High Availability Infrastructure with Disaster Recovery | Vadim Sumin

Vadim Sumin, DevOps Engineer at Magecom, spoke at Magento Meetup Online # 11 on November 11 with the topic “High Availability Infrastructure with Disaster Recovery”.

Vadim spoke about high availability infrastructures, their main features, and disaster recovery. He also provided best practices examples for developing high-availability solutions based on AWS.

00:00 Welcoming speech & agenda
00:34 What is not included in the talk
01:20 What is High Availability (HA)
03:03 Basic design principles
09:49 Murphy’s law
10:51 What is a disaster
12:33 High availability is not disaster recovery
14:00 Recovery time objective & recovery point objective
15:05 Core Disaster Recovery strategies
17:26 Backup and Restore
18:47 Pilot light
20:27 Warm Standby
21:30 Multi-site active/active
22:11 Infrastructure as code
25:05 Thank you for your attention
25:20 Q&A session
25:26 Question #1: Did you implement zero-downtime deployment on Magento 2 at release or perhaps with disaster recovery?
27:34 Question #2: Is it true that to ensure horizontal scaling with multiple frontend and application servers, it is not at all necessary to import the backend of a monolithic application to NodeJS?
28:37 Question #3: What kind of disasters did you encounter, and how did you recover from them?
29:56 Question #4: Do you use CloudWatch Alarms to notify you about disasters and attacks? Could you please provide best practices for its setting?
31:02 Question #5: Have you encountered performance degradation in cases of autoscaling? If yes, how did you solve such issues?