sites.fmind.dev

Portfolio 06 · cohort economics

AI dataset storage lifecycle calculator

Age monthly ingestion cohorts through hot and cold tiers so retention, versions, retrieval, operations, and minimum-duration penalties reconcile without averaging away boundaries.

Decision this answers: When should an AI dataset transition tiers, and does the apparent storage saving survive retrieval and early-deletion charges?

Scenario

Number of monthly cohorts to create.

measured[2]

New logical data before versions or replication.

measured[2]

Whole months each cohort remains billable.

measured[1]

Cohort age when the cold rate begins.

measured[3]

Editable provider-native GB-month rate.

published list[1]

Minimum-duration rules still apply.

published list[1]

The complete default scenario is server rendered. Interactions stay in this browser.

Decision summary

Cohort storage

$4192.50exact-derived

Sum of each active cohort in its current tier.

[1]

Operations

$7.20exact-derived

Rounded in native billing units.

[2]

Early-deletion penalty

$0.00exact-derived

Remaining cold minimum duration per deleted cohort.

[3]

Lifecycle total

$4214.70exact-derived

Storage, operations, retrieval, and penalties.

[1][2][3]
Direct-labeled result profile

The tables below are the accessible source of truth; bar lengths never carry identity alone.

Strategy ledger

CandidateTransitionRetentionMinimum durationRestore statementPrice scopeEvidence
S3 Standard to IAMonth 16 monthsTier-specificDocumented class, not guaranteeGB-month + requests + retrievalpublished list[1]
Google Cloud StorageEditable6 monthsStorage-class-specificDocumented class, not guaranteeGB/GiB displayed explicitlypublished list[2]
Azure BlobEditable6 monthsAccess-tier-specificDocumented class, not guaranteeGB-month + operations + retrievalpublished list[3]

Official surface versus this site

This offline workbench

Provider examples explain native meters one strategy at a time. This simulator follows many ingestion cohorts simultaneously and exposes penalties caused by the chosen transition and retention policy.

Sources and methodology

Every result-affecting reference is visible here without JavaScript and is retained in the JSON export.

[1] Amazon S3 pricing
Authority
Amazon Web Services
Native identifier
S3 storage classes
Region
US
Unit
USD/GB-month
Evidence
published list
Price kind
list
As of
2026-08-01
Effective from
2026-08-01
Retrieved
2026-08-01
Freshness
fresh
Normalization
Opening rates are a dated scenario snapshot; class minimums and operations remain separate.

Open the authoritative source

[2] Google Cloud Storage pricing examples
Authority
Google Cloud
Native identifier
Cloud Storage worked examples
Region
US
Unit
GB-month and operations
Evidence
published list
Price kind
list
As of
2026-08-01
Effective from
2026-08-01
Retrieved
2026-08-01
Freshness
fresh
Normalization
Worked-example units are preserved and operations round only at the documented meter.

Open the authoritative source

[3] Azure Blob access tiers
Authority
Microsoft
Native identifier
Blob access tiers
Unit
tier duration
Evidence
exact-derived
As of
2026-08-01
Retrieved
2026-08-01
Freshness
reference
Normalization
Tier behavior and restore classes are presented as documented classes rather than guarantees.

Open the authoritative source

Limits and non-claims

  • The default simulation retains 30000 GB physically at the horizon.
  • Automatic or intelligent tiering is a separate strategy with its own monitoring and transition fees.
  • Restore-time classes are not latency guarantees, and availability or residency is not evaluated.