Oracle Application Server 10g

Tuning Techniques

Donald K. BurlesonBurleson Oracle Consulting

Oracle Training by Don Burleson

www.rampant-books.comOracle Books from $9.95

Guidehorse.com

On-site custom Oracle training

Oracle Tuning & Oracle Support

Remote DBA Support

Topics: Oracle Application Server 10g Tuning

Approach

Oracle Application Server 10g Monitoring

Tuning with RAM

Load Balancing

Oracle Application Server 10g Architecture

WebCache

WebCache

WebCache

WebCache

HTTPServer

HTTPServer

HTTPServer

HTTPServer

HTTPServer

HTTPServer

Database Files

RACServer

RACServer

RACServer

RACServer

RACServer

Internet

Keys to Success

Parameter Tuning

RAM Cache Tuning

Server Tuning

Parameter Tuning

Oracle Application Server 10g parameters - Adjusting the Oracle9iAS configuration parameters for each Oracle9iAS component has influence performance and throughput.

Database parameters – Because most Oracle9iAS systems are disk I/O intensive, adjusting the Oracle database parameters for the Infrastructure database (iasdb) and the back-end database can heavily influence performance.

RAM Tuning

Data buffer tuning – Adding RAM to the database db_cache_size on the Oracle Infrastructure and back-end database can greatly reduce disk I/O and improve throughput.

Web cache tuning – Adding RAM to the Oracle9iAS web cache can improve the delivery rates of HTML and XML though the Oracle HTTP Server (OHS).

Server tuning Hardware configuration – Adding RAM of CPU

resources to existing servers will improve the throughput on the server

Hardware load balancing – The addition of new servers to the Oracle9iAS farm and relocating Oracle9iAS components. Spare servers can be configured with both Web Cache and App Server, and the appropriate components can be started as-needed.

Server parameter tuning – Adjusting the parameters on your server can have a huge impact on the performance of the Oracle Application Server 10g.

Monitoring Techniques

Response Time Monitoring – DCM and OEM

Wait Event Monitoring – Determine the source of Latency for each Component.

Server Resources – Once the farm is tuned, overloads can be addressed with dynamic server allocation.

Wait Event Monitoring (for isadb and database)

How would you tune this database?                            % Total

Event                           Waits  Time (s)  Ela Time------------------------------  -----  --------  --------CPU time                                     30     71.43db file parallel write             95         1     23.53control file sequential read       54         1      2.33log file parallel write            62         0       .95db file sequential read            20         0       .68

                            How would you tune this database? % Total

Event                           Waits  Time (s)  Ela Time------------------------------  -----  --------  --------db file sequential read   45        22     41.43db file scattered read             95        14     25.55control file sequential read       54         1      2.33log file parallel write            62         0       .95db file parallel write             20         0       .68

Oracle Application Server 10g Monitoring

Dynamic Monitoring Service (DMS)

OC4J – Measure Parse Time for Incoming Request and Free RAM in the JVM

Portal – Display Portal Metrics

Servlet – Instrument Servlets to Generate Performance Metrics

OHS – Measure Active HTTP Requests

DMS has over 300 metrics

dmstool -l |grep completed

/appsvr/OC4J:3303:6004/oc4j/default/WEBs/parseRequest.completed/appsvr/OC4J:3303:6004/oc4j/default/WEBs/processRequest.completed/appsvr/OC4J:3303:6004/oc4j/default/WEBs/resolveContext.completed

/appsvr/OC4J:3303:6004/oc4j/portal/WEBs/parseRequest.completed/appsvr/OC4J:3303:6004/oc4j/portal/WEBs/processRequest.completed/appsvr/OC4J:3303:6004/oc4j/portal/WEBs/resolveContext.completed

/ap/OC4J:3303:6004/oc4j/syndserver/WEBs/parseRequest.completed/ap/OC4J:3303:6004/oc4j/syndserver/WEBs/processRequest.completed/ap/OC4J:3303:6004/oc4j/syndserver/WEBs/resolveContext.completed

Collect 100 sets at 60 second intervals

dmstool -i 60 -c 100 \/appsvr/Apache:2534:6004/Apache/handle.completed \/appsvr/Apache:2534:6004/Apache/request.completed \/appsvr/Apache:2534:6004/Apache/handle.completed \/appsvr/Apache:2534:6004/Apache/request.completed >> t1.lst

Output ListingSun Jul 13 20:19:43 MDT 2003

/appsvr/Apache:2534:6004/Apache/handle.completed 240320 ops/appsvr/Apache:2534:6004/Apache/request.completed 146504 ops/appsvr/Apache:2534:6004/Apache/connection.completed 56908 ops

Compute delta in spreadsheet

Plot with Chart Wizard

DMS can be scripted:

#!/bin/ksh

PATH=$PATH:/home/oracle/oraportal904/binexport PATH

# Dump Stats for Later Analysisdmstool -dump >> dumparch.lst

# Dumping OHS Stats to a Filedmstool -table ohs_server >> ohs.lst

Sending OHS stats to a flat file:

# Dumping OHS Stats to a File

dmstool -table ohs_server >> ohs.lst

cat ohs.lst|grep connection.active > con_active.lstcat ohs.lst|grep request.active > req_active.lst cat ohs.lst|grep busyChildren.value > busy_child.lstcat ohs.lst|grep readyChildren.value > readyChild.lstcat ohs.lst|grep numChildren.value > det.lst

OHS Server Output

Sun Jul 13 21:01:45 MDT 2003

----------ohs_server----------busyChildren.value: 16...childStart.count: 24748 opsconnection.active: 24 threads...numChildren.value: 44...readyChildren.value: 27...request.avg: 15321 usecsrequest.completed: 150942 ops...

Plotting OHS response time

OHS Response time in milliseconds

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

1 6 11 16 21 26 31 36 41 46 51 56 61 66

Active Connections

Mill

ise

co

nd

s

Response time in milliseconds

List OHS performance metricsdmstool -table ohs_module -c 1

Name: mod_oc4j.c ...decline.count: 13487 opshandle.active: 0 threadshandle.avg: 3 usecshandle.completed: 13487 opshandle.maxTime: 8 usecshandle.minTime: 2 usecshandle.time: 43710 usecs

Name: http_core.c...decline.count: 0 opshandle.active: 0 threadshandle.avg: 0 usecshandle.completed: 0 opshandle.maxTime: 0 usecs

Hard to parse The output

Computing real response timeOne of the problems with the OHS statistics is that the one-time operations will skew

the overall averages in the ohs_response listings.

(time – min – max)real_average = ------------------------ (completed – 2)

Using the data from the previous mod_oc4j.c listing, we can compute the real response time:

(43,710 – 2 – 8)real_average = ------------------------ (13,487 – 2)

(43,700)

real_average = ------------------ = 3.24 milliseconds (13,485)

Using Aggrespy

Web Cache Monitoring

Oracle Application Server 10g Web Cache

InternetWeb

CacheWeb Server

Web Server

Database

Trigger

Programmatic

Web Cache Tuning

Static and Dynamic Information

Cacheability Rules

Cache Invalidations

Multi-version HTML

Rule for Each Page Component

Web Cache statistics:

Requests – This shows the current, average and max transaction per second. The backlog section indicates that the Web Cache is overwhelmed and another Web Cache server should be started.

Errors – This summarized the network, site busy and particle-page errors for the Web Cache.

Misses – This section shows cacheable and non-cacheable misses along with the number of refreshes for the Web Cache.

Compression – The compression sections show the total amount of RAM saved by compression and provides a great gauge of the effectiveness of the Web Cache.

Oracle Application Server 10g Load Balancing

Software Load Balancing

Web Cache to OHS – Web Cache interrogates OHS statistics and routes to least loaded.

OHS to Database Listener – OHS Distributes load to multiple Listeners

Database Listener – Listeners to Multiple Dispatchers under MTS, that load balance to least loaded RAC Instance.

Oracle Application Server 10g Load Balancing

WebCache

Web Server

Web ServerDatabase

Web ServerWeb

Cache

Instance

Instance

Instance

ApplicationServer Tier

WebCache Tier

DatabaseServer Tier

Hardware Load Balancing

WebCache

WebCache

WebCache

WebCache

HTTPServer

HTTPServer

HTTPServer

HTTPServer

HTTPServer

HTTPServer

Database Files

RACServer

RACServer

RACServer

RACServer

RACServer

InternetBlade Server Rack

OHS & WC

OHS & WC

OHS & WC

OHS & WC

Oracle RAC

Oracle RAC

Oracle RAC

Oracle RAC

Oracle RAC

8-node RAC with 8 extra nodes pre-defined:

cat cmcfg.ora

HeartBeat=15000ClusterName=TESTPollInterval=1000MissCount=210ServicePort=9998KernelModuleName=hangcheck-timerPrivateNodeNames= int-rac1 int-rac2 int-rac3 int-rac4 int-rac5 int-rac6 int-rac7 int-rac8 int-rac9 int-rac10 int-rac11 int-rac12 int-rac13 int-rac14 int-rac15 int-rac16PublicNodeNames= rac1 rac2 rac3 rac4 rac5 rac6 rac7 rac8 rac9 rac10 rac11 rac12 rac13 rac14 rac15 rac16HostName=rac1CmDiskFile=/mnt/ps/db/quorum

Dynamically add a 9th RAC node:

a) Add a Thread and enable the thread

b) Start the 9th instance

ALTER DATABASE ADD LOGFILE THREAD 9 SIZE 100M

ALTER DATABASE ADD LOGFILE THREAD 9 SIZE 100M;

ALTER DATABASE ENABLE PUBLIC THREAD 9;

startup pfile=initrac9ora

Monitoring Servers with vmstatSAMPLE_TIME=300

while truedo vmstat ${SAMPLE_TIME} 2 > /tmp/msg$$

# run vmstat and direct the output into the Oracle table . . . cat /tmp/msg$$|sed 1,3d | awk '{ printf("%s %s %s %s %s %s\n", $1, $8, $9, 14, $15, $16) }' | while read RUNQUE PAGE_IN PAGE_OUT USER_CPU SYSTEM_CPU DLE_CPU do

$ORACLE_HOME/bin/sqlplus -s perfstat/perfstat@iasdb<<EOF insert into perfstat.stats\$vmstat values ( sysdate, $SAMPLE_TIME, '$SERVER_NAME', $RUNQUE, $PAGE_IN, $PAGE_OUT, $USER_CPU, $SYSTEM_CPU, $IDLE_CPU, 0 ); EXITEOF donedone

rm /tmp/msg$$

Monitoring Servers with vmstatroot> vmstat 5 5

kthr memory page faults cpu ----- ----------- ------------------------ ------------ -----------r b avm fre re pi po fr sr cy in sy cs us sy id wa 7 5 220214 141 0 0 0 42 53 0 1724 12381 2206 19 46 28 79 5 220933 195 0 0 1 216 290 0 1952 46118 2712 40 55 0 513 5 220646 452 0 14 1 33 54 0 2130 86185 3014 38 59 0 36 5 220228 672 0 0 0 0 0 0 1929 25068 2485 25 49 16 10

Assuming an 8 CPU server:

CPU has enqueues when runqueue (r column) > cpu_count

RAM is paging when scan rate (sr) peaks before page-in (pi)

Server exception reports

Wed Dec 20                                                             page    1                                 run queue > 2                         May indicate an overloaded CPU.                    When runqueue exceeds the number of CPUs                  on the server, tasks are waiting for service.

SERVER_NAME     date       hour      runq pg_in pg_ot  usr  sys  idl       --------------- -------------------- ---- ----- ----- ---- ---- ----       AD-01           00/12/13    17          3     0     0   87    5    8      

               Whenever Unix performs a page-in, the RAM memory          on the server has been exhausted and swap pages are being used.

SERVER_NAME       date       hour      runq pg_in pg_ot  usr  sys  idl       ----------------- -------------------- ---- ----- ----- ---- ---- ----       AD-01             00/12/13    16          0     5     0    1    1   98       AD-01             00/12/14    09          0     5     0   10    2   88       AD-01             00/12/15    16          0     6     0    0    0  100       AD-01             00/12/19    20          0    29     2    1    2   98       PROD1DB           00/12/13    14          0     3    43    4    4   93       PROD1DB           00/12/19    07          0     2     0    1    3   96       PROD1DB           00/12/19    11          0     3     0    1    3   96           

Fix for Server Stress

Overloaded CPU

– Offload Task to Another Server– Add CPUs– Add Additional Instances/Servers

Overloaded RAM

– Add RAM Cheap $1k/gig– Reallocate RAM from Other Components

RAM Disk Solution

Disk I/O remains the biggest bottleneck 100 gig RAM costs $100k 6,000 times faster than disk for Oracle

Your app will still run inefficiently, but it runs 6,000 times faster!

UNIX server Monitoring rules: The UNIX vmstat utility provides a wealth of information about

the ongoing performance of the Oracle9iAS server.

The vmstat run queue value (r) can indicate a CPU shortage whenever the run queue exceeds the number of CUs on the server.

The vmstat page in values (pi) can indicate a RAM memory shortage.

You can easily define vmstat extension table to hold historical server information and use a UNIX shell script to periodically collect server performance information.

The UNIX server information can be used to generate alert reports and long-term trend reports.

Total Response Time:

Forms Server Time

FormServer

Client DatabaseServer

Time

Time

Time

Database

Client

Network

Extend iasdb for performance monitoring:

create table  FormStats (       FORM_ID VARCHAR2(120),      EVENT       VARCHAR2(120),      FSERVER  NUMBER,      DBASE       NUMBER,      NWORK      NUMBER,      CLIENT      NUMBER,      DATE        DATE) ;

Plotting response time data:

Top offending Forms:Top 10 Forms and Events that use the most Average Form Server Timewith a minimum of 10 executions and greater than 2 seconds for

execution.

1. Form: d:\prod\forms\F_END_USER_GENERATED_LETTERS.fmx Event: CLICK F_END_USER_GENERATED_LETTERS BUTTONS SAVE_BTN 1 MOUSE Avg Tm: 5.00 Seconds. Number of Executions: 62

2. Form: d:\prod\forms\F_PC_PICK_RETURNS.fmx Event: CLICK F_PC_PICK_RETURNS BUTTONS PROCESS 1 MOUSE Avg Tm: 4.00 Seconds. Number of Executions: 13

Top 10 Forms and Events that use the most Average Database Timewith a minimum of 2 executions and greater than 5 seconds for execution.

1. Form: d:\prod\forms\f_pc_case_maint.fmx Event: CLICK F_DIARY DIARY_TAB_ALLOUT DATE_OF_INCIDENT 9 Avg Tm: 472.00 Seconds. Number of Executions: 2

Conclusions

Develop a proactive, time-based performance data collection scheme. Real-time OEM and Aggrespy metrics are of little use.

Optimize by adjusting RAM resources parameters.

Once the system is optimized, server monitoring is critical

Server Load Balancing is Critical to properly scale Oracle9iAS

Current Books by Don Burleson

My web site: dba-oracle.com

E-mail me at : don@burleson.cc