[runtime] Stability of memory mapped file support

[Steve Ash <st...@gmail.com>]

Hello,

I am working on something that could benefit from using large memory mapped
files.  I have had bad experiences with the built in support, and the lack
of deterministic unmapping in the past.  I am interested in the support
offered in Commons Runtime.  Is it complete or stable?  Any production time?


Any thoughts are greatly appreciated.

Thanks!

Steve Ash

Re: [runtime] Stability of memory mapped file support

[Steve Ash <st...@gmail.com>]

Mladen,

Thanks for your response Mladen.  I'll check back in a month or so.  We will
be targeting 1.6 so that won't be a problem.  I love the idea of commons
runtime, and wish I was in a position to offer assistance with grunt work.

Steve

On Sat, Mar 12, 2011 at 12:22 AM, Mladen Turk <mt...@apache.org> wrote:

> On 03/12/2011 05:57 AM, Steve Ash wrote:
>
>> Hello,
>>
>> I am working on something that could benefit from using large memory
>> mapped
>> files.  I have had bad experiences with the built in support, and the lack
>> of deterministic unmapping in the past.  I am interested in the support
>> offered in Commons Runtime.  Is it complete or stable?  Any production
>> time?
>>
>>
> That's my sandbox project, and I'm currently reviewing the complete
> API since it collides/overlaps with Java7 API in some areas.
> Also I'll completely drop the Java5 support, so I wouldn't suggest
> you base anything on the current sandbox code any time soon.
>
> Anyhow, it's still in active development, and the original
> plan for replacing Daemon and Tomcat Native core functionality still
> stands, but since lots changed in JVM area since sandboxed, I
> had to rethink of few concepts, especially about using core Unsafe
> API which offers much higher JNI performance.
>
> Ping me back in a month or just watch the SVN activity, and I'll
> be able to give you more info then.
> Worse case? It might even go to the attic before exiting the sandbox.
> Sometimes I find developing inside ASF so frustrating, that I'm not
> sure I wish to do that again. It's not the same place I remember it
> was a decade ago.
>
>
> Regards
> --
> ^TM
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org
>
>

RE: [runtime] Stability of memory mapped file support

[Gary Gregory <GG...@seagullsoftware.com>]

> -----Original Message-----
> From: Mladen Turk [mailto:mturk@apache.org]
> Sent: Saturday, March 12, 2011 01:23
> To: user@commons.apache.org
> Subject: Re: [runtime] Stability of memory mapped file support
> 
> On 03/12/2011 05:57 AM, Steve Ash wrote:
> > Hello,
> >
> > I am working on something that could benefit from using large memory mapped
> > files.  I have had bad experiences with the built in support, and the lack
> > of deterministic unmapping in the past.  I am interested in the support
> > offered in Commons Runtime.  Is it complete or stable?  Any production time?
> >
> 
> That's my sandbox project, and I'm currently reviewing the complete
> API since it collides/overlaps with Java7 API in some areas.
> Also I'll completely drop the Java5 support, so I wouldn't suggest
> you base anything on the current sandbox code any time soon.
> 
> Anyhow, it's still in active development, and the original
> plan for replacing Daemon and Tomcat Native core functionality still
> stands, but since lots changed in JVM area since sandboxed, I
> had to rethink of few concepts, especially about using core Unsafe
> API which offers much higher JNI performance.
> 
> Ping me back in a month or just watch the SVN activity, and I'll
> be able to give you more info then.
> Worse case? It might even go to the attic before exiting the sandbox.
> Sometimes I find developing inside ASF so frustrating, that I'm not
> sure I wish to do that again. It's not the same place I remember it
> was a decade ago.

Do you have suggestions for improvements?

Gary

> 
> 
> Regards
> --
> ^TM
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org

Re: [runtime] Stability of memory mapped file support

[Mladen Turk <mt...@apache.org>]

On 03/12/2011 05:57 AM, Steve Ash wrote:
> Hello,
>
> I am working on something that could benefit from using large memory mapped
> files.  I have had bad experiences with the built in support, and the lack
> of deterministic unmapping in the past.  I am interested in the support
> offered in Commons Runtime.  Is it complete or stable?  Any production time?
>

That's my sandbox project, and I'm currently reviewing the complete
API since it collides/overlaps with Java7 API in some areas.
Also I'll completely drop the Java5 support, so I wouldn't suggest
you base anything on the current sandbox code any time soon.

Anyhow, it's still in active development, and the original
plan for replacing Daemon and Tomcat Native core functionality still
stands, but since lots changed in JVM area since sandboxed, I
had to rethink of few concepts, especially about using core Unsafe
API which offers much higher JNI performance.

Ping me back in a month or just watch the SVN activity, and I'll
be able to give you more info then.
Worse case? It might even go to the attic before exiting the sandbox.
Sometimes I find developing inside ASF so frustrating, that I'm not
sure I wish to do that again. It's not the same place I remember it
was a decade ago.


Regards
-- 
^TM

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org