GSoC 2019 Organization Application
A fundamental goal of the STE||AR Group is to build an international community of developers focused on improving application scalability. Google Summer of Code has provided us with a platform to reach out across the world and interest young developers in solving the challenges faced by an increasingly parallel world. In the past, we have had great success engaging and nurturing students which has produced several long-term, fruitful relationships. We look forward to the opportunity to engage new students who are interested in tackling programming challenges of the future.
6-10
Before the program starts, we encourage applicants to discuss their proposals with their potential mentors. This ensures that there is a clear understanding and interest in the project from both parties. Additionally, our mentors are required to get at least a weekly update of the student's progress and upcoming issues. If the mentor does not hear from the student, the mentor is responsible for contacting the student and inquiring what is leading to the lack of communication. The mentor will then, if necessary, try to work with the student to guarantee the success of the project. At the end the mentors do their best to engage students and provide the most convenient atmosphere for the work on the project.
One key feature we require in a student's application is a project plan. We ask them to lay out milestones and estimate the time it will take to deliver these milestones. This provides a road map for the mentor and the student to gauge their progress. The students are also required to setup a weekly meeting with their mentors where the student reports the progress. It is in these meetings where progress and schedule slippage can be addressed.
In order to engage students with the broader community, we encourage students to interact with their mentors using our organization's standard communication methods (mailing lists and IRC). Through these channels students have access to all of the members of the STE||AR Group. During our last GSoC, we had several discussions in our IRC channel and the mailing list initialed by students asking the community about HPX and challenges that they encountered. In addition to their project related questions, the students participate in community discussions and decisions which include feature requests, design feedback, and code review.
We have discovered that a major factor that influences whether or not a student will stay involved in the community is the personal ties developed during the program. To encourage a tighter connection with the community, we encourage our students to post questions and submit work in the same public forums that the rest of the community uses. This enables other members of our team to interact with the participants and increases the sense of community which is central to open source projects. We have had several students over the last 3 years who have continued development after the end of GSoC, they continue to contribute to the project with Pull Requests to our github project.
Yes
2018, 2017, 2016, 2015, 2014
2008
https://github.com/STEllAR-GROUP
STE||AR Group
Shaping a Scalable Future
Other
C++, Boost, python, javascript
algorithms, parallel computing, machine learning, hpc, boost
https://github.com/STEllAR-GROUP/hpx/wiki/GSoC-2019-Project-Ideas
HPX is shaking up HPC through a unique combination of C++ language development and parallelism research. Join the STE||AR Group this summer to help push the bar on what's possible.
The STE||AR Group is an international team of researchers who understand that a new approach to parallel computation is needed. Our work is crafted around the idea that we need to invent new ways to more efficiently use the resources that we have and use the knowledge that we gain to help guide the creation of the machines of tomorrow. While we develop several software products, the library which is most heavily developed and core to our team is HPX.
HPX (High Performance ParalleX) is a general purpose C++ runtime system for parallel and distributed applications of any scale. It strives to provide a unified programming model which transparently utilizes the available resources to achieve unprecedented levels of scalability.
This library strictly adheres to the C++11 Standard and leverages the
Boost C++ Libraries which makes HPX easy to use, highly optimized, and
very portable. HPX is developed for conventional architectures including
Linux-based systems, Windows, Mac, and the BlueGene/Q, as well as
accelerators such as the Xeon Phi.
We envision HPX as a library which provides services to applications which makes writing efficient, maintainable, and scalable parallel and distributed codes much simpler than current popular paradigms. Currently HPX is used to develop scientific and industrial applications and in the future we hope to expand its influence to include common applications that touch our everyday lives.
If you are looking for a project which incorporates cutting edge HPC research, runtime library development, and C++ standardization check out our ideas page and contact us either through the #ste||ar channel on IRC (Freenode) or via an email at [email protected]
Please see our complete instructions for proposal submission here
HPX, Machine Learning, Parallel Algorithms, Application, Optimization, Library
https://github.com/STEllAR-GROUP/hpx/wiki/HPX-Resource-Guide
https://mail.cct.lsu.edu/mailman/listinfo/hpx-users
N/A