We are very proud to announce the names of the 5 contributors this year who will be funded by Google to work on projects for our group through Google’s Summer of Code 2024!

These contributors represent the very best of the many excellent proposals that we had to choose from. For those unfamiliar with the program, the Google Summer of Code brings together ambitious students from around the world with open source developers by giving each mentoring organization funds to hire a set number of participants. Students then write proposals, which they submit to a mentoring organization, in hopes of having their work funded.

Below are the contributors who will be working with the STE||AR Group this summer listed with their mentors and their proposal abstracts.

Project Title: Conflict (Range-Based) Locks

Contributor: Hari Hara Naveen S, Indian Institute of Technology, Madras

Mentors: Panos Syskakis, Mikael Simberg, John Biddiscombe

In some multi-threaded algorithms, resources need to be protected using locks, but the locking mechanism may need to operate on ranges rather than individual items. For instance, imagine a scenario with a large array of N items where one task requires a small continguous subset of items to be locked while another task requires a different continguous subset. In such cases, a range-based locking mechanism is required. We need a templated range-based lock that can be applied to arrays of various types and dimensions. A successful implementation should support locking and unlocking operations on specified ranges of items and can be extended to handle multi-dimensional arrays (2D/3D, etc.) with templates that allow flexibility over dimensions and data types.

Project Title: Rustize HPX

Contributor: Dikshant Gurudutt, International Institute of Information Technology, Hyderabad

Mentors: Hartmut Kaiser, Shreyas Atre

Providing performant HPX functionality written in C++ with Rust APIs to facilitate safety as well as ease of learning HPX. Designing and implementing Rust bindings for HPX, exposing all or parts of the HPX functionality with a Rust API. Implementing rust’s ffi along with libraries like cxx.rs, bindgen to intrope hpx functions to rust.

Project Title: Adapting the Parallel Algorithms in HPX for Usage with Senders and Receivers

Contributor: Tobias Wukovitsch, University of Vienna

Mentors: Hartmut Kaiser, Isidoros Tsaousis-Seiras

HPX supports the C++ senders and receivers facilities, which were, inter alia, specified in the standard proposal P2300. However, not all of HPX’s parallel algorithms can currently be used as sender adapters. This project aims to address this issue by completing the remaining S/R implementations of HPX’s algorithms.

Project Title: Standardize and Visualize HPX Benchmarks

Contributor: Vedant Ramesh Nimje, Veermata Jijabai Technological Institute, Mumbai

Mentors: Giannis Gonidelis, Shreyas Atre

HPX, as a framework designed for high performance computing, has various benchmarks for measuring the performance of its various components, which includes parallel algorithms, its runtime system, etc. But these benchmarks (performance tests) lack a standardized format and a visualization tool that can help in analyzing performance trends over time, in different operating environments. Hence, the goal of this project is to standardize the benchmarks’ output formats within HPX, and to also add integration with an external benchmarking framework, i.e., nanobench. Additionally, a visualization tool will also be developed, which will leverage the standardized formats to display the results of the benchmarks in an intuitive manner. Expected results: (1) A unified format for HPX benchmarking using chosen benchmarking framework. (2) Automating the installation of the chosen benchmarking framework in the HPX build system. (3) A visualization tool, developed using python and matplotlib, to display the results. (4) Integration of this plotting tool with CI/CD pipelines, to track and display performance reductions or improvements

Project Title: Implement hpx::contains and hpx::contains_subrange (std::contains and std::contains_subrange).

Contributor: Zakaria Abdi, University of Bath

Mentors: Panos Syskakis, Isidoros Tsaousis-Seiras

A parallel and sequential implementation of std::contains and std::contains_subrange for hpx. To implement this, I will create a customisation point object, which will take in the parameters of the function and depending on whether or not an execution policy is passed, the CPO will dispatch a call to either the parallel or the sequential implementation of hpx::contains or hpx::contains_subrange which will be implemented within a struct called contains or contains_subrange. Both of these structs will inherit from a base class called algorithm which will take in contains or contains_subrange and a bool as template parameters. Inheriting from algorithm will give access to the call member function which will make a call to either parallel or sequential function via CRTP and type based dispatching.

Nikunj Gupta, a STE||AR Group GSoC student from 2018 and continued collaborator with our group, has received a Thesis Award from his University, IIT Roorkee, based on his research. 

IIT Roorkee allows students to pursue research in foreign universities (Foreign BTP), wherein the student should have 1 supervisor (IIT Roorkee Prof) and 1 co-supervisor, the professor under which the student pursues research. Dr. Hartmut Kaiser introduced Nikunj to Prof. Sanjay Laxmikant V. Kale, at the University of Illinois at Urbana-Champaign for the purposes of this project. Dr. Kale interviewed Nikunj and agreed to provide a research internship from Sept-Dec. Nikunj proposed this research as Foreign BTP to his University and was accepted. 

After completion of his research internship, Nikunj received an award for his project during his University’s Convocation Ceremony held on Sept 11, 2021.

The details of the project, provided by Nikunj, and are below.  Congratulations, Nikunj!  We are proud of the work you do.

Sept-Dec:

I worked with Sanjay on part 1 of my thesis, which was to write abstractions over Charm++ to generate a linear algebra library. The salient features of this library were out-of-order execution, completely asynchronous operations and almost linear distributed scaling. As a future work, I decided to optimize the library (something I’m doing right now as a graduate student here at UIUC) and to generate a frontend language that has a MATLAB like syntax.

Jan-June:

I felt that I could add more to my thesis to make it stronger, so I decided to add my resilience work to it. I contacted Hartmut regarding resiliency work to which he agreed. The work was to implement resilience execution spaces in Kokkos. I created Kokkos executors for HPX to use Kokkos facilities to achieve resilience within HPX and also added resilience execution spaces in Kokkos to achieve resilience within Kokkos. The resilience scheme within Kokkos is an auto-generative scheme that will work with all current and future execution spaces. I also included my past work with resilience in my thesis. So, the work I did back in Summer 2019 and Summer 2020 was included in my thesis too. In Summer 2019, I worked on implementing Local-Only Software Resilience in HPX which I extended to Distributed Software Resilience in Summer 2020. The resilience work has also been published last year at the FTXS workshop in Supercomputing titled “Towards distributed software resilience in asynchronous many-task programming models”. We plan on writing a paper on the Kokkos resilient execution space as well!

In the first week of July, we completed the first evaluation of our Google Summer of Code program. The students have provided summaries of their work and details of the pull requests they’ve created. Check them out below:

Akhil Nair

My GSoC work mainly involves targeting the following three issues :-

• Issue #4822 (Adapt parallel algorithms to C++20)
• Issue #1668 (Adapt all parallel algorithms to Ranges TS)
• Issue #3646 (Parallel algorithms should accept iterator/sentinel pairs)

I’ll be adapting the remaining algorithms in these issues so that they conform to the C++20 standard.

The project involves adding tag_dispatch and tag_fallback_dispatch CPO (Customization Point Object) overloads. Tag_dispatch overloads are used for the segmented overloads and tag_fallback_dispatch for the parallel overloads as this ensures that it tries to dispatch to a segmented overload even if it’s not an exact match before looking at the parallel overloads.

Range based overloads are also added as well as overloads taking in sentinel values. The base implementations are modified to support these sentinel values. Tests for these overloads as well as any missing tests for the parallel overloads have also been added. Doxygen documentation for each overload and the deprecation of the old hpx::parallel namespace overloads is also done.

The first phase of the GSoC period was great, the community was very supportive and nice. The weekly meetings are really helpful and something to look forward to every week and everyone is helpful and responsive on the irc channel. I am slowly getting to know my mentors as well as my fellow GSoC/GSoD students.

For the first phase of GSoC, I’ve adapted the following algorithms to C++ 20 :-

Merged:

• lexicographical_compare
• uninitialized_move and uninitialized_move_n
• uninitialized_default_construct and uninitialized_default_construct_n
• uninitialized_value_construct and uninitialized_value_construct_n
• unitialized_copy and uninitialized_copy_n
• unitialized_fill and uninitialized_fill_n

Open:

• exclusive_scan
• inclusive_scan
• transform_exclusive_scan
• transform_inclusive_scan

Apart from this, I’ve also created a PR to add the ranges starts_with and ends_with algorithm.

Moving forward, I would be working on adapting the remaining algorithms (not a lot left now) and working on some other issues as well such as adding the shift_left and shift_right algorithms and looking into the performance issues of the scan partitioner.

Srinivas Yadav

Add vectorization to par_unseq implementations of Parallel Algorithms

An Overview of the Project

 Currently the HPX algorithms support data parallelism through explicit vectorization using Vc library and only for few algorithms like for_each, transform and count, but recently the support for Vc library has been deprecated and it is moved to std-experimental-simd. So the aim of the project is to adapt the data-parallel support for parallel algorithms using the new  std-experimental-simd.

Work Done So far

The existing support for data parallelism is tightly entangled with the implementation of the parallel base algorithms. So the first job was to separate the existing datapar algorithms by creating Customization Point Objects (CPOs) using tag_dispatch for datapar execution policies and fallback using tag_fallback_dispatch for other execution policies to the base implementation.

At first I really did not understand how the CPOs work and why they were used in HPX, but later soon I really got into understanding how they work with the help of my mentors and their references to some nice resources for C++ metaprogramming and CRTPs and turns out CPOs was the perfect solution for separating the datapar algorithms. The existing algorithms that support data parallelism rely on two main utility algorithms called util::loop and util::transform. So I had converted these two algorithms to CPOs which successfully separated the datapar algorithms. Here is the link to PR.

This first job really made me understand the existing implementation of the data parallel algorithms and helped me a lot in future in adapting other algorithms. Then I moved on to adapt the std-experimental-simd backend. This task mainly requires implementing four basic traits i.e vector_pack_type, load_store, size and count which can be used to adapt most parallel algorithms. It was fairly simple and I implemented them, but later the CI tests were failing for clang compilers after going through the errors we found out that std-experimental-simd was partially implemented for clang compilers and was only fully  implemented for GCC so then adding extra flags which enable datapar support only with GCC during cmake resolved this issue. Then I was told by one my mentors that there was proposal to C++23 for adding simd support to parallel algorithms which is the one of the main goals of this project, so in the proposal [P0350] the author has proposed the datapar execution policies with the name of simd execution policy, so after discussing with mentors we decided to rename the datapar and dataseq policies to simdpar and simd. Then after completing  this task the existing algorithms support with new simd and simdpar execution policies. Here is the link to PR.

I was playing around with the new std-experimental-simd and was testing out the performance for a few kernels then it turns out the performance when compared to the old Vc backend was really impressive and there were some nice speed ups. I created a new repo for the performance tests which were run on different machines with different architectures.

I was testing the performance only with one metric i.e speed up, but later I was told by my mentors that the roofline model is another way to measure the performance, which I was new to and I was lucky to get an explanation on it from one of my mentors and and understood that  which essentially shows whether the performance that we got out of the newly adapted simd algorithms is reaching CPU theoretical peak performance or not and shows the kernel used is compute bound or memory bound.

The following figure shows speed plot for one the algorithms (for_each using compute bound kernel)

X_axis : Number of elements in array (In powers of 2)

Y_axis : Speed up against sequential execution policy

What’s Next ?

The next step is to implement different compute bound and memory bound kernels and get the roofline plots for them to measure the performance and I still need to clean up the repo I created for the performance benchmarks, after that I will be working on adapting the remaining algorithms here to datapar.

It is time to announce the participants for in the STE||AR Group’s 2021 Google Summer of Code! We are very proud to announce the names of the 2 students this year who will be funded by Google to work on projects for our group.

These recipients represent the very best of the many excellent proposals that we had to choose from. For those unfamiliar with the program, the Google Summer of Code brings together ambitious students from around the world with open source developers by giving each mentoring organization funds to hire a set number of participants. Students then write proposals, which they submit to a mentoring organization, in hopes of having their work funded.

Below are the students who will be working with the STE||AR Group this summer listed with their mentors and their proposal abstracts.

Participant:

Akhil J Nair, Army Institute Of Technology ( Savitribai Phule Pune University)

Mentors:

Giannis Gonidelis

Hartmut Kaiser

Project: Adapting algorithms to C++ 20 and Ranges TS

I’m Akhil J Nair, a third year undergrad studying computer engineering at AIT, Pune. I’ll be working with the STE||AR group this summer, focusing on the algorithms part of HPX, working on tasks such as adapting the parallel algorithms to C++ 20, adding range overloads, sentinel overloads etc. The algorithms would be adapted to use the tag_invoke customization point mechanism and the C++ 20 overloads will be added according to the C++ 20 Standard. The ranges overloads will also be added as proposed in the C++ extensions to ranges. Adding sentinel overloads and separating the segmented algorithms for the few remaining algorithms will also be done. I’m hoping this would serve as an entry point for me to HPX and the wider world of HPC in general and I look forward to contributing and learning a lot over the coming months.

Participant:

Srinivas Yadav, Keshav Memorial Institute of Technology, Hyderabad, India

Mentors:

Nikunj Gupta

Patrick Diehl

Project:  Add vectorization to par_unseq implementations of Parallel Algorithms

I am Srinivas Yadav currently pursuing my Bachelors in Computer Science at KMIT, Hyderabad, India. I will be working with STE | | AR group for HPX this summer in the area of vectorization for parallel algorithms. Current hpx parallel algorithms do not support vectorization. Vectorization allows the algorithms to use the cpu vector registers and hence performance may be improved by utilising most of the cpu resources and allows us to exploit another level of parallelism. This project aims to implement the support for parallel algorithms with explicit vectorization with new `simd` execution policy by using the c++ experimental simd extensions. I hope contributing to HPX would serve me as a stepping stone to the world of HPC and I am looking forward to learning and contributing more over the coming months.

The STE||AR Group is honored to be selected as one of the 2021 Google Summer of Code (GSoC) mentor organizations! This program, which pays students over the summer to work on open source projects, has been a wonderful experience for students and mentors alike. This is our 7th summer being accepted by the program!

Interested students can find out more about the details of the program on GSoC’s official website. As a mentor organization we have come up with a list of suggested topics for students to work on, however, any student can write a proposal about any topic they are interested. We find that students who engage with us on IRC (#ste||ar on freenode) or via our mailing list hpx-users@stellar-group.org have a better chance of having their proposals accepted and a better understanding of their project scope. Students may also read through our hints for successful proposals.

If you are interested in working with an international team of developers on the cutting edge of C++ parallel, task-based runtime systems please check us out!

Please find attached the final reports of our successful GSoC students::

Giannis Gonidelis, Adapt parallel algorithms to C++20: Final Report, Blog

Sayef Sakin, Time Series Updates for Traveler: Final Report/Blog

Weile Wei, Concurrent Data structure Support: Final Report/Blog

Mahesh Kale, Pip Package for Phylanx: Final Report/Blog

Pranav Gadikar, Domain decomposition, load balancing, and massively parallel solvers for the class of nonlocal models: Final Report/Blog