News for hackers. Stuff that matters.

MBkkt

TLDR It's not our choice, but it's meaningful. Because this 5GB is single data segment and literally what you will have in Elastic/etc when you have overall TBs of data. See https://www.elastic.co/docs/deploy-manage/production-guidanc... (single shard is one Lucene index that contains multiple data segments)
Re:We built a new powerful JSON data type for ClickHouse on Thursday October 24, 2024 @02:13AM (#41932455)
It's like saying that postgres was designed for distributed setups, just because there are large postgres installations. We all understand that clickhouse (and postgres) are great databases. But it's strange to call them designed for distributed setups. How about insertion not through a single master? Scalable replication? And a bunch of other important features -- not just the ability to keep independent shards that can be queried in single query
Re:The Code Review Pyramid on Sunday July 09, 2023 @12:59PM (#36656205)
> some reviewers have the need to demonstrate their superior intellect/knowledge/whatever by insisting on minutia.

> Few senior engineers are humble, and consider code reviews as a collaborative effort to deliver the best possible solution.

The difference between these two only that in first case author didn't agree with reviewer and agreed in the second.

The really worst type of review that approve without looking or IDC.

Few weeks ago on my work we spend a week (not all but still a lot of time) to find why sort with libc++ was incorrect. As result we add similar (but a simpler check).

I think always do such checks isn't good (because performance reasons), but with address sanitizer it's sounds very good for me

Re:C++ Asynchronous Framework on Saturday July 30, 2022 @11:29AM (#32287549)
Seastar is absolutely different approach, based on shared nothing architecture. It's not bad, but you cannot easy compare some async multi threaded architecture with it.

Seastar it's something like multi process architecture, where every process doesn't have synchronization, except n spsc queue per process (used to communicate between cores) and have only cooperative multitasking

So it good scales, if you don't have a lot of data to share, and have a very good work load balancer

But commonly you haven't, so go-like approach more and more easy