Oracle data block size

In this example I have experiment of oracle data block size with 8k and 16K. I performed DML operation against both 8k and 16k data block size. I created two tables table_8k under 8k tablespace tbs_8k and table_16k under 16k tablespace tbs_16k. The summary of the experiment is bigger data block cause more time while update but less time while insert operation.

A)I used OMF file system. So I set db_create_file_dest.
SQL> show parameter db_create_file_dest
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_create_file_dest string /oradata2

B)Create 8k and 16K blocksize tablespace
SQL> create tablespace tbs_8k blocksize 8k;
Tablespace created.

SQL> alter system set db_16k_cache_size=20M;
System altered.

SQL> create tablespace tbs_16k blocksize 16k;
Tablespace created.

C)create table_8k under 8k blocksize tablespace and table_16k under 16k blocksize tablespace and insert random data into them.
SQL> create table table_8k (n number ,k varchar2(15)) tablespace tbs_8K;
Table created.

SQL> begin
for i in 1 .. 100000
loop
insert into table_8k values(i,'pc-'||round(dbms_random.value(1,20000),0));
end loop;
end;
/
PL/SQL procedure successfully completed.
Elapsed: 00:00:12.51

SQL> create table table_16K (n number ,k varchar2(15)) tablespace tbs_16k;
Table created.

SQL> begin
for i in 1 .. 100000
loop
insert into table_16k values(i,'pc-'||round(dbms_random.value(1,20000),0));
end loop;
end;
/

PL/SQL procedure successfully completed.
Elapsed: 00:00:09.42

D)I created one script named flush will will flush data from buffer cache and shared pool so that we get actual result to complete operation.
SQL> !vi flush.sql
alter system flush buffer_cache;
alter system flush shared_pool;

SQL> @flush
System altered.
Elapsed: 00:00:00.03

System altered.
Elapsed: 00:00:00.03

SQL> select * from table_8k where k like '%888%';
176 rows selected.
Elapsed: 00:00:00.14

SQL> select * from table_16k where k like '%888%';
195 rows selected.
Elapsed: 00:00:00.04


So we see for 16k select is faster.
E) Now test for update operation.
SQL> @flush
SQL> update table_16k set k='Testing';
100000 rows updated.

Elapsed: 00:00:06.88

SQL> @flush
System altered.
Elapsed: 00:00:00.24
System altered.
Elapsed: 00:00:00.02

SQL> update table_8k set k='Testing';
100000 rows updated.
Elapsed: 00:00:01.91

We see update is fairly faster in 8k tablespace.
SQL> delete from table_8k;
100000 rows deleted.
Elapsed: 00:00:09.53

SQL> delete from table_16K;
100000 rows deleted.
Elapsed: 00:00:09.27

Comments

Post a Comment

Popular posts from this blog

ORA-00923: FROM keyword not found where expected

Problem Symptom While performing a simple select statement it fails with error ORA-00923: FROM keyword not found where expected like below. SQL> select 48*1024*1024*1024 decimal from dual; select 48*1024*1024*1024 decimal from dual * ERROR at line 1: ORA-00923: FROM keyword not found where expected Cause of the problem The ORA-00923 can be caused by two different reasons. 1)Oracle itself expect FROM keyword in the position but it could not find any. Like below, SQL> select 48*1024 om dual; select 48*1024 om dual * ERROR at line 1: ORA-00923: FROM keyword not found where expected As no from keyword found so error arises. Oracle expected from in place of keyword dual. 2)The secong cause can sometimes mislead you. Like in case of SQL> select 48*1024*1024*1024 decimal from dual; select 48*1024*1024*1024 decimal from dual * ERROR at line 1: ORA-00923: FROM keyword not found where exp
Read more

How to make partitioning in Oracle more Quickly

As the table size grows and full table scans happens more frequently then there is no alternative than partitioning in oracle. Partition greatly enhance performance of a query. I will start my partitioning system simply with an example. In my database the following query took 30~35 seconds to complete. After partitioning the performance increased amazingly and it then took only 1 second. My query was, SELECT DISTINCT fs.type, fs.id , fs.pid, fs.cid , fs.cr_id, fs.created_date FROM summary fs where fs.id in (select fa.id from forms fa where fa.sar in (select la.sar from login la where la.login_id=1 and fa.created_date BETWEEN to_date('1-MAY-2008') AND to_date('21-SEP-2008') and fs.created_date BETWEEN to_date('1-MAY-2008') AND to_date('21-SEP-2008') ) )and mode=0; After seeing above query I decide to make range partition on column created_date both in summary table and forms table. Below is the list of procedur
Read more

Copy files between Unix and Windows with rcp

There are many third party tools by which you can copy files between windows and unix machine. Some common tools are winscp, SSHSecureShellClient etc. Instead of using any third party tool, you can use original unix and windows built-in tool to do the task. The RCP (Remote CoPy) is a standard UNIX tool and a Microsoft Windows 2000 tool that copies files/folder to and from computer running the RCP service. Before you can use the RCP tool from a Windows-based computer you must do something in your UNIX computer. you must turn on the RCP service and update the security files to allow the Windows-based computer to connect to your unix machine over this service. The following steps will help you. Step 01: Turn on RCP service on UNIX Host Turn on the RCP service on unix machine by using inetd: 1. Log on as root. 2. Edit the file /etc/Inetd.conf. 3. Uncomment the lines that start with shell and that start with exec. 4. Save the file. 5. Use ps -ef |grep inetd and determine inetd service p
Read more