us VS FusionCache

It might be easier to think about it as a stack, rather than a tree. Each element of the stack represents a subtree -- a perfect binary tree. If you ever have two subtrees of height k, you merge them together into one subtree of height k+1. Your stack might already have another subtree of height k+1; if so, you repeat the process, until there's at most one subtree of each height.

This process is isomorphic to binary addition. Worked example: let's start with a single leaf, i.e. a subtree of height 0. Then we "add" another leaf; since we now have a pair of two equally-sized leaves, we merge them into one subtree of height 1. Then we add a third leaf; now this one doesn't have a sibling to merge with, so we just keep it. Now our "stack" contains two subtrees: one of height 1, and one of height 0.

Now the isomorphism: we start with the binary integer 1, i.e. a single bit at index 0. We add another 1 to it, and the 1s "merge" into a single 1 bit at index 1. Then we add another 1, resulting in two 1 bits at different indices: 11. If we add one more bit, we'll get 100; likewise, if we add another leaf to our BNT, we'll get a single subtree of height 2. Thus, the binary representation of the number of leaves "encodes" the structure of the BNT.

This isomorphism allows you to do some neat tricks, like calculating the size of a Merkle proof in 3 asm instructions. There's some code here if that helps: https://github.com/lukechampine/us/blob/master/merkle/stack....

You could also check out section 5.1 of the BLAKE3 paper: https://github.com/BLAKE3-team/BLAKE3-specs/blob/master/blak...

This isn't what I'm asking for - I don't care if it's baked into us, exists as a backend for minio, uses PseudoKV https://github.com/lukechampine/us/issues/67, or whatever the case may be. I see no value in sending any third party my private data in an unencrypted form (uploading to your server, even if over HTTPS, you got my data).

Want an easy to use cache with advanced resiliency features? Look no further than FusionCache. It's built for performance, good refresh rates, better auto-setup, better logs, and more. Congrats to the team on shipping your first major and stable version 🎉 and receiving over 3.8 million downloads.

I wrote an internal nuget package for our team that does similar stuff to your work, although I called mine ICachedRequest. Unlike you I denied myself the enjoyment of exploring a custom caching solution and ended up injecting FusionCache into my mediatr behavior.

The most undervalued library from that list is FusionCache. The rest is either well-known (like FluentAssertions) or pretty specific to the guy's experience (like the WPF stuff).

The instances (using FusionCache for instance) sync over Redis pub/sub.

LazyCache is amazing. Btw I'm using FusionCache and it is good too

If you are in the .NET space I suggest you to take a look at FusionCache. It has cache stampede protection built in, plus some other nice features like a fail-safe mechanism and soft/hard timeouts https://github.com/jodydonetti/ZiggyCreatures.FusionCache