a doubt of savepoint

[王 旭]

In maxdb help,there is a part of savepoint document:

As savepoints affect system performance, the database system does not
perform a new savepoint automatically at the end of this time interval. It
waits for 5,000 I/O operations and only then does it check how much time
has elapsed since the last savepoint. This means that the system does not
write a savepoint unless at least 5,000 I/O operations have taken place
since the last savepoint. With a lower number of I/O operations, more time
may elapse until the next savepoint.

And the "_Restart_time" is 600 sec!

Too long to active database writer and log writer.

In oracle,there is a concept named "checkpoint".I don't know whether
"chechpoint = savepoint"

thank!

_________________________________________________________________
与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn

[Hahn, Uwe]

Hello Liu,

MaxDB's savepoint is similar to Oracle's checkpoint.

If no other reason caused a savepoint a savepoint is written if since the last savepoint 2/3 of the logvolume is used.
If no other reason caused a savepoint and 5000 Log writes were done in at least 600 sec (default) a sapoint is triggered.

Kind regards,
Uwe

-----Original Message-----
Sent: Monday, April 18, 2005 1:30 PM
Subject: a doubt of savepoint
As savepoints affect system performance, the database
system does not
perform a new savepoint automatically at the end of this time
interval. It
waits for 5,000 I/O operations and only then does it check how
much time
has elapsed since the last savepoint. This means that the
system does not
write a savepoint unless at least 5,000 I/O operations have
taken place
since the last savepoint. With a lower number of I/O
operations, more time
may elapse until the next savepoint.
And the "_Restart_time" is 600 sec!
Too long to active database writer and log writer.
In oracle,there is a concept named "checkpoint".I don't
know whether
"chechpoint = savepoint"
thank!
_________________________________________________________________
与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn
--
MaxDB Discussion Mailing List
For list archives: http://lists.mysql.com/maxdb

[王 旭]

thanks,Hahn.

But i also have some questions.

One,IS the logvolume in the hard disk or in the memory?

Two, Do the savepoint activate the write process only?

In oracle,there are two important process:DBWR and LGWR.DBWR control
datas write operation from memory to disk and LGWR control logs write
operation from memory to disk.There are many event can trigger DBWR and
LGWR.e.g,when commit,when 1/3 of redo log buffer(in memory) be used,when
checkpoint activation .....

I understand only savepoint can activate the write process in maxdb from
your answer.I don't know if it is right for my understanding.

Where can i find the documents about the read and write mechanism of
maxdb?
thanks.

Subject: RE: a doubt of savepoint
Date: Mon, 18 Apr 2005 14:23:37 +0200
Hello Liu,
MaxDB's savepoint is similar to Oracle's checkpoint.
If no other reason caused a savepoint a savepoint is written if since the

last savepoint 2/3 of the logvolume is used.

If no other reason caused a savepoint and 5000 Log writes were done in at

least 600 sec (default) a sapoint is triggered.

Kind regards,
Uwe

-----Original Message-----
Sent: Monday, April 18, 2005 1:30 PM
Subject: a doubt of savepoint
As savepoints affect system performance, the database
system does not
perform a new savepoint automatically at the end of this time
interval. It
waits for 5,000 I/O operations and only then does it check how
much time
has elapsed since the last savepoint. This means that the
system does not
write a savepoint unless at least 5,000 I/O operations have
taken place
since the last savepoint. With a lower number of I/O
operations, more time
may elapse until the next savepoint.
And the "_Restart_time" is 600 sec!
Too long to active database writer and log writer.
In oracle,there is a concept named "checkpoint".I don't
know whether
"chechpoint = savepoint"
thank!
_________________________________________________________________
与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn
--
MaxDB Discussion Mailing List
For list archives: http://lists.mysql.com/maxdb

_________________________________________________________________
免费下载 MSN Explorer: http://explorer.msn.com/lccn

[Henrik Hempelmann]

Post by 王 旭
But i also have some questions.
One,IS the logvolume in the hard disk or in the memory?

the 'logvolume' is on disk, it's the sum of all your configured
volumes used for logging.

Post by 王 旭
Two, Do the savepoint activate the write process only?

data is written from the data cache to disk during savepoints or
when the data cache is full with changed pages and needs space
for new pages.

Post by 王 旭
Where can i find the documents about the read and write mechanism of
maxdb?

Have a look at the "Concepts of the Database System"
http://dev.mysql.com/doc/maxdb/en/79/44b2565cfb594286e2d1b3bdf7d1d9/frameset.htm
especially "Backup and Recovery"/"Savepoint"
http://dev.mysql.com/doc/maxdb/en/17/6f9b40e8192a54e10000000a1550b0/frameset.htm


henrik