Hi VIn,
See benchmarks at http://www.h2database.com/html/performance.html  to get an 
idea of maximum statement/seconds ratio,
and http://www.h2database.com/html/performance.html#database_performance_tuning.

See fast import advices at  
http://www.h2database.com/html/performance.html#fast_import.
To get an idea of the upper limit for this operation use an in-memory database 
with the other advices.

These URLs can be (in general) a good start point to experimentation: 

  
|jdbc:h2:~/test;LOG=0;CACHE_SIZE=512000;LOCK_MODE=0;MULTI_THREADED=1;||DB_CLOSE_DELAY=10|
|  
||jdbc:h2:||split:nioMapped:test||;LOG=0;CACHE_SIZE=512000;LOCK_MODE=0;MULTI_THREADED=1||;DB_CLOSE_DELAY=10||||;CACHE_TYPE=TQ|||
|  
jdbc:h2:mem:test;LOG=0;CACHE_SIZE=256000;LOCK_MODE=0;MULTI_THREADED=1||;DB_CLOSE_DELAY=10||||;CACHE_TYPE=TQ|

||You need to use proper JVM's start parameters (like -Xmx ).
|
|Please note that |CREATE TABLE(...) ... AS SELECT ...| is faster than |CREATE 
TABLE(...); INSERT INTO ... SELECT ...|.

Import a CSV file is an operation that don't benefit too much of multi-tasking 
by nature, so I don't think MULTI_THREADED=1 has much impact.

regards,
Dario

El 14/04/11 03:31, Vin escribió:
>  I did run those tests on a desktop (2.3 GHz Quad core, 2 GB RAM, 300
> GB HDD @7200).
>  I got a 70s reading for 2 M records, so I guess it was the hardware.
>
> On Apr 12, 7:29 pm, Dario Fassi <[email protected]> wrote:
>> Let me say that with the mentioned hardware you are getting very good 
>> numbers:
>> 2M records parsed and updated on DB in 176 secs.  ~  11364 tps
>> 10M records in 60 secs.  ~  167000 tps !!!
>> With witch technology are you getting 167000 tps  on a laptop ??
>>
>> El 11/04/11 15:05, Vin escribi :
>>> For loading 2 M rows into the database - it takes around 176 seconds.
>>> (Laptop Specs - 1.6 GHz Intel Pentium M, 2 GB RAM, 80 GB HDD @5400)
>

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/h2-database?hl=en.

Reply via email to