1 / 15

Porting Atmospheric Forecasting Model to HPC Platforms

John Athanaselis johnathana@mg.uoa.gr http://forecast.uoa.gr. National & Kapodistrian University of Athens  School of Physics Division of Physics of Environment Meteorology Atmospheric Modeling and Weather Forecasting Group. Porting Atmospheric Forecasting Model to HPC Platforms.

zaynah
Télécharger la présentation

Porting Atmospheric Forecasting Model to HPC Platforms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. John Athanaselis johnathana@mg.uoa.gr http://forecast.uoa.gr National & Kapodistrian University of Athens School of Physics Division of Physics of Environment Meteorology Atmospheric Modeling and Weather Forecasting Group Porting Atmospheric Forecasting Model to HPC Platforms

  2. Topics to be discussed • Why atmospheric forecasting models are important • Key characteristics that make forecasting models challenging for HPC • Current technologies we are using and their limitations • Porting forecasting models to GPU accelerated HPC platforms

  3. Atmospheric models Running on operational basis: • SKIRON • ICLAMS/RAMS • CAMx http://forecast.uoa.gr

  4. Atmospheric models

  5. Atmospheric model characteristics • weather forecasting has a real deadline • improving forecasting accuracy requires enormous computational resources • additional physical processes, increasing resolution, reducing the time step and increasing the total time interval

  6. Atmospheric model characteristics Domain -> Number of grid points -> CFL -> Time step

  7. Parallelism Technology Cluster MPI Shared memory OpenMP Multicore clusters MPI+OpenMP High Performance Computing

  8. GPU

  9. GPU

  10. GPU

  11. GPU Advantage: • Fast and Cheap • Energy efficient Disadvantage: • Not every algorithm can have theoretical speedup • Hard to program • No mature industrial/academic standard model

  12. GPU What kind of algorithms run well on this architecture? • Massive parallelism - is needed to effectively use hundreds of thread processors and provide enough slack parallelism for the fast multi-threading to effectively tolerate device memory latency and maximize device memory bandwidth utilization. • Single precision (32-bit) floating point numbers - double precision float are not universally supported on GPUs. There have been efforts to emulate double-precision floating point values on GPUs; however, the speed tradeoff negates any benefit to offloading the computation onto the GPU • Limited synchronization - thread processors within a multi-processor can synchronize quickly enough to enable coordinated vector operations like reductions, but there is virtually no ability to synchronize across multi-processors. • Locality - is needed to enable use of the hardware or user-managed data caches to minimize accesses to device memory.

  13. Success stories John Michalakes stated the following: "...the 5× to 20× increase in WSM5 performance translates into 1.25× - 1.3× increase in total application performance in total application performance (Amdahl’s law limits the total increase to 1.3×). A 1.25× improvement in model performance from a few months effort is rare. Though 1.3× is clearly not enough to support strong scaling, the initial result is still promising. Moving more computation into the GPU will yield equivalent performance from smaller more efficient clusters. Furthermore, planned improvements in GPU speed, host proximity, and programmability will allow WRF and other highly data-parallel weather and climate models to execute almost entirely on the GPU." GPU Acceleration of Numerical Weather Prediction

  14. PGI Compilers & Tools

  15. Some concluding remarks • Exploiting GPU power requires technical skills • We have progress in automated tools and standards • It looks promising

More Related