That's a good starting point. Thanks!
What I would like to suggest is that you reduce the amount of algorithms you propose to implement. The sort related parallel algorithms are non-trivial and require a significant effort in debugging and tuning. I'd rather focus on one or two than on the full list - i.e. partial_sort and stable_sort. Ok, nth_element could be a side product. Please see our implementation of sort (non-stable sort) for an example of the complexities you're looking at. Also, in addition to unit-tests I'd like to see some benchmarks written and run to understand how well the new algorithms perform. Thanks for your interest! Regards Hartmut --------------- <http://stellar.cct.lsu.edu> http://stellar.cct.lsu.edu <https://github.com/STEllAR-GROUP/hpx> https://github.com/STEllAR-GROUP/hpx From: Teng Ma <[email protected]> Sent: Tuesday, March 31, 2020 1:29 AM To: [email protected]; [email protected] Subject: Re: [hpx-users] [GSoC apply] Hi, my proposal is available on the Google Doc (https://docs.google.com/document/d/19NUcYLMAevEm5FicTh3mai96cJHi6jFZknAFLoa P7hc/edit?usp=sharing). Any feedback is welcomed. <https://docs.google.com/document/d/19NUcYLMAevEm5FicTh3mai96cJHi6jFZknAFLoa P7hc/edit?usp=sharing> <https://docs.google.com/document/d/19NUcYLMAevEm5FicTh3mai96cJHi6jFZknAFLoa P7hc/edit?usp=sharing> [GSoC] Implement missing Parallel Algorithms Personal Details Name: Teng Ma College/University: University of Chicago & Tsinghua University Course/Major: Computer Science Degree Program: Phd Email: [email protected] Homepage: https://stmatengss.github.io Github/Bitbucket: https://github.com/stmatengss Availability: How much time do you p... docs.google.com Thank you! Teng Ma _____ From: Hartmut Kaiser <[email protected]> Sent: Monday, March 30, 2020 16:12 To: Teng Ma <[email protected]>; [email protected] <[email protected]> Subject: RE: [hpx-users] [GSoC apply] Teng Ma, Please prepare a proposal and submit it to Google, then. Feel free to send it to us beforehand if you would like to have some feedback. Thanks! Regards Hartmut --------------- <http://stellar.cct.lsu.edu> http://stellar.cct.lsu.edu <https://github.com/STEllAR-GROUP/hpx> https://github.com/STEllAR-GROUP/hpx From: Teng Ma <[email protected]> Sent: Monday, March 30, 2020 3:32 PM To: [email protected]; [email protected] Subject: Re: [hpx-users] [GSoC apply] Yes, I am very familiar with sorting related algorithms. I know almost sorting algorithms like bitonic sort, k-way merge sort, etc and have tried to implement several ones. I have designed a distributed sorting algorithm with one-sided RDMA (https://github.com/stmatengss/M-RDMA/blob/master/src/app/distributed_join.h pp). Sorry for forgetting to mention my previous GSoC project. Because I don't have enough abilities and time two years ago. I pay attention to the HPX project in these two years. _____ From: Hartmut Kaiser <[email protected] <mailto:[email protected]> > Sent: Monday, March 30, 2020 8:56 To: [email protected] <mailto:[email protected]> <[email protected] <mailto:[email protected]> > Cc: Teng Ma <[email protected] <mailto:[email protected]> > Subject: RE: [hpx-users] [GSoC apply] Hey Teng Ma, Thanks for your interest in HPX. Working on the missing HPX parallel algorithms is definitely an option for you. We don't have too many of those left to implement, but the remaining ones are quite hard (mostly related to sorting). Have you some experience with parallelizing sorting related algorithms? Thanks! Regards Hartmut --------------- <http://stellar.cct.lsu.edu> http://stellar.cct.lsu.edu <https://github.com/STEllAR-GROUP/hpx> https://github.com/STEllAR-GROUP/hpx From: [email protected] <mailto:[email protected]> <[email protected] <mailto:[email protected]> > On Behalf Of Teng Ma Sent: Monday, March 30, 2020 12:52 AM To: [email protected] <mailto:[email protected]> Subject: [hpx-users] [GSoC apply] Dear all, I see the projects on the website of Google Summer of Code, and I am interested in the project "implementing missing parallel lgorithms" and "Create Generic Histogram Performance Counter". Currently, I am a fourth-year Ph.D. student at the University of Chicago. My research interests include distributed data structure, remote direct memory access (RDMA), and key-value store systems. I have ever tried to design several kinds of lock-free data structures (e.g., append-only data structure) under one-sided RDMA. On my mind, I love coding and open source, and I make the full survey about hpx because of the need for my research. Besides, I have solid C++ programming experience and know a lot about parallel programming. I am eager to make contributions to the hpx project, and I will appreciate it if you give me the chance to join this project. Can anyone give me some suggestions? Thank you! Best Wishes -- Teng Ma Tel.: 1-312-826-1481 Web: <http://stmatengss.github.io/> http://stmatengss.github.io
_______________________________________________ hpx-users mailing list [email protected] https://mail.cct.lsu.edu/mailman/listinfo/hpx-users
