昵称登录后显示

Data center network communication features for large data applications architecture research
Research Projects:
 Hadoop、Spark and Storm big data processing platform benchmark survey.
 Using profiling tools and algorithm analysis, extract microBenchmark (ex. sorting, grep, shuffle), obtain microBenchmark call graph.
 Foreach microBenchmark, analysis communication features, structure communication model and matching test.
 According to the microBenchmark call graph, combine communication models, fitting each benchmark.
 Blend Benchmarks, build DataCenter network communication simulator.
 Deploy communication simulator on interconnection simulator, evaluate new network topology.
Main Responsibility:
 For Hadoop, Spark and Storm. Finish benchmark analysis and profiling.
 Extract microBenchmark, using tcpdump measure network flow.
 Pitching pile on Hadoop(/Spark/Storm), achieve stage and data size of network production.
 Set up Benchmarks’ module.
 Run communication simulator on interconnection simulator.

GeoEast system-a typical processing module performance optimization
Research Projects:
 PetroChina geological data prestack noise algorithm design.
 Multicore optimization on E5-2658A;
 Coprocessor optimization on Intel Xeon Phi.
Main Responsibility:
 Fortran algorithm realization, c language release;
 Carry out 11.3X speed-up on CPU.
 Increase 26.5X speed-up on MIC

Deconvolution microscope performance optimization
Research Projects:
 The Brain cell image pretreatment and convolution filtering.
 Multicore optimization on E5-2658A.
 Coprocessor optimization on Intel Xeon Phi.
Main Responsibility:
 Matlab algorithm realization, c language release;
 Carry out 35.7X speed-up on CPU.
 Increase 42.9X speed-up on MIC