Paralleles Rechnen Seminar

Requirements / organisational issues

• GPU Multisplit
  
• LightScan: Faster Scan Primitive on CUDA Compatible Manycore Processors
  
• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
  
• cuDNN: Efficient Primitives for Deep Learning
  
• Code Refinement of Stencil Codes
  
• SparkBWA: Speeding Up the Alignment of High-Throughput DNA Sequencing Data
  
• Parallel Distributed Memory Construction of Suffix Arrays
  
• Shifted Hamming distance: a fast and accurate SIMD-friendly filter to accelerate alignment verification in read mapping
  
• Breadth-First Graph Traversal on GPUs
  
• High-performance Graph Analytics on Manycore Processors
  
• Julia: A fresh approach to parallel programming
  
• Self-suggested Thema (needs approval by Prof. Schmidt)
  

Additional information

1. Send E-Mail with two preferred topics to Prof. Schmidt by 22th July 2016
   
2. I will then assign a topic to you 
   
3. You are welcome to suggest new topics as well