1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l November 2002, week 5 2 20 47_Louis P Walton Jr/ASD/CSC is out of the office.17_Louis P Walton Jr16_lwaltonj@CSC.COM31_Fri, 29 Nov 2002 01:06:00 -0500536_us-ascii I will be out of the office starting 11/28/2002 and will not return until 12/02/2002.

I will respond to your message when I return.

If you need to contact me, page me at 903-621-8345

For Raytheon DB2 problems contact James Ashmore, pager 972-229-0102

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://listserv.ylassoc.com. The owners of the list can be reached at DB2-L-REQUEST@listserv.ylassoc.com. 23 21 56_Shan Leatherman/MO/americancentury is out of the office.0_54_Shan_Leatherman/MO/americancentury@AMERICANCENTURY.COM31_Fri, 29 Nov 2002 00:40:32 -0600398_us-ascii I will be out of the office starting 11/29/2002 and will not return until 12/03/2002.

If you need additional assistance, send a message to *DB. If you need immediate assistance, page the primary DBA on call at (816) 292-5449 or contact the help desk (816) 340-4250 . If your issues/questions are regarding RPS Separation, please direct those questions to RKA at (816) 340-7429 [...] 45 61 17_Omegamon DB2 v51013_Sandeep Simon33_sandeep.s.alumootil@UKTRANSCO.COM31_Fri, 29 Nov 2002 09:40:21 +0000347_us-ascii Dear Listers,

We are trying to use Omegamon Db2 Object Analysis facility to look at TS/IX with max extends so that reorgs can be scheduled for those TX/IX immediately. However it does not seem to be showing an IX with 103 extends..is there a limitation anybody knows of (say only displays upto or less than 100 extends..etc)? [...] 107 44 16_SQLJ performance15_Foweather, Iris30_Iris.Foweather@UK.EXPERIAN.COM31_Fri, 29 Nov 2002 10:24:20 -0000328_ISO-8859-1 Can anyone help regarding SQLJ performance issues?

We are DB2 V6.1 on OS/390 V2.10 I know for java we should upgrade to V7 BUT that is not an option yet! Is anyone out there using SQLJ? How can we monitor and tune performance for SQLJ? We have an application that is performing badly on a 20 row table... [...] 152 121 21_Re: Omegamon DB2 v51018_Jose Ramon Vazquez17_jrvazquez@ATCA.ES31_Fri, 29 Nov 2002 12:33:31 +0100635_us-ascii Hello,

With V 520 we see the following with a dataset of 125 extents.

________________ ZOJA1 VTM O2 V520./C DBP3 11/29/02 12:25:56 2 > Help PF1 Back PF3 Up PF7 Down PF8 Sort PF10 Zoom PF11 > > OBJECT ANALYSIS: Enter a selection letter on the top line. > > *-DATABASE ALLOCS B-DATABASE ACTIVITY C-DISPLAY OPTIONS ============================================================================== > DATABASE ALLOCATIONS OJA1 + Database = BVLP0014 Total Spaces = 2 Total Datasets = 1 + * + Spacenam Type BP Vols DSNs Exts Exts/DSN Max Exts + -------- ---- ---- ---- ---- ---- -------- -------- + EVL005 SEGM 2 0 0 [...] 274 61 21_Re: Omegamon DB2 v51013_Sandeep Simon33_sandeep.s.alumootil@UKTRANSCO.COM31_Fri, 29 Nov 2002 11:57:58 +0000710_us-ascii Hi Jose, Thank you for that We were wondering if some of the TS/IX were being fileterd off. Could you tell us what the collections ptions were ?

Thanks and Regards, Sandeep Simon ________________________________ Sandeep Simon Alumootil ATLink Application Support Office: 01455 892042 (Internal 715 32042) Mobile: 07879430971 Email:Sandeep.S.Alumootil@Uktransco.com

"Mainframe - An obsolete device still used by thousands of obsolete companies, serving billions of obsolete customers, and making huge obsolete profits, for their obsolete shareholders. And this year's mainframes run twice as fast as last year's." (The Devil's IT Dictionary ) [...] 336 18 21_Re: Omegamon DB2 v51018_Jose Ramon Vazquez17_jrvazquez@ATCA.ES31_Fri, 29 Nov 2002 13:23:17 +0100404_us-ascii Hi Simon,

I don't undestand exactly your question. Please, what screen or configuration member do you want ?



Regards.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://listserv.ylassoc.com. The owners of the list can be reached at DB2-L-REQUEST@listserv.ylassoc.com. 355 43 8_00C9009614_Mark McCormack27_mamccormack@STATESTREET.COM31_Fri, 29 Nov 2002 09:40:13 -0500514_us-ascii Marty,

You have received advice to use LOCK TABLE instead of increasing LOCKMAX. If you can tolerate the loss of concurrency, I agree with that advice.

You say that the program currently runs without COMMITs but that you intend to have that changed (eventually). If you choose to use LOCK TABLE, and if you choose to continue to use LOCK TABLE after COMMITs are added, there are a couple of things to keep in mind. Phil Grainger mentioned these briefly. I would like to expand on them. [...] 399 76 47_Doubt reg a DB2 Load utility -- (RESUME option)10_Karthick I34_karthick.iyemperumal@UKTRANSCO.COM31_Fri, 29 Nov 2002 14:58:51 +0000596_us-ascii Hello everyone,

We have a requirement to archive one of our partition table to have only two years of data. As the Load of two years of data and rebuild of indexes takes longer time we are trying to use the following option.

All our Transactions will update only last two months of data. Other data will be only referred. So, We will create another partition table with same characteristics of existing one and load all the data which will not be changed with LOAD REPLACE OPTION well before implementation. In the implementation day we will download only the changed [...] 476 57 21_Re: Omegamon DB2 v51013_Sandeep Simon33_sandeep.s.alumootil@UKTRANSCO.COM31_Fri, 29 Nov 2002 16:59:17 +0000437_us-ascii Jose,

I guess the problem is that it displays only TS/IX that were referred today.

Therefore,is it possible to use this information to produce a daily report of all TS/IX that have (say) more than 50 extends?

Thanks and Regards, Sandeep Simon ________________________________ Sandeep Simon Alumootil Office: 01455 892042 (Internal 715 32042) Mobile: 07879430971 Email:Sandeep.S.Alumootil@Uktransco.com [...] 534 122 51_Re: Doubt reg a DB2 Load utility -- (RESUME option)12_michael bell24_mbell11@WORLDNET.ATT.NET31_Fri, 29 Nov 2002 11:31:54 -0600559_iso-8859-1 I had to check but you have to have DB2 V5.1 to get the RENAME TABLE function. You can use alias or synonym to convert from the old name to the new name.

LOAD RESUME just adds the new data to the end of the tablespace. If your primary/clustering key is date or time based, this will be close to optimal placement. Even if the clustering key isn't date/time based, there will be 2 clumps of properly clustered data. This will usually require an increase in buffers utilized and additional IO for transactions. This is not as bad as pure [...] 657 15 17_STORED PROCEDURES29_Sherri Sterling DIGEST NOMIME29_Sherri.Sterling@BPD.TREAS.GOV31_Fri, 29 Nov 2002 12:47:36 -0500426_us-ascii Can a DB2 for OS/390 version 7 Stored Procedure connect to two different database subsystems at one time or with in the same Unit Of Work?

Sherri Sterling

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://listserv.ylassoc.com. The owners of the list can be reached at DB2-L-REQUEST@listserv.ylassoc.com. 673 79 20_Re: SQLJ performance15_Shields Michael28_ShieldsMichael@JOHNDEERE.COM31_Fri, 29 Nov 2002 12:40:55 -0600673_iso-8859-1 I've started testing SQLJ on our app recently. I hit a problem with jvm versions that won't allow me to use the customized version of SQLJ. However,... to answer your question.... I just used System.out.println(endtime - starttime) where endTime is System.currentTimeMillis() after the pgm executed and starttime is System.currentTimeMillis() just before execution. I noticed that SQLJ was performing twice as slow as the original dynamic statement it was going to replace. This is because (before customization and binding the statement to package/plan) it creates and executes as a prepared statement each time the SQLJ statement is executed. You can see [...] 753 161 52_Re: Doubt reg a DB2 Load utility -- (RESUME opti on)14_Grainger, Phil20_Phil.Grainger@CA.COM31_Fri, 29 Nov 2002 19:25:20 -0000340_iso-8859-1 Just adding to Michaels final suggestion, with CA Rapid Reorg and Fast Load combined you can do a rather "sneaky" thing.

We will unload the existing data, merge it with the new data and then reload the table with the merged data. All this is handled for you by the tools and is intended just for such an eventuality. [...] 915 174 52_Re: Doubt reg a DB2 Load utility -- (RESUME opti on)10_Karthick I34_karthick.iyemperumal@UKTRANSCO.COM31_Fri, 29 Nov 2002 19:40:46 +0000251_us-ascii Hi Phil & Michael, The problem is that when we try to load the complete data it takes atleast 2.5 to 3 hrs which is a very long outage that can not be afforded, whereas when we try to load just 2 months data it takes only 40 to 50 mins. [...] 1090 30 28_DB2Estimator Path not chosen0_28_sharon.fields@ADVANCEPCS.COM31_Fri, 29 Nov 2002 13:39:04 -0700399_us-ascii Louis,

I agree -- I think that the unique may be important to the optimizer. I just had a situation today where a different accesspath was chosen upon changing the bind parms ?!?!. How much detail have you fed to your optimizer regarding the rest of the index columns and partition data details ? I could see where in 'real world' the optimizer would say INDEXA is good enough. [...] 1121 213 52_Re: Doubt reg a DB2 Load utility -- (RESUME opti on)12_michael bell24_mbell11@WORLDNET.ATT.NET31_Fri, 29 Nov 2002 14:45:29 -0600551_iso-8859-1 Only 3rd party LOAD utilities completely rebuild the index as part of the load. IBM load has always done a sort merge process that is much more efficient than random inserts, but does leave page splits behind. The result of LOAD RESUME is not an optimal index but is order of magnitude better than equivalent number of INSERT. Type 2 indexes fixed a lot of the problems with INSERT but still not as good as LOAD logic. If I was on time critical install, I would NOT take the time to reorg the indexes. Might come back the next weekend [...]