I've been putting a lot of thought into the value system since it would be 
creating a lot of relatively small objects. I can see malloc/free being hit 
hard by this mechanism in C++.

Question:

in the case of multiple, in-memory db's with multiple threads accessing them, 
would it be useful to give each 'Session' or 'Database' its own Value object 
cache?


Also, it seems to me (atleast in C++-land) that the 'Value' system to be a good 
candidate to put on a memory pool rather than going through new/delete. Any 
thoughts on that?

-James

On Apr 29, 2010, at 11:24 AM, Thomas Mueller wrote:

> Hi,
> 
>> I'm studying the Value object cache mechanism. I would appear to me that 
>> multiple databases in one process would be sharing the static Value object 
>> cache on multiple threads. What keeps access to this cache serialized?
> 
> Nothing keeps it synchronized currently. A different thread will
> either see the old value, null, or the new value. This shouldn't be a
> problem. Updating references (pointers) in Java is not necessarily
> done in order, but it is atomic (otherwise it would be a security
> problem).
> 
> Regards,
> Thomas
> 
> -- 
> 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.
> 

-- 
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