Agreed. Most plausible reason they "can't remember" the good solution is because they were vibe coding and didn't really understand what they were doing. Research mode my ass.

What's it like compared to Redis or even KeyDB?

Compared to Redis, TigerBeetle has strong durability, and an LSM storage engine to page incremental updates to disk, so it can support 10+ TiB data sets without running into snapshot stalls or OOM.

If it's gonna fit on one machine I'm picking Postgres, Mongo or Redis before TigerBeetle and let's be honest it's not the difference between the 4 that will make it not fit on one machine.

And to be fair, those are all great general purpose solutions. For example, you could also use Postgres or Redis as queue, instead of say Redpanda or Kafka.

But our customers need separation of concerns in their architecture. While they could put the cash in the general purpose filing cabinet, they actually want the separation of concerns between OLGP system of reference (string database, i.e. PG) in the control plane, and OLTP system of record (integer/counting database, i.e. TB) in the data plane.

TigerBeetle doesn't shard either.

Yes, sharding would kill performance under contention, which characterizes many OLTP workloads (e.g. top ten bestseller list on Black Friday, super stocks like NVIDIA, the big 4 banks on a switch, PhonePe/Google Pay on UPI etc)

Sharding is if anything a way to reduce the contention.

Transactions typically need to hit the same set of house accounts. For example, fee accounts, omnibus accounts etc.

Or a central bank switch may only have 4 major banks.

Or a system like UPI may have 85% of transactions flowing through 2 hubs. Say 50% through PhonePe. And say 35% through Google Pay.

Good catch.

Well, Search had no chance when the sites also make money from Google ads. Google fucked their Search by creating themselves incentives for bounce rate.

I like the writing style. It is simple and effective.

The author said LLM helps. Let's lynch him!

I love his writing too! I read this post a few days ago and really liked it, so I started going through his older posts. It's no coincidence that his writing is good—he's actively working to improve it: https://evanhahn.com/economist-style-guide-book-takeaways/.

Of all the comments the post received, this one meant the most to me. Thank you.

Honestly the metric which increased the most is the marketing and astroturfing budget of the major players (OpenAI, Anthropic, Google and Deepseek).

Say what you want about Facebook but at least they released their flagship model fully open.

> model fully open.

uh-huh https://www.llama.com/llama4/license/

Amazon SES is $0.10/1k email.

I'm a huge proponent of AWS over other providers but it's also _very_ manual and detailed.

Sometimes it's nice to let someone else do the detail work and pay to get the benefits.

I think it's a proxy between your software and SQLite with a new database API. I guess "multiplexing" is a big word for saying you can someday swap SQLite for something else.

The CLI shows key-value store features.

I don't know if this software has real world savings in performance. I don't think I would ever use this software.

Like if you want your frontend to write to both Sqlite and RocksDB?