Hacker News: azurelake

New comment by azurelake in "Red Hat Woos VMware Shops with OpenShift Virtualization Engine"

azurelake — Thu, 16 Jan 2025 23:53:22 +0000

You have to understand that Broadcom isn't actually Broadcom the chipmaker. It's a private equity firm that used be named Avago Technologies before it bought Broadcom. So squeezing until there's nothing left is the plan.

https://digitstodollars.com/2022/06/15/what-has-broadcom-bec...

New comment by azurelake in "The Evolution of SRE at Google"

azurelake — Fri, 03 Jan 2025 21:50:27 +0000

> Put another way, automated tests don’t go far enough. We need yet another higher layer of abstraction. Computers are better at deciding what tests to run and when, and are also better at interpreting the results.

Sounds like you might be interested in https://antithesis.com/ (no affiliation).

New comment by azurelake in "Just use Postgres"

azurelake — Sat, 17 Aug 2024 11:14:38 +0000

MySQL has a very mature open source HA story with the flavors of group replication, as well as being able to replicate DDL. Not to mention Orchestrator and friends.

As a matter of fact, EnterpriseDB (the largest contributor to Postgres) has a paid multi master offering, so there's anti incentives in place to improve its HA story...

New comment by azurelake in "Just use Postgres"

azurelake — Sat, 17 Aug 2024 11:04:39 +0000

You can also add add GSIs (with their caveats) without any re-work.

New comment by azurelake in "Psilocybin desynchronizes the human brain"

azurelake — Thu, 18 Jul 2024 02:58:42 +0000

Your second sentence is more on the right path (speaking for myself at least). It's about labeling and observing the literal physiological sensation that is occurring in your body and choosing how to react vs. identifying strongly with the current emotion and reacting.

Easier said than done, of course!

New comment by azurelake in "Mongo but on Postgres and with strong consistency benefits"

azurelake — Sat, 13 Jul 2024 19:49:39 +0000

I’d be curious to hear more detail about the FoundationDB data loss issue that you saw? Do you remember what version / what year that you saw it?

New comment by azurelake in "Three ways to think about Go channels"

azurelake — Thu, 27 Jun 2024 01:19:34 +0000

Elixir is in the process of adding a gradual type system right now FWIW.

New comment by azurelake in "PostgreSQL Statistics, Indexes, and Pareto Data Distributions"

azurelake — Wed, 26 Jun 2024 17:28:22 +0000

One of the best articles I've seen on here this year! I didn't know about the existence of `pg_hint_plan`.

New comment by azurelake in "Aphantasia: I can not picture things in my mind"

azurelake — Sun, 23 Jun 2024 16:37:23 +0000

> Define "actual experiences".

People without aphantasia literally count sheep to go to sleep.

New comment by azurelake in "Elixir 1.17 released: set-theoretic types in patterns, durations, OTP 27"

azurelake — Wed, 12 Jun 2024 21:30:19 +0000

Same. I would pay 50 / month for an Elixir plugin that had a comparable amount of polish to RubyMine.

New comment by azurelake in "Go: Sentinel errors and errors.Is() slow your code down by 3000%"

azurelake — Sat, 01 Jun 2024 15:10:38 +0000

My argument against wrapping for backend services is that is:

1. I think that it is preferable to handle the error where it happened instead of at the top of the stack. For a backend service, there's really only three things you want to do with an error: log it, maybe bump some metrics, and return an error code and ID to the client. You have a lot more information available (including a stack trace if desired) if you handle it at this point.

2. By wrapping the error up the call stack, you're building an ad hoc stack trace. Performance wise, this is (probably, haven't measured) a lot better than an actual stack trace, but as you said yourself, the top concern is debug-ability and developer velocity.

3. Wrapping an error doesn't provide just a stack though, you can add values to the error! Except...what does that really buy you vs. just adding the values to your structured logging system going down the stack vs. doing it on the way back up in an ad-hoc way? Those wrapped error values are a lot more difficult to work with in Grafana vs. searching based on fields.

4. If I have a stack trace, structured log fields, and a correlation ID, I personally don't get any value out of messages like ("could not open file), as I can just use the stack trace to go look at exactly what the line of code is doing. You could argue that with good enough wrapping, looking at the code wouldn't even be necessary, but I think that's pretty rare in practice. It also seems like a lot of extra work to spend a minute loading up the code in an IDE.

5. As mentioned in 1), what the client gets is just an error code and trace ID anyways. In fact, we actively don't want the wrapped context to be sent back to the client since it can be a security concern. If that's the case, we need to remove it and log it anyways. Why not just log the information in the first place?

Anyways, curious to hear your thoughts. I used to advocate for wrapping errors, FWIW.

New comment by azurelake in "Autotuner: How to speed up your Rails app"

azurelake — Fri, 26 Apr 2024 16:10:14 +0000

Thank you!

New comment by azurelake in "Autotuner: How to speed up your Rails app"

azurelake — Fri, 26 Apr 2024 02:25:50 +0000

Curious what the gem is?

New comment by azurelake in "Deterministic simulation testing for our entire SaaS"

azurelake — Tue, 12 Mar 2024 22:00:48 +0000

Can’t wait!

New comment by azurelake in "Deterministic simulation testing for our entire SaaS"

azurelake — Tue, 12 Mar 2024 18:31:20 +0000

I’m curious if you’re willing and able to share: Are you using FoundationDB as the data store for Antithesis?

New comment by azurelake in "Apple built iCloud to store billions of databases"

azurelake — Wed, 17 Jan 2024 23:41:46 +0000

Vector clocks give a partial ordering. FDB version stamps give a total ordering by having a single process issue them for the entire cluster. There's a good breakdown here: https://blog.the-pans.com/notes-on-the-foundationdb-paper/

New comment by azurelake in "Database Fundamentals"

azurelake — Fri, 15 Dec 2023 19:48:38 +0000

That's more or less what Kafka is.

New comment by azurelake in "Rethinking serverless with FLAME"

azurelake — Thu, 07 Dec 2023 00:24:56 +0000

Probably not too much to say that’s specific to FLAME. Closures are serializable and can be sent as messages to actors on the BEAM with a few caveats.

From a quick look at the code, this looks the magic line: https://github.com/phoenixframework/flame/blob/main/lib/flam...

New comment by azurelake in "Noticing when an app is only hosted in us-east-1"

azurelake — Thu, 29 Jun 2023 17:24:49 +0000

They do, look up the “Backend for Frontend” pattern.

New comment by azurelake in "Why PostgreSQL High Availability Matters and How to Achieve It"

azurelake — Wed, 14 Jun 2023 17:50:31 +0000

> You can't combine two indexes to filter on two cols; you need one composite index

Can you expand on how you perceive traditional RDMSs to be different for this case?