richieartoul

5 ms·

1.
Show HN: Gobuildcache, an OSS remote build cache for Go(github.com)
2 points by richieartoul 9mo ago|0 comments
2.
The case for an iceberg-native database(warpstream.com)
28 points by richieartoul 1y ago|5 comments
3.
by richieartoul 1y ago
It's not a landing page, it's a blog, and if you read the first few sentences of the post it becomes immediately clear what service PlanetScale provides.
4.
How We Resolved a Memory Leak When Pprof Failed Us(warpstream.com)
5 points by richieartoul 2y ago|0 comments
5.
by richieartoul 2y ago
The paper says: “in companion DNA-seq data from this project, no detectable Obelisk reads are found” which I think is getting at what you’re asking, but I’m not sure if my understanding is correct. I’m also not sure if DNA-seq data refers t
6.
The Case for Shared Storage(warpstream.com)
7 points by richieartoul 2y ago|0 comments
7.
Dealing with Rejection (In Distributed Systems)(warpstream.com)
12 points by richieartoul 2y ago|0 comments
8.
The Kafka Metric You're Not Using: Stop Counting Messages, Start Measuring Time(warpstream.com)
6 points by richieartoul 2y ago|0 comments
9.
Zero Disks Is Better (For Kafka)(warpstream.com)
16 points by richieartoul 2y ago|0 comments
10.
by richieartoul 2y ago
Short answer is that MSK has (almost) all the same problems as OSS Kafka. Kinesis streams is a different beast that would require its own blog post. RE:numbers: https://www.warpstream.com/blog/warpstream-benchmarks-and
11.
by richieartoul 2y ago
I talk about this more here: https://www.warpstream.com/blog/s3-express-is-all-you-need RE: comparing to a single-zone Kafka cluster. A lot of people really dislike operating Kafka. Some people don't mind it and t
12.
by richieartoul 2y ago
(WarpStream co-founder here) You don't have to keep the data stored in S3 express one zone forever, you can just land it there and then immediately compact it to S3 standard. You still pay the higher fee to write to S3EOZ, but not the
13.
by richieartoul 2y ago
Is there a word for things that are both hilarious and tragic? I laughed out loud multiple times. Kudos to the author for making such a depressing topic so hysterical.
14.
Cloud Disks Are Expensive(warpstream.com)
5 points by richieartoul 2y ago|0 comments
15.
by richieartoul 3y ago
Storing the data in the customer's S3 bucket is a really good idea IMO. Means they just have to trust you to run the coordination reliably, but not necessarily to keep their data secure which is a much easier bar to clear.
16.
by richieartoul 3y ago
I think I just followed the official recommendations I found (which are probably stale now). I'll update it to r5, but it doesn't really matter. The price difference between the two is like 5%, but hardware only ends up representi
17.
by richieartoul 3y ago
https://www.warpstream.com/blog/rss.xml
18.
by richieartoul 3y ago
Makes sense, very cool. Thanks!
19.
by richieartoul 3y ago
This is pretty cool. I'm a bit of a noob when it comes to stuff like FUSE. There is a bunch of commentary in the blog post about how when DuckDB does X, the Differential Storage implementation does Y. If the differential storage implem
20.
by richieartoul 3y ago
Do you have any more details you can share about the performance of EFS? I've never met anyone who has actually used it in anger.
21.
Unlocking Idempotency with Retroactive Tombstones(warpstream.com)
6 points by richieartoul 3y ago|1 comments
22.
by richieartoul 3y ago
This is awesome and one of the things I’m really excited about with WASM, and specifically Wazero. The Wazero team is top notch. Now someone just needs to do this with zstd and make it fast…
23.
by richieartoul 3y ago
(WarpStream founder) This is more or less exactly what WarpStream is: https://www.warpstream.com/blog/minimizing-s3-api-costs-with... Kafka API, S3 costs and ease of use
24.
by richieartoul 3y ago
You have to do a bit more than that if you want exactly once end-to-end (I.E if Kafka itself can contain duplicates). One of my former colleagues did a good write up on how Husky does it: https://www.datadoghq.com/blog/
25.
Minimizing S3 API costs with distributed MMAP(warpstream.com)
17 points by richieartoul 3y ago|3 comments
26.
Hacking the Kafka Protocol(warpstream.com)
8 points by richieartoul 3y ago|0 comments
27.
by richieartoul 3y ago
It runs fine in AWS, Snowflake and many others run it there. The most recent FoundationDB paper goes into a lot more detail on their recovery protocol, it’s a lot more nuanced than you think, but it works extremely well
28.
by richieartoul 3y ago
Newer versions are moving towards a custom btree storage engine called Redwood
29.
by richieartoul 3y ago
Replied above
30.
by richieartoul 3y ago
Requests to the sequencer are batched heavily. If the sequencer fails, the cluster goes through a recovery and will be unavailable for 2-3 seconds and then recover.

More ›