Great exit for the PeerDB team, congratulations. However Postgres has improved so much over the years that many users don't need to move their data out of Postgres to get great performance for analytical queries.

At Tablespace https://www.tablespace.io we've built a ColumnStore Index natively as a Postgres extension that enables users to execute analytical queries in Postgres 2X faster than Clickhouse Cloud without moving their data https://www.tablespace.io/blog/postgres-columnstore-index-vs...

For transparency I'm the CEO of Tablespace.

Thanks for posting this question! I previously worked at Citus for 8 years, where we tried to bring real-time analytical capabilities to Postgres. It was common to see POCs go sideways, and several customers (including CloudFlare and Heap) moved from Citus to an analytics-specialized database. For example, CloudFlare moved from Citus to ClickHouse. This was one of the inspirations for me to build a company (PeerDB) that brings specialized OLTP and OLAP databases together.

This is not to say that Postgres cannot support larger-scale analytical workloads, but it will take time. ClickHouse has taken 10 years of effort and development to get where it is now.

I would love to understand how tablespace performs at scale in production workloads. Are there any references that you could share. :)

Please don't copy-paste comments. It makes merging threads a pain.

Probably best not to do marketing on top of someone else's announcement too... https://x.com/mitchellh/status/1759626842817069290?t=57h-biZ...

Congrats on Tablespace! it's good to see innovation in that space and I always love to see ClickBench mentioned in the wild :) (I work for ClickHouse).

I'm curious to hear your take about the tradeoffs that the HTAP model introduces? any impact on ingest times or query throughput for example?

Olo, CEO of https://www.tablespace.io here. We've built a columnstore extension for Postgres that is faster than ClikcHouse for real-time analytics in their own ClickBench Benchmarks. Feel free to check it out - https://www.tablespace.io/blog/postgres-columnstore-index-vs...

Hi there CEO of Tablespace here, no we are comparing 16CPU/32GB which Tablespace used to 24CPU/96GiB which ClickHouse cloud uses.Tablepsace is 1.6x faster running the benchmark even though we are using a much smaller compute shape.

Hi there CEO of Tablespace here. I'll address your points. The disk space that Postgres on the Tablespace platform uses is similar to that Vanilla Postgres uses. Also the ingestion time is similar as well - https://n9.cl/g4woz Remember we are still in beta and the indexing time will only be improved as well. We are not optimising for disk space usage, we are optimising execution for speed. Further, although Tablespace uses more disk space as you say, it is still 50% less expensive than ClickHouse Cloud. This cost saving is very conservative as well because users of ClickHouse Cloud still need to maintain a separate transactional database. Tablespace is fully HTAP.

The 1.6x speed-up figure is indeed correct. Have a look at the relative times for Tablespace 1.44 vs. ClickHouse (AWS) 2.34 and ClickHouse (GCP) 2.22 this works out to a 1.6x speed-up.

Hi there. CEO of Tablespace here. We have indeed made something that people can try ;) We built a Columnstore index for Postgres which allows users to run fast OLAP analytics on Postgres. We just launched in open beta and are looking for feedback, comments, questions. The blog post is a comparison of Tablespace vs. ClickHouse for OLAP workloads. We also included links to the benchmark scripts that people can run themselves as well.

You can post the blog post as a regular submission, you can post whatever it is you want people to try as a Show HN, you can't post a blog post as a Show HN. That's in the thing I linked:

Off topic: blog posts, sign-up pages, newsletters, lists, and other reading material. Those can't be tried out, so can't be Show HNs. Make a regular submission instead.

Ah got you. Will do.