cube2222

78 ms·

931.
by cube2222 5y ago
> you should just bite the bullet and make everything red I agree! It's also what I've been seeing in i.e. Scala, where everything is converging to a single Monad. And in that case, monads are basically unnecessary. The point
932.
by cube2222 5y ago
I've been working through the Statistical Rethinking[0] course with Julia recently and so far the language has been really intuitive, same for the libraries (Turing, Distributions). And it's nice that you don't have to vector
933.
by cube2222 5y ago
A nice practice is to finish all such nitpicky comments with the strength of your opinion / how much you care. This way, if most of those are "very little" then it's up to the author to decide whether to do it or not, an
934.
by cube2222 5y ago
This looks great! Definitely looks useful and with no status quo solving it well right now. I'd love to see what else people are using right now for this, and what the cons of that are compared to this - the real-time, team-based suppo
935.
by cube2222 5y ago
I'm just going through this journey myself (migrating some Go event handling stuff to Lambda), will definitely have a blog post up in a few months. Lots of pain, but also lots of gain. Some advice: Don't use zip files. There'
936.
by cube2222 5y ago
I've been doing forks of modules and using replace to use them in my code base extensively. I had 0 problems, it works marvelously. If you're doing some kind of global find/replace on fork, then something's definitely no
937.
A Collaborative Hub for Infrastructure as Code(stackoverflow.blog)
4 points by cube2222 5y ago|0 comments
938.
by cube2222 5y ago
It's not new, at least not as far as the theory goes. This model is basically how the standard proof goes that pushdown automata with two stacks are equal in power to a Turing Machine. Here's for example a stackexchange answer fro
939.
by cube2222 5y ago
Spacelift | Remote | Europe | Full-time | Senior Software Engineer | $80-120k We're a VC-funded startup building an automation platform for Infrastructure-as-Code, adding a Policy-as-Code layer above it, in order to make IaC usable in
940.
by cube2222 5y ago
[0]: https://github.com/cube2222/octosql
941.
by cube2222 5y ago
Congrats on the Show HN! I can confirm Infracost is very easy to integrate with various automation as we’ve been able to implement it in Spacelift[0] using their CLI tool and even integrated[1] it into our OPA-based policy system. Great stu
942.
by cube2222 5y ago
Nice one! I like the SQL extension to denote the output format ("TO json"), great idea! It's also interesting that you allow calling python methods on values, really neat! If you're interested in tools like this, also ma
943.
by cube2222 5y ago
Just keep in mind SELECT * and COUNT(*) are only visually similar. The first one needs to read all data if used at the top level and the second one needs no actual data - just the overall row count, in whatever way is most optimal to calcul
944.
by cube2222 5y ago
> Say you stopped using Photoshop years ago but suddenly need to export one of your old documents. You’re now forced to pay rent to Adobe for at least one month to access one of your own files. And Adobe won’t make it easy to unsubscribe
945.
by cube2222 5y ago
I don't really share OP's impressions. I think the move to the subscription model is one of the best things that happened to our industry in the recent years. It's a model which lets you sustainably sell software and keep wor
946.
Twitter Introduces NFT Profile Pictures(theverge.com)
8 points by cube2222 5y ago|3 comments
947.
by cube2222 5y ago
I do think the parent thinks that. But it's simply not true based on my research. COUNT(*) doesn't need any actual data from the rows, so I'm wondering if there's something else. First result in Google confirms that at l
948.
by cube2222 5y ago
I've never said I have null values, I've just used it to explain what performance optimization opportunities COUNT(*) has vs COUNT(id). It would be great if you'd explain why what I've written is not the case or provided
949.
by cube2222 5y ago
That's harsh, could you explain why, please? As far as I understand (and I'm the author) `COUNT(*)` means "count the number of rows, regardless of the values". `COUNT(id)` means "count all non-NULL id's" w
950.
by cube2222 5y ago
I obviously don't know its exact internal architecture, but based on my digging and trying out different PostgreSQL-native features, it looks like it's PostgreSQL instances as query executors with a custom (high-availability) stor
951.
by cube2222 5y ago
Not really. But it's mostly because the support for other features is basically non-existant. It also seems like all development is happening on Aurora Serverless v2 (which I think is in preview state?). The scaling also isn't sea
952.
by cube2222 5y ago
Yep, you could indeed use a partial index with the two relevant conditions to solve this query well. We're using those all over the place as well.
953.
by cube2222 5y ago
To be fair, we're using Aurora Serverless, so I'll have to dig deeper if we can use these kinds of hints. But thanks for the tip, this could be really useful!
954.
by cube2222 5y ago
Thanks for the tip! I didn't know EXISTS can provide such benefits. However, I've tried it out and it resulted in a 10x performance loss. Looking at the query plan, it actually got PostgreSQL to be too smart and undo most of the o
955.
by cube2222 5y ago
That's more custom logic and additional code to maintain though. In this case we're using the Datadog Agent with its custom_queries directive to automatically create metrics based on periodically evaluated SQL queries.
956.
by cube2222 5y ago
Haven't run it, but based on the query plan in the first picture, PostgreSQL already does that part kind of right. It precalculates it for all accounts and then does a three-way hash join.
957.
by cube2222 5y ago
The downside is that a PostgreSQL version update can change internal optimizer rules, which may result in the plan being changed to a less performant one again. In that case it's good to have monitoring in place to catch something like
958.
by cube2222 5y ago
Hey, author here. The article describes a fun debugging and optimization session I’ve had recently. Happy to answer any questions!
959.
Tricking PostgreSQL into using an insane, but faster, query plan(spacelift.io)
259 points by cube2222 5y ago|132 comments
960.
by cube2222 5y ago
Currently we deem it not desirable, so it's alright to me if we price it very high, thus stimulating investment in this area to bring the cost down or even make it profitable.

More ›