openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bryan Noll <bwn...@gmail.com>
Subject Re: Data + Query cache
Date Fri, 25 Aug 2006 17:54:14 GMT
inline...

Pinaki Poddar wrote:
> DataCache or L2 cache being a "performance optimization" feature not
> mandated by the spec, makes it a candidate for *not* being on by
> default. 
> a) If the user explictly turns the feature on, it makes him/her aware of
> its advantages as well as behavioral difference e.g. refresh().     
> b) If the user turns it on and observes a performance improvement --
> that makes the benefit of the feature obvious.
> c) It keeps the option of product level differentiation open
>
> On the other hand, if the feature is "on by default" what would be the
> likely reason to turn it off? Some JavaEE portal may then report how
> "default behaviour of OpenJPA is not spec-compliant" :(
>
> On performance front, initial results indicate that OpenJPA is
> significantly better than other alternatives against industry standard
> benchmarks -- so even leaving DataCache off would not give a reason to
> report otherwise.   
>   
Wow... cool.  Do you have or are you planning to make these 'initial 
results' available publicly?  It would be handy for people to have that 
kind of information in their back pocket if they need it.

Thanks...
>
> Pinaki Poddar
> BEA Systems
> 415.402.7317  
>
>
> -----Original Message-----
> From: Marc Logemann [mailto:ml@logemann.org] 
> Sent: Friday, August 25, 2006 10:18 AM
> To: open-jpa-dev@incubator.apache.org
> Subject: Re: Data + Query cache
>
> Hi,
>
> of course my intention wasnt to ask about kodo, but i was confused that
> a feature will be "on" by default that was formerly on sale.
> I will look into the packages to get an idea.
>
> The other question regarding the cache or a method like isCached() was
> to solve the issue that users could get irritated where the results come
> from.
> Without such a method, there is no way to solve the irritation for
> certain users then ;-)
>
> But the other thing is, if you dont turn it on by default, you can be
> sure to see a review that says that hibernate is faster by default on
> some JavaEE portal.
>
> --
> regards
> Marc Logemann
> [blog] http://www.logemann.org
> [busn] http://www.logentis.de
>
>
> Am 25.08.2006 um 18:31 schrieb Patrick Linskey:
>
>   
>> (Performance pack is Kodo parlance, and this is a Kodo-related email.)
>>
>> Actually, we're getting rid of Performance Pack altogether, at long 
>> last, and replacing with a Professional Edition. (I never much liked 
>> the Performance Pack name, given that the other two were editions.) I 
>> don't remember the details of what lands where, but we aren't going to
>>     
>
>   
>> be selling things that are available in OpenJPA.
>>
>> You can get a feeling for what the JDO bindings for Kodo look like by 
>> taking a look at the openjpa-persistence and openjpa-persistence-jdbc 
>> modules in OpenJPA -- you'll notice that the bindings largely add spec
>>     
>
>   
>> behavior to the core kernel. So the features available in the core 
>> kernel are available to all bindings.
>>
>> -Patrick
>>
>> --
>> Patrick Linskey
>> BEA Systems, Inc.
>>
>>     
>>> -----Original Message-----
>>> From: Marc Logemann [mailto:ml@logemann.org]
>>> Sent: Friday, August 25, 2006 7:01 AM
>>> To: open-jpa-dev@incubator.apache.org
>>> Subject: Re: Data + Query cache
>>>
>>> Hi,
>>>
>>> does that mean that the performance pack gets obsolete or how do you 
>>> want to control the cache then?
>>> It would be weird to sell the cache to JDO users while its free for 
>>> JPA users right?
>>>
>>> --
>>> regards
>>> Marc Logemann
>>> [blog] http://www.logemann.org
>>> [busn] http://www.logentis.de
>>>
>>>
>>> Am 24.08.2006 um 22:54 schrieb Abe White:
>>>
>>>       
>>>> How does the list feel about turning the L2 data and query caches on
>>>>         
>
>   
>>>> by default?  Traditionally, Kodo always left them off by default, 
>>>> primarily because they required a "performance pack"
>>>> license to run.
>>>>         
>>>       
>> ______________________________________________________________________
>> _
>> Notice:  This email message, together with any attachments, may 
>> contain
>> information  of  BEA Systems,  Inc.,  its subsidiaries  and   
>> affiliated
>> entities,  that may be confidential,  proprietary,  copyrighted   
>> and/or
>> legally privileged, and is intended solely for the use of the 
>> individual or entity named in this message. If you are not the 
>> intended recipient, and have received this message in error, please 
>> immediately return this by email and then delete it.
>>     
>
> _______________________________________________________________________
> Notice:  This email message, together with any attachments, may contain
> information  of  BEA Systems,  Inc.,  its subsidiaries  and  affiliated
> entities,  that may be confidential,  proprietary,  copyrighted  and/or
> legally privileged, and is intended solely for the use of the individual
> or entity named in this message. If you are not the intended recipient,
> and have received this message in error, please immediately return this
> by email and then delete it.
>
>   

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message