pgvector HNSW on a small VPS: how much RAM?
Size a pgvector HNSW index before you build it: the byte arithmetic, maintenance_work_mem, ef_search, and what to do when it stops fitting in RAM.
Filtering by topic #memory · clear
Size a pgvector HNSW index before you build it: the byte arithmetic, maintenance_work_mem, ef_search, and what to do when it stops fitting in RAM.
Your PHP-FPM pool size is a division you run on your own box: measure one worker, subtract what the database needs, then read the number that is left.
Most cloud images ship with no swap. Here is when a small VPS needs a swap file, how big to make it, what vm.swappiness trades, and when zram wins.
Headless Chrome is the biggest process on a small VPS. Install Moli, serve CDP on loopback, point an agent at it, and learn where it stops working.
Four different failures look like n8n going offline. Learn to tell the websocket banner, a restart loop, an out of memory kill and a dead schedule apart.
Anthropic publishes no memory token price. Remembered text is replayed as input tokens, so your cost tracks how much you replay and whether it is cached.
Ollama truncates long prompts at a small default context window. Set num_ctx per request or server wide, and size the KV cache RAM before you raise it.
A capped process can still stall your whole VPS. Set MemoryHigh, MemoryMax, CPUQuota and TasksMax on a systemd unit, and read the OOM kill after.
Docker on a VPS is the same engine with less room: RAM runs out, published ports skip UFW, containers stay down after a reboot, and disk fills up.
Memmy gives your AI agents one shared memory store. Build it from source on Ubuntu, run the memory service on port 18960, and keep every note local.
Recall is a fully local plugin that logs each Claude Code session and condenses it into a resumable summary. Set it up on a VPS and measure the saving.
Set memory and CPU limits in Docker Compose so one greedy container cannot take down your VPS: deploy.resources, mem_limit, exit 137, swap, sizing.