A sharded, infinite-cardinality database built for time-series data.
Cardinality is where time-series databases stop. Add enough unique series — a container ID, a device serial, a player session — and InfluxDB v1 goes from milliseconds to seconds. Your options are dropping the tags you needed, downsampling data you’ll want later, or a pricing tier. HyperbyteDB shards across nodes so cardinality is a capacity question instead, and speaks InfluxDB v1 line protocol and InfluxQL, so switching costs you one URL.
HyperbyteDB
Open source. Sharded for cardinality that doesn’t cap out.
Same /write and /query endpoints, same InfluxQL, same Telegraf and Grafana. Point them at HyperbyteDB and go — no rewrite, no licence conversation.
On a single node today: 1.5M points/sec sustained at sub-second p95. Sharding lands this year.
HyperbyteDB Cloud
The same database, run by us.
Point Telegraf at a URL and we handle the tuning, replication, upgrades and capacity planning. Early access is open to a small number of design partners.
[email protected]
Questions about moving off InfluxDB v1, or about early access? An actual person reads it.