derekperkins

8 ms·

31.
by derekperkins 3y ago
Postgres does several things better than MySQL, but tooling isn't among them.
32.
by derekperkins 3y ago
You have been able to do that with DuckDB for 18 months. https://duckdb.org/2022/09/30/postgres-scanner.html Clickhouse supports it too https://clickhouse.com/docs/en/sql-reference&#
33.
by derekperkins 3y ago
At often 10x the latency and significant query compatibility issues. That's a hard pill to swallow
34.
by derekperkins 3y ago
It finally supports them, which is nice. It doesn't support deferred checks to the commit, like postgres, but are otherwise good. I've appreciated using it to enforce json schema at the db level
35.
by derekperkins 3y ago
We do both and it works great
36.
by derekperkins 3y ago
They have since migrated to Vitess
37.
by derekperkins 3y ago
That's clearly not true, and one of the primary design choices Go makes. It trades concise code for more readable code.
38.
by derekperkins 3y ago
Let's call him another name...
39.
by derekperkins 3y ago
I'm surprised Envoy isn't mentioned anywhere
40.
by derekperkins 3y ago
I've never seen that functionality in JetBrains, which has been my main driver for years. I hope it's there
41.
by derekperkins 3y ago
You're probably thinking of Spanner, not BigQuery
42.
by derekperkins 3y ago
I wouldn't classify type assertion as reflection. It's a super fast type check of the object header, not dynamic reflection.
43.
by derekperkins 3y ago
Even Cloudflare isn't building their own datacenters
44.
by derekperkins 3y ago
If you don't need recursive CTEs on sharded databases, they'll work today. We're actively using them
45.
by derekperkins 3y ago
Easily handled by adding a linter. Like you pointed out, that's an easy place to make mistakes, and this regularly catches them for us https://github.com/nishanths/exhaustive
46.
by derekperkins 3y ago
"show me your fat fruit friend" https://youtube.com/shorts/EL4LY1WPEes
47.
by derekperkins 3y ago
GCP also has a hosted vector db https://cloud.google.com/vertex-ai/docs/vector-search/overvi...
48.
by derekperkins 3y ago
Very interested in the details
49.
by derekperkins 3y ago
That's true for any database at that scale
50.
by derekperkins 3y ago
Thanks, and we're very excited!
51.
by derekperkins 3y ago
That PR was created 15 months ago. Any rough idea when we might expect tooltips to be available?
52.
by derekperkins 4y ago
The activation process uses freemius, which is a joke. I've been trying to activate for 2 weeks and still haven't been able to get it working, and their support is a joke. No shade on the plugin itself, just the third party licens
53.
by derekperkins 4y ago
Your argument is that they own it despite courts denying it?
54.
by derekperkins 4y ago
It already does that. Click on the individual commit to only see those changes, then click next to continue. That's the only way I review commits, and it has been available for at least a few years
55.
by derekperkins 4y ago
Setting MaxIdleConns to be the same as MaxOpenConns isn't sufficient? https://pkg.go.dev/database/sql#DB.SetMaxIdleConns Otherwise, I guess you'd have to poll DBStats and run dummy queries to keep the pool fu
56.
by derekperkins 4y ago
If not most
57.
by derekperkins 4y ago
GitHub is using Vitess internally to avoid applications needing to be shard aware
58.
Authzed Joins the CNCF(authzed.com)
5 points by derekperkins 4y ago|0 comments
59.
by derekperkins 4y ago
And Vitess supports shard local foreign key constraints, even if Planetscale disallows them. They don't work with all the cool online ddl tech.
60.
by derekperkins 4y ago
That matches many, if not most most real world workloads. When hotspots happen, you just continue to shard, as they don't have to be equally sized, even if you shard a single warehouse id to a dedicated node and vertically scale it up.

More ›