Abdullahi Yunus
Abdullahi Yunus
> can you create some stress tests to see whether union is actually better ? I added a benchmark as the first commit, the following is what I found after...
Thanks @MPins, for taking a look. I've addressed the feedback.
Thanks for taking a look @GustavoStingelin. From the output, we can see that indeed the `union_all` performs better, took less time to plan and execute.
I cherry-picked f9078e552a9a1652d432408f96dd472f8ceedef1 from https://github.com/lightningnetwork/lnd/pull/10356 adding the benchmark for `IsPublicNode` call and below is how it compares: ```bash go test -tags=test_db_sqlite -bench=BenchmarkIsPublicNode -v ``` | Scenario | Iterations (`b.N`) |...
I notice the `itests` are all failing. Let me check to see why...
> The cache only stores public nodes. Once a node is in the cache, it's considered public (until LRU eviction or restart). This makes sense. I will drop the cache...
I noticed that unit tests for `postgres and sqlite` backends are failing. It is expected because `TestNodeIsPublic` assumes that a node can transition from `public` to `private` after its channels...
Nice analysis @ziggie1984, I'll be taking on this. cc: @saubyk 🙏
> Test with realistic graph sizes (80K+ channels) Hey @ziggie1984, curious if there's a tool/script you use to populate the graph with channels for test purposes?
Thanks, @gijswijs, for creating this comprehensive tracker. I've been doing some research/learning on Onion Message (BOLT4) recently. I will keenly follow the progress of this project and hopefully contribute to...