End-to-End Computation on the GPU

Presented at the SF Data Mining Meetup: https://www.meetup.com/Data-Mining/events/244034591/

Bill Maimone, the VP of Engineering of MapD, will discuss how to create common frameworks for enabling intra-GPU communication, and will explain how these frameworks will enable developers and statistical researchers to accelerate machine learning workflows by allowing systems to interchange data seamlessly.

As VP of Engineering at MapD, Bill Maimone is responsible for leading the engineering team in architecting and delivering the the company’s suite of database, visualization, and GIS solutions. Prior to MapD, Bill was the VP of Engineering at Anaplan and held similar roles with Salesforce and Actian. Bill began his career at Oracle, where he spent two decades, finishing as a VP with over five hundred members of the R&D team across four continents reporting to him. He holds a M.S. and a B.S. in Computer Science and a B.S. in Journalism from MIT.

Transcript

1. 1 End-to-End Computation on the GPU Bill Maimone - VP Engineering MapD
2. 2 2 AGENDA • Value of GPUs • What is MapD • MapD in AWS • Benchmarks • GOAI (GPU Open Analytics Initiative) • Demo • Get Started
3. 3 Life After Moore’s Law 40 Years of Microprocessor Trend Data 102 103 104 105 106 107 Single-threaded perf 1.5X per year 1.1X per year Transistors (thousands)
4. 4 102 103 104 105 106 107 Single-threaded perf 1.5X per year 1.1X per year GPU-Computing perf 1.5X per year 1000X By 2025 Rise of GPU computing
5. 5 Performance Gap Continues to Widen TFLOPS
6. 6 MapD: Software Optimized For The Fastest Hardware + 100x Faster Queries Speed of Thought Visualization MapD Core MapD Immerse A fast, relational, column store database powered by GPUs A visual analytics engine that leverages the speed + rendering capabilities of MapD Core
7. 7 Mark Litwintschik benchmarked MapD vs. major CPU systems on a billion-row taxi data set and found it to be 6x to 12,500x faster than the fastest CPU databases. MapD GPU-Powered Queries are the Fastest MapD Comparative Query Acceleration* System Query 1 Query 2 Query 3 Query 4 BrytlytDB & 2-node p2.16xlarge cluster 36x 47x 25x 12x ClickHouse, Intel Core i5 4670K 49x 58x 32x 25x Redshift, 6-node ds2.8xlarge cluster 74x 24x 14x 6x BigQuery 95x 38x 6x 6x Presto, 50-node n1-standard-4 cluster 190x 75x 61x 41x Amazon Athena 305x 117x 37x 13x Elasticsearch (heavily tuned) 386x 343x n/a n/a Spark 2.1, 11 x m3.xlarge cluster w/ HDFS 485x 153x 119x 169x Presto, 10-node n1-standard-4 cluster 524x 189x 127x 61x Vertica, Intel Core i5 4670K 685x 607x 203x 132x Elasticsearch (lightly tuned) 1,642x 1,194x n/a n/a Presto, 5-node m3.xlarge cluster w/ HDFS 1,667x 735x 388x 159x Presto, 50-node m3.xlarge cluster w/ S3 2,048x 849x 164x 86x PostgreSQL 9.5 & cstore_fdw 7,238x 3,302x 1,424x 722x Spark 1.6, 5-node m3.xlarge cluster w/ S3 12,571x 5,906x 3,758x 1,884x *All speed comparisons are to the “MapD & 8 Nvidia Pascal Titan Xs” benchmark Source: http://tech.marksblogg.com/benchmarks.html
8. 8 AWS Announces P2 Instances MapD Selected as One of Three Example Use Cases
9. 9 MapD AWS GPU Hours
10. 10 Where Does MapD Fit In? Complementing Your Entire Data Ecosystem
11. 11 DGX SYSTEMS CLOUD Servers in Every Shape and Size TESLA The Essential AI Tool for Instant Productivity Everywhere NVIDIA Deep Learning Platform Everywhere, Anywhere
12. 12 GOAI: End-to-end Analytics on the GPU GPU Open Analytics Initiative – Fusing Machine Learning And GPU Analytics
13. 13 GOAI: End-to-end Analytics on the GPU (1) GPU Open Analytics Initiative – Fusing Machine Learning and GPU Analytics CPU Memory GPU ML Pipeline (Before) MapD (GPU) Python (GPU) ML (GPU)
14. 14 GOAI: End-to-end Analytics on the GPU (2) GPU Open Analytics Initiative – Fusing Machine Learning and GPU Analytics GPU VRAM MapD (GPU) Python (GPU) ML (GPU) GPU Data Frame GPU ML Pipeline (After)
15. 15 DEMONSTRATION
16. 16 Get Involved! • GOAI website: gpuopenanalytics.com/ • GitHub: github.com/gpuopenanalytics github.com/mapd/mapd-core github.com/mapd/pymapd • Discussion Groups: groups.google.com/forum/#!forum/gpuopenanalytics https://community.mapd.com/
17. 17 MapD, Now Open Source