Why keep buying servers for RAM? Lavik combines NVMe SSD capacity with throughput above Redis on DRAM. Explore the architecture, benchmarks & economics of a 300-to-3 consolidation.

What on earth is running on that, Panantirs spying database ?

Do you mean the use case of a 300-node Redis Cluster?

It’s very common. Since Redis typically runs with 2 vCores per node, a 300-node cluster can only store about 2.4 TB of data.

It’s common in human-centric feature stores and profile stores. But it’s even more common in agent profile stores, since agent profiles are a much larger category.

what is anagent profile store that is deployed at the scale of terabytes? Do you mean, for example, logs of agent sessions that run on providers?

> 2 vCores per node, a 300-node cluster can only store about 2.4 TB of data.

Haven't seen anything more "hyperscaler-brained" in a while.

Do you have comparisons against Dragonfly?

Yes, we compared Lavik with DragonflyDB on flash

https://lavik.dev/en/compare/lavik-vs-dragonfly/

Does anyone read websites like these? AI scrapers maybe?

I'm the author... Codex is co-author but the blog “Migrate Your 300-Node Redis Cluster to Just 3 Lavik Nodes” is reviewed by myself.

I swear I've seen this link a couple of days ago.

Yes, I think it's Hackernews admin help to promote this one.

GitHub link: https://github.com/eloqdata/lavik

The three node setup doesn't make sense to me. Why not offer a user-selectable RAID-like choice instead (or even no redundancy)? Also, I don't like the idea of having any node labeled as Primary.