The AWS cost vocabulary, demystified

A 1- or 3-year AWS commitment to a steady level of compute usage ($/hour) in exchange for a lower rate. Flexible across instance family, size, OS, region, and tenancy.

An older AWS commitment model — pre-pay for a specific instance type / region / tenancy / OS for 1 or 3 years in exchange for a discount of up to ~72%.

The most flexible AWS Savings Plan variant — applies to EC2, Fargate, and Lambda usage across any region, instance family, OS, and tenancy.

A spare EC2 capacity instance priced at up to ~90% off on-demand, with the trade-off that AWS can reclaim it with 2 minutes' notice.

A key-value AWS tag that has been activated for cost reporting — once active, it appears as a column in the Cost and Usage Report and Cost Explorer.

The most detailed AWS billing data export — line-item-level usage and cost, written hourly to an S3 bucket as Parquet or CSV.

A statistically significant unexpected spike or drop in AWS cost for a given service / account / region, relative to its recent baseline.

Matching the instance type and size to actual workload demand — avoiding both over-provisioning (paying for idle capacity) and under-provisioning (performance pain).

An AWS resource that is provisioned and billed but doing no real work — common offenders: unattached EBS volumes, idle EC2, unused elastic IPs, abandoned load balancers.

The letter-and-number prefix of an EC2 instance type (m, c, r, t, etc.) that maps to a specific generation, processor, and resource ratio (CPU vs memory).

A tier of S3 storage trading retrieval latency, durability, and minimum-storage-duration for price — Standard, Standard-IA, Intelligent-Tiering, One Zone-IA, Glacier flavors.

An S3 storage class that automatically moves objects between access tiers based on usage patterns — no retrieval fees, small monitoring charge per object.

The performance / cost tiers of Amazon Elastic Block Store — gp3 (default modern general-purpose), gp2 (legacy), io2 (provisioned IOPS), st1 / sc1 (throughput / cold HDD).

A managed AWS service that gives private-subnet instances outbound internet access — and silently accumulates cost on hourly + per-GB-processed charges.

AWS-native dashboard offering cost, security, fault-tolerance, performance, and service-limit checks — some free, most requiring Business or Enterprise Support.