BrainStore
In productionStorage engine for security telemetry
The engine underneath BrainSOC, built for the one shape security telemetry always has: written once, never changed, read as questions about a time range.
Why BrainStore exists
General-purpose databases are built to update rows. Security telemetry never updates anything — it arrives, it is never touched again, and it is read as aggregates over windows of time. BrainStore is built for that shape alone, which is why a query over an afternoon does not have to read a year.
It is not a database you install; it is the engine inside BrainSOC, and that is where it runs — holding our live telemetry and taking writes continuously. If it ever becomes something you can deploy on its own, that will be a deliberate decision, and this page will say so rather than leave you to assume it.
In operation
What is running today
The precise scope of its use right now — not what it could do.
- Embedded as the storage engine inside BrainSOC on Braincap infrastructure
- Holding hundreds of gigabytes of live telemetry, taking writes continuously
- Not available as a standalone product — it ships as part of BrainSOC
Capabilities
What it does
Nothing acknowledged before it is safe
Every write lands in a durable log before it is confirmed, and every block carries a checksum verified on the way back out. A power cut costs you the last few milliseconds, not the afternoon.
Storage that pays for itself
Compression is chosen per column against the data actually in it rather than set once globally — which is why long retention on high-volume telemetry stays affordable instead of becoming the reason you shorten it.
Queries that skip what they cannot match
Indexes let a search rule out most of the data before reading any of it. The practical effect: an investigation over a narrow window returns while the analyst is still thinking, not after they have moved on.
It keeps its own house
Compaction, retention expiry, deduplication and backup run in the background. Retention is a policy you set, not a job somebody has to remember.
Roadmap
What comes next
- Replication between sites
- Tiered storage with a long-term archive
Interested in BrainStore?
Tell us about the problem you are trying to solve — an engineer answers, not a form.