Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA...

15
March 25, 2014 Srinivas Reddy, CTO [email protected]

Transcript of Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA...

Page 1: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

March  25,  2014  

Srinivas  Reddy,  CTO  [email protected]

Page 2: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

The Problem  Domain

 Geospatial  Data-centric

  Incredibly large volumes   Small but requiring massively intensive

computations  Computational analysis  Real-time results  Uncompromised Speed and Accuracy

Page 3: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

The Problem Details   Fictional border control example…well, sorta

fictional ;)   Drone based system (Self-guided and Self-

tracking)   Aware of terrain and structures   Producing telemetry and sensor data in real-time   Primary responsibility ○  To remain within its host country’s border ○  To autonomously track heat sources deployed in

their zone   Alert command center upon acquiring heat source

  Real-time processing of data and spatial location of the drone

Page 4: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Current Methods

 Database system approach  Cost metrics - $1.2 M  Not real-time processing   Found drift with more information

Page 5: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Current Methods

 CPU based systems  Cost metrics - $600 k  Could NOT maintain true real-time posture

with volume of data to be processed

Page 6: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Hypothesis  Develop algorithms on NVIDIA GPUs

  Enormous scale GIS operations performed in sub-second times ○  Highly parallelized and deeply pipelined

  Increase efficiency  Reactive computing (Interactive and real-

time)   Enhanced responsiveness

Page 7: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Our Solution  Kernels Implemented

 Great Circle Distance [Haversine] ○  Accurate to ~ 0.3% ○  Light and simple computations

  Vincenty’s Distance ○  Accurate to 0.5mm ○  Slight decrease in speed and more

computationally intensive [Ideal for GPUs]

Page 8: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Our Solution  Kernels Implemented

  Bearing  Midpoint   Equi-rectangular approximation  Destination point-- given distance and

bearing from start point   Intersection of two paths-- given start points

and bearings  Cross-track distance

Page 9: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Architecture

Dell R720   2 CPU @ 2.0

GHz each   8 cores each   64 GB

memory

Page 10: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Architecture

NVIDIA K20x x2   3.52 teraflops of

single precision floating point each

  2496 CUDA cores each

  5 GB GDDR5 memory each

Page 11: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Method  CUDA Algorithm on GPUs

  Scalable parallel programming model   Software environment for parallel computing

Page 12: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Findings  CPU/GPU Comparisons

 CPU 4 processes: 80-98/sec  CPU 8 processes: ~200/sec  GPU 1 total process: ~10,000/sec  GPU 2 total processes: ~20,000/sec  CPU 8 processes/ GPU 2 processes:

~20,400/sec

Page 13: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Implications   Biometrics [Fingerprint, Palm, Face, Iris]

  Real-time processing for denial of access or suspect alerting against national databases

  Big Data Text Analytics   Unstructured text--Word and geo-hashing for

content similarity and recommenders   Predictive analysis based on unstructured text

content   Medical

  Near real-time diagnosis and analytic judgments from DNA sampling or blood work

Page 14: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

Next Steps  Pursue new ideas based on ever-

growing CUDA GIS algorithms   Integrate SRIS GIS CUDA algorithms

with alternate database technologies

Page 15: Real-Time Geospatial Processing with NVIDIA® GPUs and CUDA ...on-demand.gputechconf.com/gtc/2014/presentations/S... · The Problem Details Fictional border control example…well,

www.sriscompany.com [email protected]