Gogala, Folks
GOAL - To Compare performances (monitored readings), of the Primary Database in various Standby modes , with each other & also against a Primary database WITHOUT Standby (performance baseline)
Qs Is the following approach suitable?
Approach – 1) On WAN, Only 2 (of 3) Physical Standby modes of Dataguard Benchmarked. 2) Application Transaction Completion Timings & Transaction per Second TPS achieved ( Transactions OLTP in nature ) 3) CPU & Memory Consumptions on Primary & Standby Databases
4) Using a Solaris Tool (netload) Utilization on the Network Card (hme0) being monitored :-
Sample Output :-
Inpackets Outpackets Inbytes Outbytes Collisions Utilization 634 1300 42216 1385402 2 1.17 1170 2341 77402 2502201 3 2.12 850 1624 56394 1731692 1 1.47
5) Using ethereal Freeware Tool (www.ethereal.com) :- Volume of Data (Traffic) Flowing between the Primary & Standby Database machines captured
6) On LAN, All 3 Physical Standby modes of Dataguard yet to be Benchmarked 7) Logical Standby Database Benchmark NOT being considered
Qs – How may the following Waits be addressed?
WAN Simulator Setting for the below results :-
Thruput = 2 MHz = 2 MBPS Delay = 0 ms
WAN Simulator present between Primary Database machine & Standby Database machine
SAME Application Transaction Load Fired on the Database in the various modes NOTE – All Other conditions kept Same
CASE 1 RUN WITHOUT Standby Database (dataguard)
Top 5 Timed Events ~~~~~~~~~~~~~~~~~~ % Total Event Waits Time (s) Ela Time -------------------------------------------- ------------ ----------- -------- log file sync 22,601 181 34.76 library cache pin 62 121 23.17 CPU time 110 21.10 db file parallel write 391 51 9.85 db file sequential read 9,951 28 5.34 -------------------------------------------------------------
CASE 2 RUN with Dataguard mode set to MAXIMIZE PERFORMANCE
i.e. log_archive_dest_2 = 'SERVICE=phy arch'
RESULT - Average APPLICATION Transaction Time & TPS is SAME as that of Without Standby Database (Case1)
Top 5 Timed Events ~~~~~~~~~~~~~~~~~~ % Total Event Waits Time (s) Ela Time -------------------------------------------- ------------ ----------- -------- ARCH wait on SENDREQ 135 389 52.05 log file sync 18,630 165 22.07 CPU time 94 12.52 library cache pin 18 39 5.24 db file parallel write 283 17 2.31 -------------------------------------------------------------
CASE 3 Dataguard mode Set to MAXIMIZE AVAILABILITY
i.e. log_archive_dest_2 = 'SERVICE=phy LGWR Sync Affirm'
RESULT - Average APPLICATION Transaction Time & TPS is MUCH Higher than of Cases 1 & 2
Top 5 Timed Events ~~~~~~~~~~~~~~~~~~ % Total Event Waits Time (s) Ela Time -------------------------------------------- ------------ ----------- -------- log file sync 17,355 4,129 74.13 LGWR wait on LNS 92,198 368 6.61 LGWR-LNS wait on channel 42,991 367 6.60 LNS wait on SENDREQ 6,187 358 6.43 library cache pin 53 96 1.71 -------------------------------------------------------------
Will Provide any data needed
Thanks
-----Original Message-----
No. I mean the goals of the benchmark. What are you measuring and why.
On 2003.09.16 06:10, VIVEK_SHARMA wrote: > Hi > > Plan of the Benchmark = ? If you mean an action plan of important steps > ? > > We have just started. I Shall send it to you shortly. > > Monitoring of Actual Bandwidth thruput (MBPS) used versus that set on > WAN will be done . > > (other than actual application transaction thruput rate (TPS) , OS > resources Utilizations ) > > > > Thanks indeed > > -----Original Message----- > Sent: Tuesday, September 16, 2003 2:59 AM > To: Multiple recipients of list ORACLE-L > > > > Did you plan that benchmark? What did you decide to monitor in the > planning phase? > > -- > Mladen Gogala > Oracle DBA > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > Behalf Of VIVEK_SHARMA > Sent: Monday, September 15, 2003 1:49 PM > To: Multiple recipients of list ORACLE-L > Subject: Dataguard Benchmark > > We are doing a Data guard Benchmark. > > INFO.:- > > ----- > > WAN Simulator :- > > We have a WAN Simulator with 2 routers at either ends of it. > > Thruputs from 0 to 2 MBPS can be manually set as is required by > the run. > > > > Application = Banking :- > > Transactions mainly OLTP in nature (Both DML & SELECTS) . > > We can do CPU intensive batch Transactions too if advised by you > folks > > > > Machines = 2 machines of 4 CPUs each > > Memory = 8 GB on each machine > > O.S. = Solaris 9 > > Oracle = 9.2 > > Sniffer network tool ( to get volume of bytes transferred over > the WAN ) > > Dataguard Setup will transfer Data thru the listener services :- > > i.e. init.ora - LOG_ARCHIVE_DEST_2 = > 'SERVICE=<SERVICE_NAME_FROM_TNSNAMES.ORA>,. . .' > > > > Execution methodology:- > > > > Run Same Transaction's Volume in BOTH Logical & Physical > (Maximum Protection , Maximum Availability , Maximum performance ) > standby modes > > Qs What readings to be particularly monitored & measured? > > Qs What thruput bandwidths should be benchmarked? > > Qs Does total Size of Existent Database-in-use matter to the > Benchmark? > > Current Database Size = 3 GB > > Qs Does RMAN setup add any value to the Dataguard benchmark in > some way? > > Else we will do Without RMAN, manually altering the various > modes > > Qs Any Sample Docs, Links on existing Dataguard Benchmarks? > > Qs Any else that will enable us to bring out a paper of > reasonable standard?
|