The VIEW is pretty much forced to behave this way.. it's really just
a run-time SELECT which is amended with your additional query
parameters.. what you want is a MERGE table
This allow your collection of identically structured MyISAM tables to
remain distinct, storage-wise, while acting as an aggregate at query
time. This will give you the behaviour you are expecting.
A TEMP table is just that.. it is never visible to any except the
connection that creates it and it disappears when that connection is
broken. They need to be build from scratch within the connection
life-cycle. They have their uses but, I suspect, they have little to
do with the functionality you seek.
- michael dykman
On Tue, Jan 18, 2011 at 12:42 PM, Sairam Krishnamurthy
> I have another problem with the partition. Once I create a partition,
> the first table contains 700 million rows and I started adding new
> values to the second table. And I have a View with the following
> CREATE VIEW view_name AS SELECT * FROM table_1 UNION SELECT * from table_2;
> Now, the problem is, when I query the view, MySql is creating a temp
> table from these two tables and queries the temp table. Because of the
> size of the original tables, the temp table creation is very slow. I
> looked at the processing algorithms for views and I not sure about the
> difference between MERGE, and TEMPTABLE.
> Can you guys please advise on what is the best way to do this?
> On Fri, Jan 14, 2011 at 2:08 PM, Sairam Krishnamurthy
> <kmsram420@stripped> wrote:
>> The type is MyISAM. So i guess size is not a matter for now. I am well over
>> the limit for MyISAM. I will partition the table and check.
>> Sairam Krishnamurthy
>> +1 612 859 8161
>> On 01/14/2011 01:27 AM, Yogesh Kore wrote:
>> What is the table type for Table?
>> Firstly check with queries and index if required. Check if queries using
>> this table can be fine tuned. Check if table getting locked.
>> If size of table is problem and if the table type is innodb check for
>> innodb_file_per_table options. Also have a look for portioning.
>> On Fri, Jan 14, 2011 at 4:18 AM, Sairam Krishnamurthy
>>> I have a very large table. It has about 1 billion rows. Initially
>>> everything was fine. But now the table is a bit slow. Loaded takes a lot of
>>> time. I usually load in chunks of 22 million rows.
>>> Is the size of the table any way related to the performance? I am not sure
>>> about this.
>>> Will splitting the table and having a "view" from multiple table increase
>>> the performance?
>>> Thanks in advance.
>>> Sairam Krishnamurthy
>>> +1 612 859 8161
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe: http://lists.mysql.com/mysql?unsub=1
- michael dykman
May the Source be with you.