Teaching

Paralleles Rechnen Seminar

Instructors: Christian Hundt; Univ-Prof. Dr. Bertil Schmidt
Shortname: 08.079.590
Course No.: 08.079.590
Course Type: Seminar

Requirements / organisational issues

Themenauswahl:

E-Mail mit 2 bevorzugten Themen bis spätestens 8.7.2018 an Prof. Bertil Schmidt schicken.
Ich werde daraufhin Ihnen ein Thema zuweisen
Es ist auch möglich eigenen Themen vorzuschlagen

Voraussetzungen:

Erfolgreiche Teilnahme an PAA oder HPC

Contents

Themen:

1. Quotient Filters: Approximate Membership Queries on the GPU
2. Design Principles for Sparse Matrix Multiplication on the GPU
3. A Dynamic Dictionary Data Structure for the GPU
4. Fast Equi-Join Algorithms on GPUs: Design and Implementation
5. Rethinking SIMD Vectorization for In-Memory Databases
6. Parallel Programming with Pictures is a Snap!
7. A Framework for the Automatic Vectorization of Parallel Sort on x86-based Processors
8. Generic accelerated sequence alignment in Seqan using vectorization and multithreading
9. Highly Ef?cient Compensation-based Parallelism for Wavefront Loops on GPUs
10. Fast algorithms for Convolutional Neural Networks
11. Darwin: A Genomics Co-processor Provides up to 15,000× acceleration on long read assembly
12. A Domain Specific Language for Developing Computational Genomics Applications
13. Kokkos: Enabling Maycore Performance Portability through Polymorphic Memory Access Patterns
14. Isoefficiency in Practice
15. self-suggested topic (requires approval by Prof. Schmidt)