Skip to main content

Disaster Recovery & Migrations.

info

These were the topics I created flashcards for (Remnote) and would revise them using spaced repetition. The formatting is an export from Remnote.

  • Disaster Recovery
    • RPO stands for?―Recovery Point Objective (Data loss)
    • RTO stands for?―Recovery Time Objective (Downtime)
    • Strategies
      • List the 4 strategies: ↓
        • Backup & Restore
          • restoring from backups
          • longest RTO
          • cheap
        • Pilot Light
          • critical only resources running in standby
          • quicker RTO
          • cheaper
        • Warm Standby
          • full architecture in standby
          • running at minimum levels
          • quick to boost to PROD level
          • costs more...
        • Hot Site (Multi-Site)
          • full prod level site running in stanby
          • most expensive
          • shortest RTO
    • Recovering
      • Backups - enable automatic everywhere, keen schedule regular
      • High Availability - use Route53, RDS Multi-AZ and other systems (Site-to-Site VPN for DX backup) setup so your RTO is minimal.
      • Replication - use services with multi-region setups e.g. Aurora Global, and on-prem to AWS replication with RDS, storage gateway etc... replicate between Regions, Premises etc.
      • Automation ‒ enable things to speed up recovery like CloudFormation & Elastic Beanstalk, recover & reboot on EC2
  • Database Migration Service (DMS)
    • source DB ⇒ DMS (on EC2) ⇒ dest DB
    • Schema Conversion Tool (SCT)
      • when do you need SCT?―when migrating between different DB engines
      • when do you NOT need SCT?―if you're migrating same engine e.g. MySQL to RDS for MySQL
    • RDS MySQL to Aurora MySQL
      • list options to migrate RDS ⇒ Aurora? ↓
        • snapshot RDS ⇒ restore to Aurora
        • Aurora Read Replica using RDS then promote it as its own DB Cluster
    • External MySQL to Aurora MySQL
      • What can you use to create a file backup in S3, and then an Aurora MySQL DB from S3?―Percona XtraBackup
      • if both databases are up & running, use...?―Database Migration Service (DMS)
  • On-Premise to AWS Strategies
    • List known on-prem to AWS Cloud strategies: ↓
      • Download & Run Amazon Linux 2 AMI as VM
      • VM Import & Export
      • AWS Application Directory Service
        • gathers info of on-prem servers and plans migration
      • AWS Database Migration Service
      • AWS Server Migration Service
  • AWS Backup
    • manage and automate backups across AWS Services in a central place.
    • services supported by AWS Backup (hint: E S R D E)? ↓
      • EC2 & EBS
      • S3
      • RDS
      • DocumentDB
      • EFS
      • Storage Gateway
    • Support cross-region backups?―Yes.
    • Support cross-account backups?―Yes.
    • What are backup policies otherwise known as?―Backup Plans.
    • AWS Backup Vault Lock
      • What does WORM stand for?―Write Once Read Memory)
      • Can root user delete backups when vault lock is enabled?―Yes.
  • AWS Application Discovery Service
    • plan migrations by running the ADS agent on on-prem servers and gathering information.
    • two types of Discovery you can run....?― ↓
      • Agent-less Discovery
      • Agent-based Discovery
  • AWS Application Migration Service
    • "lift-and-shift" approach
    • minimum {{downtime}}, reduced {{costs}}.
  • VMWare Cloud on AWS
    • use case?―Customer wants to extend their data centre but keep using their on-prem VMWare setup.
  • Transfer Large Data
    • what are your data transfer options from on-prem to AWS Cloud? ↓
      • Over public internet or Site-to-Site VPN
        • setup = immediate
        • speed = 100Mbps
      • Over private network aka DX
        • setup = over 1 x month
        • speed = 1Gbps
      • Over snowball (hardware)