[GSoC Project Announce] Support Strong APA Scheduling in SMP Configurations

[Darshit Shah]

Hi,

I'd like to introduce myself and the project that I'll be working on 
over the next couple of months with RTEMS under the Google Summer of 
Code program. 

** Project Abstract **
RTEMS exposes a Linux-like Arbitrary Processor Affinity (APA) API that 
allows applications to restrict themselves to being scheduled on only a 
subset of the available processor cores. However, the current priority 
SMP scheduler is not very efficient at this task. Hence, the aim of this 
project is to implement a new scheduling policy that works similar to 
the priority scheduler but includes an efficient way to provide Strong 
APA scheduling using a bipartite graph as described in [1].

This project will help RTEMS gain a better SMP scheduler that requires 
significantly lesser overhead for scheduling. The new algorithm should 
cause a scheduling overhead of O(m^2) where m is the number of 
processors in the system, whereas the existing algorithm causes 
overheads greater than O(n^2) where n is the number of tasks in the 
system.

The proposal which outlines the exact scope and other details of this 
project can be found at	[2].

I would appreciate any suggestions / ideas about this project from your 
side as well. Also, do let us know if you think this will be useful to 
you in the future and how you plan on using it.

[1]: http://www.mpi-sws.org/~bbb/papers/pdf/rtss14f.pdf
[2]: 
https://docs.google.com/document/d/1kYjC3LT9piCXwMP1mTrBKRgYpMpuwVpu-t7OAPOKxF8/edit?usp=sharing

-- 
Thanking You,
Darshit Shah
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.rtems.org/pipermail/users/attachments/20160425/72e80728/attachment.bin>