The blog author founded altinity. Altinity's main product offering is a hosted clickhouse service. The top 10 committers to clickhouse all seem to be clickhouse employees. Looking at altinity on github, they contribute much less open source. If clickhouse the company are spending 40%+ of their money to build the product, then others including altinity spend 5% dev and 80% marketing, they will get more customers. That isn't sustainable. How do you solve that? Other than fencing off exclusive enterprise features.
This would be a shame and also a mistake in my opinion.<p>Clickhouse is instantly differentiated from Snowflake, Databricks, BigQuery and RedShift with the open source offering that you can deploy yourself. There are lots of other options but Clickhouse has the most mindshare and is the techies choice.<p>I find myself rooting for them and recommending them for that before you even get into any technical comparison.
As a user of clickhouse since 2018 I'm fully aligned with the content of this article. This technology is one of the best I've been using in my career.<p>The choice of clickhouse for a new project in my company has always been a no-brainer, but the recent move from clickhouse.inc to a closed source version has made this choice less straightforward.
Anyone familiar with Databend, Starrocks, or ByConity? They all focus on shared storage with separate compute. Currently checking out ByConity. Been using Clickhouse for quite a while and these were on my radar
The inherited advantage of being closed source in the first places is that you will not be accused of “moving away” from open source. We never see Microsoft office or Apple MacOS moving away from open source.
Open Source doesn’t pay. Companies need to make money. Any open source product owned and developed by a for profit company is at danger of it moving away from open source.<p>If you want open source go fund non profit organisations and/or charities. The fact we don’t see developers do that tells me a lot.
Here is a similar Thin-Crust Open Core model <a href="https://reactflow.dev/blog/asking-for-money-for-open-source/" rel="nofollow noreferrer">https://reactflow.dev/blog/asking-for-money-for-open-source/</a>. But I hear CH and understand their move. I think OSS Sponsorship has failed; it does not generate enough money to pay a team of top engineers. The best move would have been to implement better paid support model and allow more users to pay for support. Currently CH charges immense amount for their support, so only large corporations could afford.
Clickhouse open source and cloud user. My understanding is that the cloud version uses S3. Which would mean that they have very specific tenant pattern and code to run that version. This may be why lightweight work a specific way in that environment, or they need a way to test it at scale that would be hard through a feature flag in the open source product. Lightweight deletes were released to both, and previous roadmaps listed updates as upcoming.<p>You should be using ReplacingMergeTree if you are doing updates at the current moment.
Here's a list of other open source OLAP systems out there. Clickhouse is on the list along with others like StarRocks. <a href="https://atwong.medium.com/top-open-source-alternatives-to-olap-databases-snowflake-redshift-and-bigquery-21b4f0f0cd31" rel="nofollow noreferrer">https://atwong.medium.com/top-open-source-alternatives-to-ol...</a>
I do not think ClickHouse Inc will Abandon to Open Source, but following current trend it may focus on restricting Innovations to their proprietary cloud only product.<p>We see it with Oracle (MySQL) where most of innovations is happening in cloud only "Heatwave" or MongoDB where MongoDB Atlas increasingly getting features not available in their Community (SSPL) version
New column store alternative : <a href="https://github.com/hydradatabase/hydra">https://github.com/hydradatabase/hydra</a><p>HN: <a href="https://news.ycombinator.com/item?id=37571974">https://news.ycombinator.com/item?id=37571974</a>