Ten million uniques
Where the saving against a set is enormous
cardinality: 10000000
See what a HyperLogLog costs against an exact set at your cardinality, and what the approximation actually means in absolute terms.
Or drop a file anywhere on this panel. Nothing is uploaded: the analysis runs in this tab.
The answer appears here
Paste on the left and press Calculate. Nothing leaves this tab.
Nothing else to flag.
No formatting problems, and nothing the rules object to. Worth remembering what that covers: this reads the file you pasted, not the account or cluster it will be applied to.
No finding matches that filter.
Real input you can load into the tool above. Each one shows a different thing going wrong, because that is what the tool is for.
Where the saving against a set is enormous
cardinality: 10000000
Where a set is exact and barely larger, so the approximation buys nothing
cardinality: 500
These are the ones that fail silently. The config is accepted, nothing raises an error, and the consequence arrives later.
It is an estimate with a 0.81% standard error. At ten million that is roughly plus or minus eighty thousand.
Instead:Never use it for billing, reconciliation or anything audited.
A set is exact and costs about the same at that scale.
Instead:Switch to a set. The saving only appears at large cardinalities.
Daily HyperLogLogs merge into weekly and monthly with no extra loss, which storing the members cannot do cheaply.
Instead:Store one per period and merge on read.