Tuning your applications with the Quest Central Performance pack

1
Health Check your Database A Performance Tuning
Methodology
Robert Wijnbelt
2
WHAT IS PERFORMANCE TUNING?

• A combination of identifying and reacting to
  performance problems
• Proactively preventing such problems from
  occurring
• Key responsibility for Oracle DBAs
• Ensures that service levels are being met
• Avoids costly hardware upgrades

3
Why is performance tuning important?

• At the core of business logic, and at the core
  of business data for most production
  applications, is a relational database management
  system
• Ray Paquet of Gartner, July, 2002
• No access no data no business
• Includes poor performance, missed SLAs

4
MAJOR CAUSES OF PROBLEMS

• Poorly tuned application code SQL PL/SQL
• Contention for internal Oracle Resources locks,
  latches buffers
• IO bottlenecks
• Inadequate hardware resources

5
Procrastination?

• Only when the application and RDBMS is tuned can
  the possibility of adding hardware be considered
• Don Burleson
• Number one resolution to performance issues in
  past three years was to upgrade hardware, add
  memory and add disk.
• Problems still persist!

6
Health CheckBest Practices?

• Where are the most common problems past and
  present?
• How do I resolve those problems?
• How much benefit can I get from fixing the
  problems identified?
• How much effort is involved in resolution?
• What can I do to prevent future issues?

7
Tuning Methodology Best Practices

• Identify issues happening now
• Resolve quickly
• Find problems in the past
• Resolve methodically
• How to prevent future occurrences
• Set Goals
• What and How to tune

8
APPLICATION TUNING Resolutions

• The greatest and most common performance gains
  are attained through SQL tuning
• A relative effort is modifying improper parameter
  settings
• Just as important is resolving management of
  database space and data layout

9
4 Step Action Plan For a Database Healthcheck
10
Step 1 Optimize the Application Workload

• Effectively use the Oracle Optimizer
• Set optimizer mode to RULE, COST or CHOOSE
• Optimize SQL Statements
• Set Optimizer initialization parameters
• Determine SQL Code needs optimization
• SQL Coding Best Practices
• PL/SQL coding Best Practices

11
Step 2 Reduce Contention

• Identify and reduce Latch contention
• Identify and reduce Lock contention
• Optimize Redo Log configuration
• Reduce other internal contention

12
Identify and reduce Latch contention

• Types of Latches in the SGA that suffer most
  contention
• Buffer Cache Latches, caused by
• SQL statements with very high logical or physical
  I/O, due to unselective indexes (large index
  range scans) or many full table scans.
• DBWR not keeping up with the dirty workload,
  which forces the foreground process to hold the
  latch longer looking for a free buffer.
• Undersized buffer cache.
• Library Cache Latches
• Misses on this latch occur when SQL is executed
  at very high rates. There is little you can do
  to reduce the load on this latch, although using
  private rather than public synonyms (or even
  direct object references such as OWNER.TABLE) can
  help.
• Use bind variables in your code

13
Identify and reduce Lock contention

• Often caused in the application code and Data
  model
• Code for Concurrent use
• Index Foreign Key relations
• Reduce Ad-Hoc querying
• Try to reduce Updates

14
Optimize Redo Log configuration

• Use Fast disks
• avoid RAID 5, use 01 or Filesystem level
  replication
• Multiplex Your RedoLogs
• Use NOLOGGING table operations
• Use multiple Disks for different Group members
• Adjust Log Buffer size

15
Reduce Shared Pool contention

• Rewrite SQL
• Pin PL/SQL objects or SQL cursor objects in the
  shared pool, use the DBMS_SHARED_POOL.KEEP
• Fully qualify tables and objects with the schema
  owner name. This will eliminate some reparsing
  requirements.
• Increase the shared pool size when shared cursors
  are being flushed out
• Decrease the shared pool size when the
  application does not use bind variables and when
  cursors are not shared and reused.

16
Reduce Shared Pool contention continued

• consider setting the parameter cursor_sharing
  force (first available in 8.1.6). When this
  parameter is set, Oracle replaces embedded
  literal values with bind variables prior to
  parsing the statement, to avoid additional hard
  parses for SQL statements that differ only by
  literal values.
• Add cursors to stored procedures/packages. Stored
  procedures/packages can be pinned in the shared
  pool, which prevents them from being flushed and
  subsequently reparsed.
• Use the same bind variable length and array size
  in SQL statements. When an SQL statement with
  different bind variable lengths is used
  throughout an application, it cannot be shared.

17
Step 3 Physical IO Optimization

• Tune Sorting parameters (Sort_Area)
• Tune SQL (avoid Full Table Scans)
• Reduce Row Chaining and Migration
• Balance IO
• Increase number of datafile devices
• Implement OFA
• Increase DB Block size

18
Step 4 Consider Best Practices

• Tune Top Down
• Start with Network, IO, Instance and Database
  design before tuning SQL
• Set Benchmarks
• Identify some SQL statements, and rerun them as
  test
• Identify and monitor critical modules
• Quantify the problem and the expectations.
• Make sure there is not some problem with the
  infrastructure, get everyone involved.
• Involve the developers in your tuning efforts.

19
Best Practices continued

• Understand all the features of the database you
  are using. Use as many standard Oracle provided
  functions as possible.
• Learn how to generate and read execution plans.
• Tune to reduce logical IOs. Reducing the number
  of LIOs will reduce both overall CPU usage as
  well as physical IO requirements. Generally,
  index usage will reduce LIO requirements, but
  sometimes the reverse it true.

20
Best Practices continued

• Consider that sometimes writing PL/SQL can
  ultimately improve performance over SQL.
• If you have lots of large queries doing full
  table scans, make sure that parallel query is
  configured correctly.
• If Oracle is performing many hash joins, make
  sure the database parameters associated with hash
  joins (e.g. hash_area_size) are set correctly.

21
Quest Central Health Check

• A free service from Quest in which our
  consultants use Quest Central Performance
  management to diagnose the health of your database

22
What does it do?

• Collects Metrics using PL/SQL packages at a user
  defined interval
• Stores that information in a repository
• Allows you to run analysis against any number of
  those stored collections
• Produces problems, recommendations, solutions,
  reports, and goals for tuning

23
WHAT IS QUEST CENTRAL

• An integrated comprehensive solution for Oracle
  DBAs
• Real-time and Historical Diagnostics and Analysis
  
• Space Management
• SQL Tuning
• Database Administration

24
THANK YOU FOR LISTENING
Register for a Free Database Health Check
http//www.quest.com/healthcheck/
Robert Wijnbelt