Backup Rotation Planner

Preset:

Pick a policy shape, or Custom to keep your current tier counts.

Daily incremental size:

Enter the average protected change per recovery point; units are MB, GB, or TB.

Full backup size:

Use one protected dataset copy, such as 500 GB or 2 TB.

Daily incrementals: {{ params.dailyRetention }} copies

Enter whole recovery points, for example 7, 14, or 30.

copies

Weekly full backups: {{ params.weeklyRetention }} copies

Enter retained weekly copies, usually 4, 8, or 12.

copies

Monthly full backups: {{ params.monthlyRetention }} copies

Enter retained monthly copies, such as 6, 12, or 24.

copies

Yearly archives: {{ params.yearlyRetention }} copies

Enter yearly full copies; use 0 when the policy has no annual archive.

copies

Monthly change rate: {{ params.changeRatePercent }}%

Slider range: 0% to 50% monthly change.

Full compression ratio: {{ params.fullCompression }}:1

Enter stored-size ratios such as 1.5 for 1.5:1.

Incremental compression: {{ params.incrementalCompression }}:1

Enter ratios such as 2 for 2:1 incremental reduction.

Global dedupe savings: {{ params.dedupePercent }}%

Slider range: 0% to 80%; use 0% for no dedupe credit.

Tier	Frequency	Copies	Per backup	Storage (raw)	Storage (net)	Coverage	Copy
{{ row.tier }}	{{ row.frequency }}	{{ row.copies }}	{{ row.sizePerBackup }}	{{ row.rawStorage }}	{{ row.netStorage }}	{{ row.coverage }}
Total		{{ totalBackups }}		{{ totalStorageRawReadable }}	{{ totalStorageReadable }}	{{ coverageHorizon }}

Signal	Status	Audit note	Copy
{{ row.signal }}	{{ row.status }}	{{ row.detail }}

Configure retention targets to estimate backup storage footprint and coverage.

Embed:

Customize

Include current inputs

Size

Advanced

Width

Height

Aspect ratio

Max height

Collapsible embed

Allow fullscreen

Referrer policy

Sandbox tokens

Backup rotation is the discipline of deciding which recovery points survive after the first few days, weeks, months, and years of backups have passed. The storage bill is only one part of the decision. The same policy also decides how quickly a team can recover from an accidental deletion, a failed migration, a slow data-corruption discovery, or an audit request that asks for an older system state.

Recent restore points usually need to be dense because everyday mistakes are discovered quickly. Older restore points are often kept more sparsely because the goal changes from quick rollback to investigation, compliance, or disaster recovery. A policy with fourteen daily incrementals, several weekly full backups, monthly full backups, and a small set of yearly archives can therefore have a long reach without keeping every daily copy forever.

Full backup: A point-in-time copy that represents the whole protected dataset. Long-term GFS tiers are commonly modeled as full copies.
Incremental backup: A smaller point that stores changed data relative to an earlier backup point or chain state.
Coverage horizon: The farthest modeled retained point, which may be much longer than the dense daily rollback window.

Grandfather-Father-Son retention, usually shortened to GFS, is a practical way to express that taper. The "son" layer keeps frequent recent backups, the "father" layer carries selected weekly or monthly full backups farther back, and the "grandfather" layer preserves the longest archive points. Backup products implement the calendar details differently, but the planning questions stay consistent: how many copies are retained, how large each retained copy becomes, and which tier consumes most of the repository.

Backup rotation planning checks
Planning check	What it protects against
Enough recent points	Routine file mistakes, bad edits, failed updates, and short-lived corruption.
Enough long-term reach	Delayed investigations, audit requests, legal holds, and infrequent disaster-recovery needs.
Defensible size assumptions	Capacity requests that depend too heavily on compression, dedupe, or slow-growth assumptions.

The common mistake is to treat the horizon as the whole story. Five years of yearly archives does not mean five years of daily rollback. It means at least one modeled point reaches five years, while daily and weekly points may expire much earlier. Storage sizing also needs caution because synthetic fulls, active fulls, dependent increments, immutability holds, archive movement, and delayed pruning can make a real repository larger or more complex than a simple tier count suggests.

How to Use This Tool:

Start with measured backup sizes when you have them. The presets are useful policy shapes, but the estimate becomes stronger when the full and incremental sizes match a real protected dataset.

Choose Preset to load a starting policy. GFS 7-4-12-2, GFS 14-8-12-4, Compliance 30-12-7-5, and Cloud snapshot 14-6-6-0 set different tier counts and reduction assumptions. Editing any value changes the policy to Custom.
Enter Daily incremental size and Full backup size, then choose MB, GB, or TB. If the summary still says Configure retention targets, a required size or retained tier is still zero.
Set Daily incrementals, Weekly full backups, Monthly full backups, and Yearly archives. The tier sequence above the fields shows whether the policy is dense near the present and sparse in older history.
Open Advanced for Monthly change rate, Full compression ratio, Incremental compression, and Global dedupe savings. Use conservative values when the repository has not yet proven those savings.
Compare Backup Footprint with Raw, Coverage horizon, total copies, and the dedupe badge. A policy that fits only after a large dedupe credit should also be checked with Global dedupe savings set to 0%.
Review Capacity Breakdown for tier storage, Retention Audit for warnings, Storage Mix Chart for which tier dominates, Retention Horizon Chart for reach, and JSON when the result needs to be copied into planning records.

Interpreting Results:

Backup Footprint is the estimated stored size after compression, growth, and global dedupe savings. Raw is the same modeled policy before dedupe savings are applied. The difference between those two numbers is an assumption to defend, not guaranteed space that every storage platform will return.

How to interpret backup rotation outputs
Output	Read it as	Verify before relying on it
Coverage horizon	The longest modeled reach among the enabled tiers.	Confirm the daily and weekly restore density inside that horizon, not just the oldest point.
Largest tier	The tier consuming the biggest share of net storage.	Check whether that tier matches the recovery, audit, or archive reason for keeping it.
Average copy	Total net storage divided across all retained copies.	Do not use it to size a single full backup, one incremental chain, or a vendor repository block.
Retention Audit	Planning notes for high dedupe, growth, long horizons, dense rollback gaps, or missing GFS bridges.	Use the notes to decide what the backup product, storage tier, and restore tests must prove.

A low net footprint can be false confidence when dedupe is high or the protected data changes quickly. For budgeting, keep the net size, raw size, dedupe percentage, largest tier, and horizon together so reviewers can see which assumption moves the request.

Technical Details:

The model treats daily points as incremental copies and the weekly, monthly, and yearly tiers as full copies. Entered units use binary storage multiples, so 1 GB is 1024 MB and 1 TB is 1024 GB. Copy counts are handled as whole nonnegative counts, zero-copy tiers are omitted, and a result appears only when at least one retained tier has a positive stored size.

Compression is applied to each copy before global dedupe savings. Monthly change rate is modeled as a simple average-age adjustment: a tier covering more days receives more growth credit because its retained points represent a wider period of data change.

Formula Core:

\begin{array}{lcl} S_{t} & = & \frac{E_{t}}{C_{t}} \\ H_{t} & = & n_{t} \times d_{t} \\ G_{t} & = & 1 + \frac{r}{100} \times \frac{H_{t} / 30}{2} \\ N_{t} & = & n_{t} \times S_{t} \times G_{t} \times (1 - \frac{p}{100}) \end{array}

Backup rotation formula symbols
Symbol	Meaning
E_t	Entered size for tier t: incremental size for daily points, full backup size for weekly, monthly, and yearly points.
C_t	Compression ratio for tier t. A 1.5 ratio stores the entered size divided by 1.5.
n_t	Retained copy count for tier t.
d_t	Coverage days per retained copy: 1 for daily, 7 for weekly, 30 for monthly, and 365 for yearly.
r	Monthly change rate percentage, with negative values treated as zero.
p	Global dedupe savings percentage, capped to the 0% to 80% control range.

Tier rules for backup rotation capacity planning
Tier	Copy range	Base size	Coverage rule
Daily incrementals	0 to 60 copies	Daily incremental size after incremental compression.	`copies * 1 day`
Weekly full backups	0 to 52 copies	Full backup size after full compression.	`copies * 7 days`
Monthly full backups	0 to 120 copies	Full backup size after full compression.	`copies * 30 days`
Yearly archives	0 to 30 copies	Full backup size after full compression.	`copies * 365 days`

For example, a 500 GB full backup with a 1.5:1 full compression ratio has a compressed base of about 333.3 GB. Four weekly full backups cover 28 days. With 5% monthly growth, the weekly growth factor is 1 + 0.05 * (28 / 30) / 2, or about 1.023. Before dedupe, that tier is about 4 * 333.3 GB * 1.023, which rounds to 1.3 TB. With 30% dedupe savings, the same tier contributes about 955 GB net.

Displayed storage rounds to a readable binary unit. Coverage is shown as days, weeks, months, or years for scanning, while the tier math uses the day counts above. The charts and tables use the same tier data, so a disagreement between a chart and table usually means the active policy was changed after an earlier reading.

Limitations and Privacy Notes:

Use the estimate as a planning model, not proof that a repository is recoverable. The calculation does not inspect backup catalogs, read repository blocks, upload backup files, or validate a restore chain.

Backup products can handle synthetic fulls, active fulls, immutable copies, archive movement, dependent increments, and deletion delays differently.
Dedupe is applied as one global percentage. Real savings depend on block reuse, encryption, compression order, churn, repository design, and backup software.
Monthly change rate is an average-age model, so fast growth, seasonal load, and one-time migrations need separate review.
Downloaded tables, documents, chart images, and JSON may reveal protected data size and retention policy details. Handle them as operational planning records.

Worked Examples:

Small Office Rollback:

A 350 GB full backup, 45 GB daily incremental size, 10 daily copies, 4 weekly full backups, 6 monthly full backups, no yearly archives, 3% monthly change, 1.5:1 full compression, 2:1 incremental compression, and 25% dedupe produces 20 retained copies. The modeled Backup Footprint is about 2.0 TB net, with 2.6 TB raw and a Coverage horizon of 6.0 months. The monthly tier is the main capacity check because it carries the longest enabled history.

Compliance-Heavy Archive:

Using the Compliance 30-12-7-5 shape with a 1.5 TB full backup and 100 GB daily incremental size creates 54 retained copies. With the preset's 10% monthly change rate and 20% dedupe, the estimate reaches about 38.5 TB net and 48.1 TB raw. The yearly archive is the tier to challenge because five yearly full copies plus growth can outweigh the short-term rollback window.

No Result After Clearing Values:

If all tier counts are zero, or both entered backup sizes are zero, the result area stays at Configure retention targets. Restore a missing size, then set at least one copy count such as 7 daily incrementals or 4 weekly full backups. Once total copies and stored bytes are greater than zero, the summary, tables, audit notes, charts, and JSON update from the same policy.

FAQ:

Does a five-year horizon mean daily restores for five years?

No. Coverage horizon is the farthest modeled retained point. Daily incrementals, weekly full backups, monthly full backups, and yearly archives can each expire on different schedules.

Why can yearly archives dominate storage?

Yearly archives use the full backup size, not the daily incremental size, and their growth factor is based on 365 days per copy. A small number of yearly copies can therefore become the largest tier.

Should I budget from Raw or Backup Footprint?

Use Raw when dedupe savings are unproven. Use Backup Footprint only when the selected dedupe percentage is backed by repository history, vendor sizing, or test data.

Why did the preset change to Custom?

The preset becomes Custom after you edit a size, tier count, growth rate, compression ratio, or dedupe value. That prevents the label from implying that the original preset values are still intact.

Can the estimate replace a restore test?

No. Capacity tables and audit notes help plan storage, but restore testing must still confirm chain integrity, recovery time, permissions, application consistency, and whether older retained points are usable.

Glossary:

GFS: Grandfather-Father-Son retention, a tiered pattern that keeps frequent recent points and selected older full backups.
Full backup: A backup copy representing the whole protected dataset at a point in time.
Incremental backup: A backup copy that stores changed data relative to a previous recovery point or chain state.
Dedupe: Deduplication savings from storing repeated data blocks once across retained copies.
Coverage horizon: The farthest modeled point in the retained policy, expressed as days, weeks, months, or years.

References:

Security Guidelines for Storage Infrastructure, National Institute of Standards and Technology.
Long-Term Retention Policy (GFS), Veeam Backup & Replication User Guide.
Grandfather-Father-Son (GFS) Data Retention, Commvault Documentation.
Grandfather-Father-Son scheme, Acronis Backup & Recovery documentation.

Backup Rotation Planner

Backup Footprint

Plan status

Introduction:

How to Use This Tool:

Interpreting Results:

Technical Details:

Formula Core:

Limitations and Privacy Notes:

Worked Examples:

Small Office Rollback:

Compliance-Heavy Archive:

No Result After Clearing Values:

FAQ:

Glossary:

References: