Thursday, September 11, 2014

Tuning individual Oracle SQL statements

For a complete understanding of SQL tuning steps, try the Advanced Oracle SQL Tuning, a 3-Day or 5-day on-site Oracle course.

Tuning individual Oracle SQL statements

The acronym SQL stands for Structured Query Language. SQL is an industry standard database query language that was adopted in the mid-1980s. It should not be confused with commercial products such as Microsoft SQL Server or open source products such as MySQL, both of which use the acronym as part of the title of their products.

Do this before you start individual SQL statement tuning

This broad-brush approach can save thousands of hours of tedious SQL tuning because you can hundreds of queries at once.  Remember, you MUST do this first, else later changes to the optimizer parameters or statistics may un-tune your SQL.
 
Remember, you must ALWAYS start with system-level SQL tuning, else later changes might undo your tuned execution plans:
  • Optimize the server kernel - You must always tune your disk and network I/O subsystem (RAID, DASD bandwidth, network) to optimize the I/O time, network packet size and dispatching frequency.
     
  • Adjusting your optimizer statistics - You must always collect and store optimizer statistics to allow the optimizer to learn more about the distribution of your data to take more intelligent execution plans.  Also, histograms can hypercharge SQL in cases of determining optimal table join order, and when making access decisions on skewed WHERE clause predicates.
     
  • Adjust optimizer parameters - Optimizer optimizer_mode, optimizer_index_caching, optimizer_index_cost_adj.
     
  • Optimize your instance - Your choice of db_block_size, db_cache_size, and OS parameters (db_file_multiblock_read_count, cpu_count, &c), can influence SQL performance.
     
  • Tune your SQL Access workload with physical indexes and materialized views - Just as the 10g SQLAccess advisor recommends missing indexes and missing materialized views, you should always optimize your SQL workload with indexes, especially function-based indexes, a Godsend for SQL tuning. 
11g Note:  The Oracle 11g SQL Performance Analyzer (SPA), is primarily designed to speed up the holistic SQL tuning process.

Once you create a workload (called a SQL Tuning Set, or STS), Oracle will repeatedly execute the workload, using sophisticated predictive models (using a regression testing approach) to accurately identify the salient changes to SQL execution plans, based on your environmental changes. Using SPA, we can predict the impact of system changes on a workload, and we can forecast changes in response times for SQL after making any change, like parameter changes, schema changes, hardware changes, OS changes, or Oracle upgrades.  For details, see the book Oracle 11g New Features.
Once the environment, instance, and objects have been tuned, the Oracle administrator can focus on what is probably the single most important aspect of tuning an Oracle database: tuning the individual SQL statements. In this final article in my series on Oracle tuning, I will share some general guidelines for tuning individual SQL statements to improve Oracle performance.

Oracle SQL tuning goals

Oracle SQL tuning is a phenomenally complex subject. Entire books have been written about the nuances of Oracle SQL tuning; however, there are some general guidelines that every Oracle DBA follows in order to improve the performance of their systems.  Again, see the book "Oracle Tuning: The Definitive Reference", for complete details.

The goals of SQL tuning focus on improving the execution plan to fetch the rows with the smallest number of database "touches" (LIO buffer gets and PIO physical reads). 

  • Remove unnecessary large-table full-table scans—Unnecessary full-table scans cause a huge amount of unnecessary I/O and can drag-down an entire database. The tuning expert first evaluates the SQL based on the number of rows returned by the query. The most common tuning remedy for unnecessary full-table scans is adding indexes. Standard b-tree indexes can be added to tables, and bitmapped and function-based indexes can also eliminate full-table scans. In some cases, an unnecessary full-table scan can be forced to use an index by adding an index hint to the SQL statement.
     
  • Cache small-table full-table scans—In cases where a full-table scan is the fastest access method, the administrator should ensure that a dedicated data buffer is available for the rows.  In Oracle8 and beyond, a small table can be cached by forcing it into the KEEP pool.
     
  • Verify optimal index usage—Oracle sometimes has a choice of indexes, and the tuning professional must examine each index and ensure that Oracle is using the proper index.
     
  • Materialize your aggregations and summaries for static tables - One features of the Oracle 10g SQLAccess advisor is recommendations for new indexes and suggestions for materialized views.  Materialized views pre-join tables and pre-summarize data, a real silver bullet for data mart reporting databases where the data is only updated daily. Again, see the book "Oracle Tuning: The Definitive Reference", for complete details on SQL tuning with materialized views.
These are the goals of SQL tuning in a nutshell. However, they are deceptively simple, and to effectively meet them, we need to have a through understanding of the internals of Oracle SQL. Let's begin with an overview of the Oracle SQL optimizers.

Oracle SQL optimizers

One of the first things the Oracle DBA looks at is the default optimizer mode for the database. The Oracle initialization parameters offer many cost-based optimizer modes as well as the deprecated yet useful rule-based hint:
The cost-based optimizer uses 'statistics' that are collected from the table using the 'analyze table' command. Oracle uses these metrics about the tables in order to intelligently determine the most efficient way of servicing the SQL query. It is important to recognize that in many cases, the cost-based optimizer may not make the proper decision in terms of the speed of the query. The cost-based optimizer is constantly being improved, but there are still many cases in which the rule-based optimizer will result in faster Oracle queries.
Prior to Oracle 10g, Oracle's default optimizer mode was called 'choose.' In the choose optimizer mode, Oracle will execute the rule-based optimizer if there are no statistics present for the table; it will execute the cost-based optimizer if statistics are present. The danger with using the choose optimizer mode is that problems can occur in cases where one Oracle table in a complex query has statistics and the other tables do not. 
Starting in Oracle 10g, the default optimizer mode is all_rows, favoring full-table scans over index access.  The all_rows optimizer mode is designed to minimize computing resources and it favors full-table scans.  Index access (first_rows_n) adds additional I/O overhead, but they return rows faster, back to the originating query:

Full-table scans touch all data blocks
Hence, many OLTP shops will choose first_rows, first_rows_100 or first_rows_10, asking Oracle to use indexes to reduce block touches:

Index scans return rows fast by doing additional I/O
Note:  Staring in Oracle9i release 2, the Oracle performance tuning guide says that the first_rows optimizer mode has been deprecated and to use first_rows_n instead.
When only some tables contain CBO statistics, Oracle will use the cost-based optimization and estimate statistics for the other tables in the query at runtime. This can cause significant slowdown in the performance of the individual query.
In sum, the Oracle database administrator will always try changing the optimizer mode for queries as the very first step in Oracle tuning. The foremost tenet of Oracle SQL tuning is avoiding the dreaded full-table scan. One of the hallmarks of an inefficient SQL statement is the failure of the SQL statement to use all of the indexes that are present within the Oracle database in order to speed up the query.

Of course, there are times when a full-table scan is appropriate for a query, such as when you are doing aggregate operations such as a sum or an average, and the majority of the rows within the Oracle table must be read to get the query results. The task of the SQL tuning expert is to evaluate each full-table scan and see if the performance can be improved by adding an index.

In most Oracle systems, a SQL statement will be retrieving only a small subset of the rows within the table. The Oracle optimizers are programmed to check for indexes and to use them whenever possible to avoid excessive I/O. However, if the formulation of a query is inefficient, the cost-based optimizer becomes confused about the best access path to the data, and the cost-based optimizer will sometimes choose to do a full-table scan against the table. Again, the general rule is for the Oracle database administrator to interrogate the SQL and always look for full-table scans. 
For the full story, see my book "Oracle Tuning: The Definitive Reference" for details on choosing the right optimizer mode.
A strategic plan for Oracle SQL tuning

Many people ask where they should start when tuning Oracle SQL. Tuning Oracle SQL is like fishing. You must first fish in the Oracle library cache to extract SQL statements and rank the statements by their amount of activity.

Step 1:  Identify high-impact SQL

The SQL statements will be ranked according the number of executions and will be tuned in this order. The executions column of the v$sqlarea view and the stats$sql_summary or the dba_hist_sql_summary table can be used to locate the most frequently used SQL.  Note that we can display SQL statements by:
  • Rows processed:  Queries that process a large number of rows will have high I/O and may also have impact on the TEMP tablespace.
     
  • Buffer gets:  High buffer gets may indicate a resource-intensive query.
     
  • Disk reads:  High disk reads indicate a query that is causing excessive I/O.
     
  • Memory KB:  The memory allocation of a SQL statement is useful for identifying statements that are doing in-memory table joins.
     
  • CPU secs:  This identifies the SQL statements that use the most processor resources.
     
  • Sorts:  Sorts can be a huge slowdown, especially if they're being done on a disk in the TEMP tablespace.
     
  • Executions:  The more frequently executed SQL statements should be tuned first, since they will have the greatest impact on overall performance.

Step 2:  Determine the execution plan for SQL

As each SQL statement is identified, it will be 'explained' to determine its existing execution plan. There are a host of third-party tools on the market that show the execution plan for SQL statements. The most common way of determining the execution plan for a SQL statement is to use Oracle's explain plan utility. By using explain plan, the Oracle DBA can ask Oracle to parse the statement and display the execution class path without actually executing the SQL statement.

To see the output of an explain plan, you must first create a 'plan table.' Oracle provides a script in $ORACLE_HOME/rdbms/admin called utlxplan.sql. Execute utlxplan.sql and create a public synonym for the plan_table:

 
sqlplus > @utlxplan
Table created.
 
sqlplus > create public synonym plan_table for sys.plan_table;
Synonym created.

Most relational databases use an explain utility that takes the SQL statement as input, runs the SQL optimizer, and outputs the access path information into a plan_table, which can then be interrogated to see the access methods. Listing 1 runs a complex query against a database.

EXPLAIN PLAN SET STATEMENT_ID = 'test1' FOR
SET STATEMENT_ID = 'RUN1'
INTO plan_table
FOR
SELECT   'T'||plansnet.terr_code, 'P'||detplan.pac1
|| detplan.pac2 || detplan.pac3, 'P1', sum(plansnet.ytd_d_ly_tm),
 sum(plansnet.ytd_d_ty_tm),
 sum(plansnet.jan_d_ly),
 sum(plansnet.jan_d_ty),
FROM plansnet, detplan
WHERE
    plansnet.mgc = detplan.mktgpm
AND
    detplan.pac1 in ('N33','192','195','201','BAI',
    'P51','Q27','180','181','183','184','186','188',
    '198','204','207','209','211')
GROUP BY 'T'||plansnet.terr_code, 'P'||detplan.pac1 || detplan.pac2 || detplan.pac3;
This syntax is piped into the SQL optimizer, which will analyze the query and store the plan information in a row in the plan table identified by RUN1. Please note that the query will not execute; it will only create the internal access information in the plan table. The plan tables contains the following fields:
  • operation:  The type of access being performed. Usually table access, table merge, sort, or index operation
     
  • options:  Modifiers to the operation, specifying a full table, a range table, or a join
     
  • object_name:  The name of the table being used by the query component
     
  • Process ID:  The identifier for the query component
     
  • Parent_ID:  The parent of the query component. Note that several query components may have the same parent.
Now that the plan_table has been created and populated, you may interrogate it to see your output by running the following query in Listing 2.
plan.sql - displays contents of the explain plan table
SET PAGES 9999;
SELECT  lpad(' ',2*(level-1))||operation operation,
        options,
        object_name,
        position
FROM plan_table
START WITH id=0
AND
statement_id = 'RUN1'
CONNECT BY prior id = parent_id
AND
statement_id = 'RUN1';
Listing 3 shows the output from the plan table shown in Listing 1. This is the execution plan for the statement and shows the steps and the order in which they will be executed.
SQL> @list_explain_plan
 
OPERATION
-------------------------------------------------------------------------------------
OPTIONS                           OBJECT_NAME                    POSITION
------------------------------ -------------------------------------------------------
SELECT STATEMENT
SORT
GROUP BY                                                      1
       CONCATENATION                                   1
NESTED LOOPS                                    1
TABLE ACCESS FULL         PLANSNET                   1
TABLE ACCESS BY ROWID     DETPLAN                    2
          INDEX RANGE SCAN       DETPLAN_INDEX5             1
NESTED LOOPS

From this output, we can see the dreaded TABLE ACCESS FULL on the PLANSNET table. To diagnose the reason for this full-table scan, we return to the SQL and look for any plansnet columns in the WHERE clause. There, we see that the plansnet column called 'mgc' is being used as a join column in the query, indicating that an index is necessary on plansnet.mgc to alleviate the full-table scan.

While the plan table is useful for determining the access path to the data, it does not tell the entire story. The configuration of the data is also a consideration. The SQL optimizer is aware of the number of rows in each table (the cardinality) and the presence of indexes on fields, but it is not aware of data distribution factors such as the number of expected rows returned from each query component.
 

Step 3:  Tune the SQL statement

For those SQL statements that possess a sub-optimal execution plan, the SQL will be tuned by one of the following methods:
 

  • Adding SQL 'hints' to modify the execution plan
     
  • Rewriting the SQL in PL/SQL. For certain queries this can result in more than a 20x performance improvement. The SQL would be replaced with a call to a PL/SQL package that contained a stored procedure to perform the query.
Using hints to tune Oracle SQL

Among the most common tools for tuning SQL statements are hints. A hint is a directive that is added to the SQL statement to modify the access path for a SQL query.
Troubleshooting tip!  For testing, you can quickly test the effect of another optimizer parameter value at the query level without using an 'alter session' command, using the new opt_param SQL hint:
select /*+ opt_param('optimizer_mode','first_rows_10') */ col1, col2 . . .
select /*+ opt_param('optimizer_index_cost_adj',20) */ col1, col2 . .
Oracle publishes many dozens of SQL hints, and hints become increasingly more complicated through the various releases of Oracle and on into Oracle.
Note:  Hints are only used for de-bugging and you should adjust your optimizer statistics to make the CBO replicate the hinted SQL.  Let's look at the most common hints to improve tuning:
  • Self-order the table joins - If you find that Oracle is joining the tables together in a sub-optimal order, you can use the ORDERED hint to force the tables to be joined in the order that they appear in the FROM clause.  See
     
  • Try a first_rows_n hint. Oracle has two cost-based optimizer modes, first_rows_n and all_rows. The first_rows mode will execute to begin returning rows as soon as possible, whereas the all_rows mode is designed to optimize the resources on the entire query before returning rows.  
    SELECT /*+ first_rows */
     

    A case study in SQL tuning

    One of the historic problems with SQL involves formulating SQL queries. Simple queries can be written in many different ways, each variant of the query producing the same result, but with widely different access methods and query speeds.

    For example, a simple query such as 'What students received an A last semester'' can be written in three ways, as shown in below, each returning an identical result.
  •  
    A standard join:
     
    SELECT *
    FROM STUDENT, REGISTRATION
    WHERE
        STUDENT.student_id = REGISTRATION.student_id
    AND
        REGISTRATION.grade = 'A';
     
    A nested query:

    SELECT *
    FROM STUDENT
    WHERE
        student_id =
        (SELECT student_id
            FROM REGISTRATION
            WHERE
           grade = 'A'
        );

    A correlated subquery:

    SELECT *
    FROM STUDENT
    WHERE
        0 <
        (SELECT count(*)
            FROM REGISTRATION
            WHERE
            grade = 'A'
            AND
            student_id = STUDENT.student_id
        );
    Let's wind up with a review of the basic components of a SQL query and see how to optimize a query for remote execution.

    Tips for writing more efficient SQL
    Space doesn't permit me to discuss every detail of Oracle tuning, but I can share some general rules for writing efficient SQL in Oracle regardless of the optimizer that is chosen. These rules may seem simplistic but following them in a diligent manner will generally relieve more than half of the SQL tuning problems that are experienced:

    • Rewrite complex subqueries with temporary tables - Oracle created the global temporary table (GTT) and the SQL WITH operator to help divide-and-conquer complex SQL sub-queries (especially those with with WHERE clause subqueries, SELECT clause scalar subqueries and FROM clause in-line views).  Tuning SQL with temporary tables (and materializations in the WITH clause) can result in amazing performance improvements.
       
    • Use minus instead of EXISTS subqueries - Some say that using the minus operator instead of NOT IN and NOT Exists will result in a faster execution plan.
       
    • Use SQL analytic functions - The Oracle analytic functions can do multiple aggregations (e.g. rollup by cube) with a single pass through the tables, making them very fast for reporting SQL.
       
    • Re-write NOT EXISTS and NOT EXISTS subqueries as outer joins - In many cases of NOT queries (but ONLY where a column is defined as NULL), you can re-write the uncorrelated subqueries into outer joins with IS NULL tests.  Note that this is a non-correlated sub-query, but it could be re-written as an outer join.
    select book_key from book
    where
    book_key NOT IN (select book_key from sales);
    Below we combine the outer join with a NULL test in the WHERE clause without using a sub-query, giving a faster execution plan.
    select b.book_key from book b, sales s
    where
       b.book_key = s.book_key(+)
    and
       s.book_key IS NULL;
    • Index your NULL values - If you have SQL that frequently tests for NULL, consider creating an index on NULL values.  To get around the optimization of SQL queries that choose NULL column values (i.e. where emp_name IS NULL), we can create a function-based index using the null value built-in SQL function to index only on the NULL columns. 
       
    • Leave column names alone - Never do a calculation on an indexed column unless you have a matching function-based index (a.k.a. FBI).  Better yet, re-design the schema so that common where clause predicates do not need transformation with a BIF:
    where salary*5            > :myvalue
    where substr(ssn,7,4)     = "1234"
    where to_char(mydate,mon) = "january"
    • Avoid the use of NOT IN or HAVING. Instead, a NOT EXISTS subquery may run faster (when appropriate).
       
    • Avoid the LIKE predicate = Always replace a "like" with an equality, when appropriate.
       
    • Never mix data types - If a WHERE clause column predicate is numeric, do not to use quotes. For char index columns, always use quotes. There are mixed data type predicates:
    where cust_nbr = "123"
    where substr(ssn,7,4) = 1234
    • Use decode and case - Performing complex aggregations with the "decode" or "case" functions can minimize the number of times a table has to be selected.
       
    • Don't fear full-table scans - Not all OLTP queries are optimal when they uses indexes.  If your query will return a large percentage of the table rows, a full-table scan may be faster than an index scan.  This depends on many factors, including your configuration (values for db_file_multiblock_read_count, db_block_size), query parallelism and the number of table/index blocks in the buffer cache.
       
    • Use those aliases - Always use table aliases when referencing columns.
    Also, see these related SQL tuning notes:

    No comments:

    Post a Comment