Data Pump uses the PARALLEL parameter as the maximum allowed parallelism for the job, with the maximum number of active workers and Parallel Execution Processes at any given time. If a user is monitoring the workers, there will be times when workers will be busy and other times when some workers may be idle. This is expected and means that Data Pump is behaving properly.[/size][/size][/font]
juvenile Posted June 9, 2013 Author Report Posted June 9, 2013 Ento bhayya !! Strange behaviour .. Import> status Job: SYS_IMPORT_TABLE_01 Operation: IMPORT Mode: TABLE State: EXECUTING Bytes Processed: 0 Current Parallelism: 16 Job Error Count: 0 Worker 1 Status: Process Name: DW00 State: EXECUTING Object Schema: PRD Object Name: DBTABLOG Object Type: TABLE_EXPORT/TABLE/TABLE_DATA Completed Objects: 1 Completed Rows: 13,734,868 Completed Bytes: 96,162,811,144 Percent Done: 16 Worker Parallelism: 1 Current parallelism - 16 chuistundi .. still i see only one worker .. and one worker parallelism !!
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 chadivava bhayya paina cheppindi... at times you see only one worker works, but eventually, when the table structures are built by the master slave, parallel slaves trigger in to load.. also, same with indexes too..
juvenile Posted June 9, 2013 Author Report Posted June 9, 2013 [quote name='CBZ' timestamp='1370752077' post='1303838313'] [font=Garamond][size=2][size=5]Data Pump uses the PARALLEL parameter as the maximum allowed parallelism for the job, with the maximum number of active workers and Parallel Execution Processes at any given time. If a user is monitoring the workers, there will be times when workers will be busy and other times when some workers may be idle. This is expected and means that Data Pump is behaving properly.[/size][/size][/font] [/quote] Thanks bhayya !! but i can see data getting imported ... aa stage lo parallelism undali ..
juvenile Posted June 9, 2013 Author Report Posted June 9, 2013 [quote name='CBZ' timestamp='1370752213' post='1303838318'] chadivava bhayya paina cheppindi... at times you see only one worker works, but eventually, when the table structures are built by the master slave, parallel slaves trigger in to load.. also, same with indexes too.. [/quote] yeah bhayya .. currently data import phase lo undi .. aa time lo parallel threads undaali kada ... thikinggg
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 ee table lo aite load chestundo, aa table data type enti?? if it is long raw or blob's, then I read, parallelism won't work on these data types...
juvenile Posted June 9, 2013 Author Report Posted June 9, 2013 [quote name='CBZ' timestamp='1370752443' post='1303838326'] ee table lo aite load chestundo, aa table data type enti?? if it is long raw or blob's, then I read, parallelism won't work on these data types... [/quote] LONG RAW undi bhayyaaaaa .. Edaina Bug aa vatilo parallelism work avvakapovadam ?
peru_edaite_enti Posted June 9, 2013 Report Posted June 9, 2013 [quote name='AggiRamudu' timestamp='1370752512' post='1303838329'] LONG RAW undi bhayyaaaaa .. Edaina Bug aa vatilo parallelism work avvakapovadam ? [/quote] perfect.. this is what I was expecting ba.... yes long raw aite impdp handles different.. http://pingshx.itpub.net/post/39434/484690 ee link lo chudu.. they had the same issue... data type is LONG RAW
Recommended Posts