On 6 Apr, Robert Treat wrote:
> On Tue, 2004-04-06 at 04:19, David Costa wrote:
>>
>> On Apr 6, 2004, at 1:02 AM, scott.marlowe wrote:
>>
>> >
>> > I think that if we are going to benchmark the two, and I'm all for it,
>> > we
>> > should benchmark them running something fairly standard, like the OSDL
>> > tests. And MySQL should be using innodb tables for everything, since
>> > postgresql pretty much always carries the load of running mvcc and
>> > write
>> > ahead logging and transactions for everything it does.
>> >
>>
>> Let's do it ;)
>>
>
> One potential problem... I don't think that my$ql can actually handle
> the osdl tests. On the bright side iirc the osdl folks switched to
> studying postgresql instead of sapdb (aka max$ql) because they could not
> get the performance they needed from sapdb, so you might be able to do a
> comparison there. But this has always been a problem with benchmarking
> my$ql with postgresql... any test complex enough to show off
> postgresql's capabilities tends to cause my$ql to fall over... not
> saying you shouldn't do it, just saying finding a good benchmark might
> be tricky.
>
> BTW David, I'm sure you've seen
> http://www.phpbuilder.com/columns/tim20001112.php3?page=1 right?
One of the most important reasons for switching to PostgreSQL was that,
for whatever reason, the SAPDB folks didn't really offer the support we
were hoping for.
Peter Zaitsev has told me they are porting DBT-2 to work with MySQL, but
I haven't heard from him in a while. I don't believe it's too hard to
port it, but we don't have the time here. If someone here wanted to
give it a shot, I can definitely aid what needs to be done.
A quick search on google shows it might be done:
http://seclists.org/lists/linux-kernel/2004/Mar/0818.html
Mark
