This Meetup is past

62 people went

Location image of event venue

Details

The 2nd South Bay PostgreSQL Meetup.

This is not a competitive event to the PG Meetups in SF, but rather aims to provide a platform for anyone in the South Bay who wants to avoid the traffic jams in SF.

We aim to host a Meetup every other month.

Adobe offered to host this Meetup. You need to sign up for this Meetup, only the names on the registration list will be let in by security. There is a maximum number of guests, sign up quickly.

Logistics:

If you want access to Wifi, you MUST send an email to [masked] by April 24th, with the subject: "SBPUG". Credentials will be send to your email address.

Parking is available: Under 321 Park Ave.

Speakers:

CK Tan: Vitesse DB: 100% PostgreSQL, 100 Times Faster for Analytics

Vitesse DB is PostgreSQL for data warehouse. Rapid advances in memory, SSD, and core-count have produced affordable, powerful computers that comfortably fit the working sets of most DW queries on a single machine without resorting to MPP. Unfortunately, processing speed and single-threaded limitations have precluded PostgreSQL from DW applications.

In this talk, we will describe how we made TPCH Q1 run more than 100X faster on a Postgres instance. In particular, how we dramatically increase per-core efficiency of the Executor using LLVM for dynamic query compilation, and how we thread and scale query execution across many cores efficiently, yet still maintain 100% compatibility with PostgreSQL.

Heikki Linnakangas: pg_rewind - resynchronizing servers after failover

pg_rewind is a tool that makes it possible to resynchronize an old master server after a failover, witout having to make a new base backup. In this presentation, I will explain the problems involved with failover and failback, how pg_rewind works, and demonstrate how to use it.

About CK Tan:

I am Founder at Vitesse Data. Prior to that, I worked at Elance-oDesk as Director of Data Services, and at Greenplum as Sr. Architect. Experienced in data systems, and in particular, MPP data warehouse systems dating back to Informix XPS.

About Heikki Linnakangas:

Heikki is a long-time PostgreSQL developer and committer. He has worked on many different parts of the PostgreSQL backend, most recently on WAL logging. He is currently working for Pivotal.