Anyone know this paper?

[Joel Sherrill <joel@OARcorp.com>]

gregory.menke at gsfc.nasa.gov wrote:
> "Joel Sherrill <joel at OARcorp.com>" <joel.sherrill at OARcorp.com> writes:
>  > http://vlsi.ee.duth.gr/amdrel/papers/date04.pdf
>  > 
>  > It is about dynmic memory allocation but I think they
>  > missed the boat on the RTEMS reference.  They appear
>  > to think that the RTEMS region/heap manager is based
>  > on a paper from 2001 which also used the term region.
>  > 
>  > The RTEMS heap design dates back to the earliest days
>  > of RTEMS and certainly predates a paper from 2001.
>  > 
> 
> I think they missed a lot of boats.  Its a pretty superficial
> treatment of memory allocation.  

Glad to know it just isn't me being stupid looking at it.

Can you even tell what algorithm/data structure they are proposing?

> The cites here
> 
> http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/b/Bonwick:Jeff.html
> 
> and also referenced
> 
> http://www.os-help.org/slab_allocators-542520-4686-a.html
> 
> treat the issue exhaustively, from userspace all the way down to
> scalability on SMP systems and cache-friendliness.  I think the people
> who did the paper you identified didn't do enough homework ahead of
> time.

Thanks.

I believe that there is no perfect algorithm for all real-time
embedded systems.  It would be great to be able to choose from
a variety of heap, CPU scheduling, disk scheduling and disk
caching algorithms based upon your application requirements.

It seems that no algorithm is perfect for every situation.

--joel