FPGA: an Easier Path to Parallelism for Postgres?

It’s easier than you think! Accelerating query performance 55x with FPGA
Schedule

Presented by:

Andy ellicott

Andrew Ellicott

Swarm64

Andy Ellicott has spent over 20 years launching and growing adoption of new databases. He's held executive product strategy and developer advocacy roles at Crate.io, Cloudant/IBM, Vertica, VoltDB, and others. He enjoys being on the forefront of DBMS technology because he enjoys working with the early adopters--the software development vanguard who are transforming the way we work and live.

Wleytjpl 400x400

Sebastian Dressler

Swarm64

Sebastian loves to play with different technologies and bring them together to build creative solutions. In his role as Solution Architect at Swarm64 he helps customers to performance-optimize their PostgreSQL and understand how to fit Swarm64 DA into their environments.

Amazon recently announced FPGA acceleration for Redshift, most clouds now offer FPGA-equipped instances, and FPGA prices are dropping.

Is it time for you to consider hardware acceleration for Postgres?

Adding an FPGA processor to your free, open source Postgres server, along with the Swarm64 FPGA extension for Postgres, enhances your database with 100+ additional SQL reader & writer processes working in parallel to speed up queries and ingestion. It’s an easy way to add massive parallelism to Postgres...“scaling inside,” vs. scaling out or up.

In this webinar, we will explain FPGA acceleration for Postgres:

  • How FPGA database acceleration works
  • Why it’s likely to be a hot DBMS topic in 2020
  • How to apply it in your Postgres deployments
  • Demonstrating a reference architecture for FPGA-accelerated Postgres and showing how it increases query performance by 55x
  • Scaling-in before you scale-out to simplify distributed Postgres deployments

If you care about Postgres performance, and especially if you’d like to use standard Postgres for more analytic projects, you’ll learn new techniques to speed up and scale Postgres.

Date:
2020 February 19 11:00 UTC
Duration:
1 h
Room:
Online
Conference:
Postgres Webinar Series
Language:
Track:
Ops
Difficulty:
Medium
Requires Registration:
Yes (Registered: 87)