Distributed Computing Seminar

In this seminar we will investigate programming paradigms for distributed computing beyond MPI.

People

Roland Leißa, Simon Moll, Sebastian Hack, Richard Membarth, Philipp Slusallek

General Information

When Wednesdays at 16:15 (sine tempore) (AKA 16:00 (cum tempore))
Where Room 401, E1 3

The first meeting will be on 18 April 2018.

Modus Operandi

Participants will be divided into six groups (two persons per group). Each group will implement a distributed conjugate gradient solver which requires a distributed sparse matrix-vector multiplication. Each group will use a different programming technique.

Moreover, each participant will moderate the discussion for one scientific paper. We will have meetings during the semester in which we will discuss two of the papers. The discussion will be managed by the student to whom the paper was assigned. She/He is responsible for giving a short summary on the paper and for structuring the following discussion.

Prior to each meeting, every student has to write a summary (max. 500 words) for each paper. This summary should include open questions and is to be submitted to Roland Leißa two days before the corresponding meeting (23:59). The summaries of all participants will be made available and can be used by the moderator to structure the discussion in the following meeting.

Registration

Closed.

Please remember to register in the HISPOS!

Coding

Please register in our GitLab and email Roland Leißa your GitLab username and desired group name. You will then get access to a GitLab project matching your group's name. You can use this project for coding. Additionally, you will obtain read access to the paper list in our Wiki.

Schedule

Date Speaker Topic Comments
2018-04-18 - Discussion of SPMV first meeting
2018-04-25 - Introduction of the RRZE cluster
2018-05-23 Timo Gühring HPX - A Task Based Programming Model in a Global Address Space
Guruprasad Hegde Kokkos: Enabling manycore performance portability through polymorphic memory access patterns
2018-05-30 Dil Jose Paul UPC++: A PGAS Extension for C++
Julian Wolter X10: An Object-Oriented Approach to Non-Uniform Cluster Computing
2018-06-13 Carl-Lucien Schwan Parallel Programmability and the Chapel Language
Sami Shalayel Caching Puts and Gets in a PGAS Language Runtime
2018-06-20 Jannic Warken Enhanced memory management for scalable MPI intra-node communication on many-core processor
Shrey Sharma Generating efficient data movement code for heterogeneous architectures with distributed-memory
2018-06-27 Dominik Luche Regent: A High-Productivity Programming Language for HPC with Logical Regions
Thorsten Klößner Control Replication: Compiling Implicit Parallelism to Efficient SPMD with Logical Regions
2018-07-04 Fei Chen Dryad: Distributed Data-Parallel Programs from Sequential Building Blocks
- -
2018-07-18 all Project Presentation