RDS or: How I Learned to Stop Worrying and Love the Cloud
Gavin M. Roy has over 20 years of experience in the Internet industry as a CTO, CIO, VP of Architecture and technology leader. Gavin has been responsible for the technical architecture and operational stability of several large-scale web properties including MeetMe.com. Currently, Gavin is the VP of Architecture at AWeber Communications and is writing a book for Manning Publications entitled RabbitMQ in Depth.
Gavin started his career in Internet development in the mid-90’s developing websites and working as an open-source advocate. Out of his experience in this area, Gavin co-founded ReadySetNet, a Hollywood based web development firm servicing many top-tier corporate and entertainment industry clients.
Gavin is active in several open source projects and communities including Python, PostgreSQL and RabbitMQ. Gavin is the maintainer of pika, the semi-official Python client-library for RabbitMQ. As a contributor to the PostgreSQL, Gavin has been the keynote speaker at pgCon, the annual PostgreSQL developer conference. In addition, Gavin is a repeat speaker at the annual Python Software Foundation run PyCon conference. Gavin is a member of several professional organizations including the PSF, MAAWG and the IEEE.
No video of the event yet, sorry!
At first blush, running PostgreSQL in the cloud can be a scary proposition. When using Amazon Web Services RDS for PostgreSQL, one loses low-level control for the operational management of PostgreSQL. Conversely, when you go in understanding the shortcomings and how to work around them, it can be a very liberating experience. In 2016, I became a cloud convert. Where I was hesitant to embrace Postgres in the cloud, I now endorse the operational outsourcing RDS provides. In this talk I will explain the path that I took to get here. Moving 15 TB of heavy OLTP data to PostgreSQL in AWS was not without pains, problems, and compromises. This talk will provide a realistic view of running PostgreSQL in AWS, detailing design considerations, operational challenges, and the numerous upsides that can be found.
- 50 min
- PGConf US 2017 [PgConf.US]