Total Cost of Ownership (TCO) when moving from on-premise to cloud, CAPEX vs OPEX, staffing impact of migration, and the AWS migration tooling: Pricing Calculator, Migration Evaluator, VM Import/Export, and Database Migration Service.

Total Cost of Ownership (TCO)

  • TCO = a financial estimate to help buyers/owners determine the direct and indirect costs of a product or service; data center examples: hardware, monitoring, installation, IT personnel, training, software, security, licensing, taxes 11:34:26
  • Creating a TCO is useful when migrating from on-premise to cloud 11:34:49
  • Gartner research case: an organization moved 2,500 VMs to Amazon EC2 — costs (baseline ~$400K) initially went up due to migration cost, then settled at a 55% reduction once migration completed 11:35:13, 11:35:44
  • Iceberg model: visible costs (software license/subscription fees) look similar on-prem vs cloud; TCO means counting the hidden costs below the waterline 11:36:09
    • On-premise hidden costs: implementation, configuration, training, physical security, hardware, IT personnel, maintenance; on AWS only implementation, configuration, and training remain 11:36:30
  • AWS's old TCO calculator promised 75% savings; realistically savings vary by migration strategy (e.g., ~50% over a three-year period with an initial cost spike); after migration the eliminated cost categories become AWS's responsibility 11:36:57, 11:37:23

CAPEX vs OPEX

  • CAPEX (capital expenditure): spend money upfront on physical infrastructure, deduct from taxes over time — the on-premise model; companies often resist cloud because of these tax breaks 11:37:23
    • CAPEX costs: servers, storage, network, backups/archives, disaster recovery, data center, technical personnel — you must guess upfront what you'll spend 11:38:04
  • OPEX (operational expenditure): physical costs shift to the service provider; customer pays only non-physical costs — software leasing/customization, training, cloud support, billing based on usage metrics (compute, storage) 11:38:04
    • With OPEX you can try a product or service without investing in equipment; CAPEX ≈ on-premise thinking, OPEX ≈ cloud/AWS thinking 11:38:34

Does cloud make IT personnel redundant?

  • No — the answer is about shifting responsibilities 11:39:03:
    • IT personnel are needed during the migration phase (the Gartner report showed roughly a year of elevated effort) 11:39:03
    • Roles can transition to cloud roles (e.g., traditional CCNA networking → cloud networking); hybrid approaches always need both traditional and cloud IT teams 11:39:38
    • Exam answer: a company can change employees' activities from managing infrastructure to revenue-generating activities 11:40:01

AWS Pricing Calculator

  • Free cost-estimate tool in the browser, no AWS account needed, at calculator.aws; it replaced the retired TCO calculator 11:40:39
  • Contains 100+ configurable services; to compute TCO, compare your known existing costs against the AWS estimate; export the final estimate as CSV 11:40:39, 11:41:12
  • Demo: per-service quick estimate (ballpark) vs advanced estimate (workload details like daily traffic spikes, baseline/peak) 11:41:42, 11:42:04; estimates can be cloned, edited, exported, or shared via a public link 11:43:28

Migration Evaluator

  • Formerly TSO Logic (company acquired by AWS): estimates an organization's existing on-premise costs to compare against projected AWS costs for a planned migration 11:44:00
  • Uses an agentless collector to gather data from on-premise infrastructure (works with VMware, Microsoft, SQL Server, etc.) 11:44:00, 11:44:38

VM Import/Export

  • Imports virtual machines into EC2; supports VMware, Citrix, Microsoft Hyper-V, and Windows/Linux VHDs from Azure 11:44:38
  • Flow: prepare the VM image → upload it to an S3 bucket → import with the AWS CLI → produces an AMI → launch EC2 from the AMI 11:45:12, 11:45:37

Database Migration Service (DMS)

  • Quickly and securely migrates one database to another — e.g., on-premise database to AWS 11:45:37
  • Architecture: source database → source endpoint → replication instance (an EC2 instance replicating the data) → target endpoint → target database 11:46:03
  • Sources include: Oracle, Microsoft SQL, MySQL, MariaDB, PostgreSQL, MongoDB, SAP ASE, IBM DB2, Azure SQL Database, Amazon RDS, S3, Aurora, DocumentDB 11:46:03; targets add Redshift, DynamoDB, OpenSearch Service, ElastiCache for Redis, Neptune, Apache Kafka — a very flexible service 11:46:30
  • AWS Schema Conversion Tool (SCT): automatically (or semi-automatically) converts a source database schema to a target schema — needed for dissimilar engines (e.g., relational → NoSQL); easy paths like MySQL → PostgreSQL don't need much, and not all source/target/version combinations are possible 11:46:30
  • Instructor experience: DMS migrated a very large database, super fast and not hard to use — remember it for migrations 11:47:42