What does this calculator do?
Log and event platforms — SIEM, observability stacks, audit archives — are sized by daily volume multiplied by retention, then adjusted for compression, replication and expected growth. This calculator does that in the order the deductions actually apply, and lets you start either from an event rate and an average event size or from a known GB-per-day figure. Results are given in both TB and TiB, because storage vendors quote one and operating systems usually display the other. Units follow the standards exactly: kB, MB, GB and TB are powers of ten (1 GB = 10⁹ bytes) while KiB, MiB, GiB and TiB are powers of two (1 GiB = 2³⁰ bytes). The two ladders are never mixed, so "1 GB = 1024 MB" is not treated as a universal rule.
Formula
Daily volume is compressed first, then multiplied out over retention and copies.
dailyBytes = eventsPerSecond × bytesPerEvent × 86400 storedPerDay = dailyBytes × (1 − compression ÷ 100) retained = storedPerDay × retentionDays replicated = retained × replicationFactor total = replicated × (1 + growthBuffer ÷ 100)
How to use it
Choose how you know your volume. In event mode, enter the sustained events per second and the average event size in bytes — a syslog line is often 200–500 bytes, while a verbose JSON event can be several kilobytes. In daily mode, enter the GB per day your pipeline already reports. Then set retention in days, and use compression, replication and growth buffer only with figures you can justify: an index that expands raw data rather than compressing it needs a compression of 0.
Worked example
500 events per second at 500 bytes each: 500 × 500 × 86,400 = 21,600,000,000 bytes, or 21.6 GB per day. Over 90 days of retention with no compression that is 1.944 TB, which stays 1.944 TB with a single copy and no growth buffer — about 1.7681 TiB as an operating system would report it. Add a replication factor of 3 and the requirement becomes 5.832 TB.
Frequently asked questions
Should I use compression for an indexed log platform?
Be careful. Raw compressed archives shrink a great deal, but a search index adds its own structures on top and some platforms end up storing more than the raw volume. Use a figure measured on your own deployment, or leave compression at 0 and size conservatively.
What is the growth buffer for?
Log volume rarely stays flat — new services, more verbose levels and higher traffic all push it up. The buffer adds headroom on top of the calculated requirement so the cluster does not need emergency expansion mid-retention.
Why report both TB and TiB?
A vendor selling 10 TB means 10 × 10¹² bytes. A filesystem showing 10 TB usually means 10 TiB, which is 10 × 2⁴⁰ bytes — about 10% more. Planning in one unit and buying in the other is a common way to end up short.
Sources
- SNIA Dictionary — vendor-neutral definitions of RAID levels, stripe, mirror, parity, IOPS and raw versus usable capacity.
- NIST — Prefixes for binary multiples (IEC 80000-13): a drive sold as 2 TB (2 × 10¹² bytes) is about 1.82 TiB as reported by most operating systems, which is why raw and usable figures are shown in the unit you choose.
- Nominal equal-drive capacity model used here: RAID 0 = N × S, RAID 1 = S, RAID 5 = (N−1) × S, RAID 6 = (N−2) × S, RAID 10 = (N/2) × S. Filesystem overhead, metadata, hot spares, vendor reservations, compression and deduplication are not modelled unless you enter them.
- Reviewed: 2026-08-10
Related calculators
Last updated: 2026-08-10 · v1.0.0 · Report an error