Hi experts,

I have a doubt,

in my setup we currectly have RHEL 5.5 32 bit, and DB2 9.7 32bit,

I have the backup daily wise, and the DB is connected from Storage SAN,

So now we are planning for a migration to 64bit, Now the doubt is after migration the OS and DB to 64 bit.

if i re-mount the Storage to the server, and do a restore of DB
will i be able to access the DB,

or a catalog of DB is enough bring it up.

Kindly share all your suggestion and views.

I would opt for doing an offline backup of the 32 bit system and restoring it on the 64 bit system. This would automatically migrate the database for you and you should be good to go right after the restore. Although you will probably want to change some configuration parameters to take advantage of the 64 bit system and more memory.

Andy

Thanks andy,

it worked the same way in test env... lemme try the same in my PROD.

can you tell what are the key config parms.

and i also need to know how to tune the DB for avoiding shared memory error
as we were facing this in 32bit hence migrating to 64bit

.. help me

What is the exact error you are getting? How much memory is on the 32 bit server? How much memory will you have on the 64 bit server? What is the current DBM, DB, and bufferpool config for the 32 bit system?

Andy

In my 32 bit (DB2 Workgroup) server i have 32Gb RAM, which will not be used by DB2 because of the limitation that it can use only 4Gb.

We have 2 Database in our env

this is the bufferpool configuration of DB-1(which is the critical DB)

[db2inst1@WESSPTDW01 ~]$ db2pd -db warehous -bufferpools

Database Partition 0 -- Database WAREHOUS -- Active -- Up 0 days 14:11:26 -- Date 05/12/2011 09:12:51

Bufferpools:
First Active Pool ID 4096
Max Bufferpool ID 0
Max Bufferpool ID on Disk 7
Num Bufferpools 4

Address Id Name PageSz PA-NumPgs BA-NumPgs BlkSize NumTbsp PgsToRemov CurrentSz PostAlter SuspndTSCt Automatic
0x18BEE380 4096 IBMSYSTEMBP4K 4096 16 0 0 5 0 16 16 0 False
0x19DBD730 4097 IBMSYSTEMBP8K 8192 16 0 0 1 0 16 16 0 False
0x19DBE7E0 4098 IBMSYSTEMBP16K 16384 16 0 0 0 0 16 16 0 False
0x1A1B0060 4099 IBMSYSTEMBP32K 32768 16 0 0 7 0 16 16 0 False

Bufferpool Statistics for all bufferpools (when BUFFERPOOL monitor switch is ON):

BPID DatLRds DatPRds HitRatio TmpDatLRds TmpDatPRds HitRatio IdxLRds IdxPRds HitRatio TmpIdxLRds TmpIdxPRds HitRatio
4096 209687343 104421878 50.20% 573089 19446 96.61% 2698649 139657 94.82% 0 0 00.00%
4097 0 0 00.00% 4947 24 99.51% 0 0 00.00% 0 0 00.00%
4098 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%
4099 104200431 50384855 51.65% 0 0 00.00% 1420427 52396 96.31% 0 0 00.00%

BPID DataWrts IdxWrts DirRds DirRdReqs DirRdTime DirWrts DirWrtReqs DirWrtTime
4096 44927 8734 17414986 9292 1333877 0 0 0
4097 44 0 0 0 0 0 0 0
4098 0 0 0 0 0 0 0 0
4099 13445 6107 12589056 5187 709690 0 0 0

BPID AsDatRds AsDatRdReq AsIdxRds AsIdxRdReq AsRdTime AsDatWrts AsIdxWrts AsWrtTime
4096 0 0 11 11 6 29511 2767 5386
4097 0 0 0 0 0 38 0 0
4098 0 0 0 0 0 0 0 0
4099 0 0 63 57 869 8559 1574 5864

BPID TotRdTime TotWrtTime VectIORds VectIOReq BlockIORds BlockIOReq FilesClose NoVictAvl UnRdPFetch
4096 40910016 152086 11 11 0 0 0 0 268983
4097 0 8 0 0 0 0 0 0 24
4098 0 0 0 0 0 0 0 0 0
4099 44878163 201491 63 57 0 0 0 0 186948
[db2inst1@WESSPTDW01 ~]$


--Where i can see only the hidden pools are activated instead of the

this is the bufferpool configuration of DB-2

[db2inst1@WESSPTDW01 ~]$ db2pd -db reporter -bufferpools

Database Partition 0 -- Database REPORTER -- Active -- Up 0 days 14:14:22 -- Date 05/12/2011 09:15:06

Bufferpools:
First Active Pool ID 1
Max Bufferpool ID 3
Max Bufferpool ID on Disk 3
Num Bufferpools 7

Address Id Name PageSz PA-NumPgs BA-NumPgs BlkSize NumTbsp PgsToRemov CurrentSz PostAlter SuspndTSCt Automatic
0x9B341110 1 IBMDEFAULTBP 8192 1000 0 0 5 0 1000 1000 0 True
0x9B3421C0 2 BUFPB16K 16384 1000 0 0 2 0 1000 1000 0 False
0x9B343270 3 REPBUF32K 32768 1000 0 0 1 0 1000 1000 0 False
0x99D7E380 4096 IBMSYSTEMBP4K 4096 16 0 0 0 0 16 16 0 False
0x9AF4D2B0 4097 IBMSYSTEMBP8K 8192 16 0 0 0 0 16 16 0 False
0x9AF4E360 4098 IBMSYSTEMBP16K 16384 16 0 0 0 0 16 16 0 False
0x9B340060 4099 IBMSYSTEMBP32K 32768 16 0 0 0 0 16 16 0 False

Bufferpool Statistics for all bufferpools (when BUFFERPOOL monitor switch is ON):

BPID DatLRds DatPRds HitRatio TmpDatLRds TmpDatPRds HitRatio IdxLRds IdxPRds HitRatio TmpIdxLRds TmpIdxPRds HitRatio
1 42563 366 99.14% 0 0 00.00% 13558 611 95.49% 0 0 00.00%
2 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%
3 190439 262 99.86% 0 0 00.00% 189808 48 99.97% 0 0 00.00%
4096 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%
4097 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%
4098 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%
4099 0 0 00.00% 0 0 00.00% 0 0 00.00% 0 0 00.00%

BPID DataWrts IdxWrts DirRds DirRdReqs DirRdTime DirWrts DirWrtReqs DirWrtTime
1 90 0 240368 1064 12249 0 0 0
2 0 0 0 0 0 0 0 0
3 219 3 757760 186 8948 0 0 0
4096 0 0 0 0 0 0 0 0
4097 0 0 0 0 0 0 0 0
4098 0 0 0 0 0 0 0 0
4099 0 0 0 0 0 0 0 0

BPID AsDatRds AsDatRdReq AsIdxRds AsIdxRdReq AsRdTime AsDatWrts AsIdxWrts AsWrtTime
1 12 4 0 0 42 90 0 5
2 0 0 0 0 0 0 0 0
3 0 0 0 0 0 190 3 27
4096 0 0 0 0 0 0 0 0
4097 0 0 0 0 0 0 0 0
4098 0 0 0 0 0 0 0 0
4099 0 0 0 0 0 0 0 0

BPID TotRdTime TotWrtTime VectIORds VectIOReq BlockIORds BlockIOReq FilesClose NoVictAvl UnRdPFetch
1 4092 5 12 4 0 0 0 0 0
2 0 0 0 0 0 0 0 0 0
3 1474 64 0 0 0 0 0 0 0
4096 0 0 0 0 0 0 0 0 0
4097 0 0 0 0 0 0 0 0 0
4098 0 0 0 0 0 0 0 0 0
4099 0 0 0 0 0 0 0 0 0
[db2inst1@WESSPTDW01 ~]$


------------

Any suggestion on how can it be configured, And when moving to 64bit, and having only 1 DB under this instance will have any improvement in performance .

And my Kernel configuration in 32bit is

[db2inst1@WESSPTDW01 ~]$ ipcs -l

------ Shared Memory Limits --------
max number of segments = 8192
max seg size (kbytes) = 4194303
max total shared memory (kbytes) = 1073741824
min seg size (bytes) = 1

------ Semaphore Limits --------
max number of arrays = 8192
max semaphores per array = 250
max semaphores system wide = 256000
max ops per semop call = 32
semaphore max value = 32767

------ Messages: Limits --------
max queues system wide = 32768
max size of message (bytes) = 65536
default max size of queue (bytes) = 65536

[db2inst1@WESSPTDW01 ~]$

---

what is suggested for 64 bit with the same RAM and memory in place.

The latest fixpacks of 9.7 (I would recommend at least FP3a or better would be FP4) will set most of the Linux kernel parms for you automatically, but you can see the recommened settings if you search the DB2 InfoCenter for "linux kernel".

Thanks let me try ...

i have one more doubt, i read in a document that we can assign 50% of physical RAM to buffer pool (all)

i have 32 Gb RAM ...

so if i assign 20000 8K pages then it becomes only 160 MB, as the same i have 4 more buffer pools which all togecther can come only upto 3 GB or max 4 GB, But i can use 16 GB,

in this case can i give more to these pages say it 50000 pages ??

----
correct me if wrong

If you take into account all of the DB2 databases on the server, and there are no other applications running directly on the server (it is dedicated to DB2 except for some OS processes), then you should be able to allocate about 50% of the physical memory to DB2 bufferpools (for all databases combined). Just make sure that DB2 is not using any virtual memory to handle your bufferpool allocations.

Keep in mind that it does no good to allocate more memory to a bufferpool than the total size of the tablespaces (which contain tables and indexes) assigned to that bufferpool.

By limiting bufferpool memory to about 50% of physical memory, you should have plenty of memory for DB2 instances (including application connections) and other database memory allocations. If 50% was still not enough for optimum performance, you probably could go higher for bufferpools so long as DB2 does not run out of real memory (not using virtual memory).

Obviously, this assumes you have 64-bit OS and 64-bit DB2.

Thanks for all your comments, now i have upgraded and have assigned the buffer pool a good value, so will wait for any bottleneck and come back for the tuning details.

thanks again