[Starpu-devel] Questions about StarPU worker recongnization

[Fangli Pi <hpcfapix@hlrs.de>]

Hello,

I am interested in StarPU and now trying to use it in my project. However 
there are some questions about how StarPU do tasks scheduling and allocation:

1. The platform, where we want to use starpu is composed of 3 computation 
nodes (different CPU type and with multiple cores inside), so I want to know 
can StarPU do tasks allocation on different nodes? If yes, are there extra 
inputs need to be given to starpu?

2. From the source code of StarPU, it seems that only task size (eg. data 
length) is considered for performance model (history_based and regression 
based), to predict the expectation time. And do you have some flexibility to 
add other parameters? For example, performance parameters and power 
parameters? I have heard that StarPU would also be extended to energy and 
power prediction. If this is true, it would be very interesting for our 
project now and we would be willing to contribute to this field if needed.

3. About scheduling, there is another thing not clear from the description. 
For example, if "dm" scheduling policy is used, the tasks are pushed to one 
central queue or a queue per worker?
It is written in your documentation that "it schedules tasks where their 
termination time will be minimal", does this mean that all the tasks in the 
queue are compared by their predicted execution time and the one with minimal 
execution time will be popped to the worker and be executed? But is it 
possible to know which worker it is scheduled to, and can we, as a user have 
the possibility to determined which worker to use?

Thanks for providing such helpful tool and looking forward to your reply soon.

Best regards,
Fangli

-- 
High Performance Computing Center (HLRS)   
University of Stuttgart                    
Nobelstr. 19                               
D-70569 Stuttgart, Germany

phone: ++49-711-685-60442