Sanjana,

> This is to remind you of the above mail.

I apologize for not answering right away. I'll cc our mailing list to this 
response.

> I am a second year undergraduate in Department of Computer Science and
> Engineering in Indian Institute of Technology, Kanpur, India.
> 
> I am interested in contributing to your organisation Ste||ar Project
> especially for the project "Implement a Map/Reduce Framework". I have done
> a course on Operating Systems and Computer Organisation and next semester
> I'll be doing one on Computer Architecture. I am proficient in C and C++.

First of all, the work we're doing is all open source, thus you don't have to 
ask anybody in order to participate.

Map/Reduce is a venerable and proven technology which unfortunately has severe 
shortcomings. Google has
recently announced that it moved away from it, replacing it with dataflow 
techniques. Thus I'd suggest we
look into applying dataflow like techniques for replacing Map/Reduce which also 
perfectly aligns with
HPX. Would you be interested in such a topic instead?

The main source for google's work is here: https://cloud.google.com/dataflow/, 
(the SDK is here:
https://github.com/GoogleCloudPlatform/DataflowJavaSDK), a description of their 
programming
model is here: https://cloud.google.com/dataflow/model/programming-model). But 
all of this is
really just one possible way of looking into applying dataflow techniques to 
distributed computing, it
would require some changes to be usable with HPX. 

For some high level introduction to HPX you might want to read these papers: 
http://stellar.cct.lsu.edu/pubs/pgas14.pdf and 
http://stellar.cct.lsu.edu/pubs/executors_espm2_2015.pdf.

HTH
Regards Hartmut
---------------
http://boost-spirit.com
http://stellar.cct.lsu.edu



_______________________________________________
hpx-users mailing list
[email protected]
https://mail.cct.lsu.edu/mailman/listinfo/hpx-users

Reply via email to