Read Replica Sizing Calculator

Work out read replica sizing instantly with clear inputs, formula shown and shareable results.

About Read Replica Sizing Calculator

Replicas only help read-heavy workloads, because every write is still applied on every replica — so adding replicas increases total write work rather than reducing it. Sizing starts by subtracting write load from the primary's capacity to find what read capacity remains, then divides the excess across replicas at a safe 70 percent utilisation.

Formula

Replica count

primary read capacity = per-instance capacity - write rate; reads to offload = total reads - primary capacity; replicas = ceil(offload / (capacity x 0.7))

Worked example

With the default values below, Read Replica Sizing Calculator returns Read replicas required of 3. Change any input above and every figure here is recalculated in your browser.

Inputs used

Read queries per secondminimum 0.001: 4,000
Write queries per secondminimum 0.001: 600
Query capacity per instanceminimum 0.001: 1,500
Acceptable replica lagminimum 0.001 s: 5 s

Results

Read replicas required: 3
Reads offloaded to replicas: 3,100 q/s
Read to write ratio: 6.7 :1
Consistency consideration: Lag tolerance allows most reads to be served from replicas

Step by step

1Values used
2Replica count
3Read replicas required= 3
4Reads offloaded to replicas= 3,100 q/s
5Read to write ratio= 6.7 :1
6Consistency consideration= Lag tolerance allows most reads to be served from replicas

Assumptions and limits

The result is derived only from the 4 inputs listed above. Anything not among them is not part of the calculation.
In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
Figures are rounded for display to 1 decimal place, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.

How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.

Frequently Asked Questions

Do replicas help write throughput?

No, they make it slightly worse, since each replica must apply the full write stream. Write scaling requires sharding or a different storage engine.

How is read-after-write consistency handled?

Route reads that must see the just-written value to the primary, or use a session consistency token. Blindly sending all reads to replicas produces confusing missing-data bugs.