Relative significance
Insert events dynamically relative to their neighbours —
ABOVE, BELOW, or
BETWEEN — with no rigid, static importance scale to
maintain.
Finite memory, done right
Hippocampus is a finite storage engine for log retention, audit trails, and context management. Instead of expiring data by age, it continuously weighs significance, recall, and relationships — keeping the highest-value context and gracefully forgetting the noise.
Traditional stores lean on TTLs and FIFO queues, where age alone decides what survives. Hippocampus applies principles of human memory consolidation instead.
Insert events dynamically relative to their neighbours —
ABOVE, BELOW, or
BETWEEN — with no rigid, static importance scale to
maintain.
Querying or accessing a record strengthens its retention weight, protecting high-demand operational data from decay.
Periodic background cycles apply decay models, compact space, and distil clusters of episodic detail into compact semantic summaries.
Embedded or centralised, backed by SQLite, PostgreSQL, or MySQL — with minimum retention floors that hold regardless of storage pressure.
Both run a compressed decay clock, so forgetting, summarisation, and
eviction play out in minutes rather than days. Sign in as
reader-demo, writer-demo, or
admin-demo — the password is the role.
Great Expectations is reloaded daily, then summarised and left to decay. Watch clusters of episodic memories consolidate into semantic summaries as the store forgets the rest.
Open the book consoleA continuous log trickle flows in and is reaped by consolidation and capacity eviction, so an ever-growing stream stays bounded while the most significant events survive.
Open the logs consoleNew sign-ins may take a moment while the demo stacks warm up. Telemetry for both runs in a shared Grafana.
Clone the repo and start the embedded stack — pure-Go SQLite, zero external dependencies.
git clone https://github.com/fastbean-au/hippocampus.git
cd hippocampus
./demo/run.sh