philippemnoel

17 ms·

1. 211.▲
   by philippemnoel 3y ago

   Check out https://github.com/paradedb/paradedb/tree/dev/pg_analytics , we're shipping this week for column-based storage in Postgres. It's not LSM, though, but Parquet/Arrow
2. 212.▲
   by philippemnoel 3y ago

   Hey! I'm one of the makers of pg_analytics/ParadeDB. We store tables column-based. You can do CREATE TABLE (...) USING deltalake; when using ParadeDB and it will be stored column-based (and soon, with support for also querying clo
3. 213.▲
   by philippemnoel 3y ago

   This is interesting. We've been seeing a pretty split desire where some companies would prefer ParadeDB installed as an extension in their existing database, while others would prefer ParadeDB as a separate search node (for independent
4. 214.▲
   by philippemnoel 3y ago

   This is a great post. As an extension developer myself (pg_bm25), seeing more native support for search-related functionalities that extensions can harness would be really great. I'm surprised nothing was mentioned around pgBouncer&#x2
5. 215.▲
   by philippemnoel 3y ago

   I think bringing RocksDB into Postgres as a Table access method would really be a powerful feature, as an alternative to heap
6. 216.▲
   by philippemnoel 3y ago

   Hi there! ParadeDB author here. We don't believe AGPL should prevent you from self-hosting for your own internal usage. We see ourselves as becoming a Postgres-based Elasticsearch alternative, primarily used internally for observabilit
7. 217.▲
   by philippemnoel 3y ago

   ParadeDB author here. You're right, currently we only do the "Search" in Elastic, but we're planning to do real-time analytics, logs processing, etc. eventually
8. 218.▲
   Prometheus_fdw: Seamless Monitoring in Postgres(tembo.io)

   1 points by philippemnoel 3y ago|0 comments
9. 219.▲
   by philippemnoel 3y ago

   ParadeDB developer here -- we're developing a way for users to no zero-ETL sync between any managed Postgres and ParadeDB, so that you can get a dedicated search node on Postgres without slowing down your production DB. We expect this
10. 220.▲
    by philippemnoel 3y ago

    You can even do Elastic-level full text search in Postgres with pg_bm25 (disclaimer: I am one of the makers of pg_bm25). Postgres truly rules, agree on the rec :)
11. 221.▲
    by philippemnoel 3y ago

    I just got an email saying they're moving to pre-paid billing... Seems like Sam Altman might've hidden some very large financial costs that the board just discovered?
12. 222.▲
    Ask HN: Do you use Postgres and ClickHouse together?

    8 points by philippemnoel 3y ago|0 comments
13. 223.▲
    by philippemnoel 3y ago

    ParadeDB and pg_bm25 offer weak consistency. pg_bm25 doesn't slow down transactions for indexing, and like ElasticSearch it becomes become eventually consistent shortly after (typically at most a few seconds, altough your mileage may v
14. 224.▲
    by philippemnoel 3y ago

    Hey! Absolutely, we would love to offer as many cloud providers as possible for our compute backend. We're starting with AWS, and will be adding other clouds based on demand. I've added Scaleway to our list, and if you'd like
15. 225.▲
    by philippemnoel 3y ago

    Our goal is for one day ParadeDB to be a viable alternative to AWS RDS/Aurora, so that like you say, you don't need to keep data in-sync and can just use one system (ParadeDB). Soon it will be possible for you to have ParadeDB run
16. 226.▲
    by philippemnoel 3y ago

    Yes! We have another extension, pg_search, which is specifically for hybrid search using pg_bm25+pgvector. You can find it here: https://github.com/paradedb/paradedb/tree/dev/pg_search
17. 227.▲
    by philippemnoel 3y ago

    pg_bm25/ParadeDB author here. What we're doing is building an opinionated alternative within PostgreSQL. If you are not using Postgres, or want your system to be separate, Elastic is still the best choice and will likely remain so
18. 228.▲
    by philippemnoel 3y ago

    Thanks for sharing, will look to add a benchmark for that as well
19. 229.▲
    by philippemnoel 3y ago

    pgrx is awesome and making pg_bm25 would've been infinitely more challenging without it. Check them out if you want to make a Postgres extension, we can't recommend them enough
20. 230.▲
    by philippemnoel 3y ago

    That's right, we do support pgvector (it is pre-installed on ParadeDB) and support full HNSW. In fact, we even have another extension, called pg_search, which is the combination of searching on pgvector and pg_bm25 for better results!
21. 231.▲
    by philippemnoel 3y ago

    ParadeDB author here -- correct! We plan to offer a hosted version soon and the idea behind picking AGPL is to be as permissive as possible so that people can use the product for free, but also protect ourselves from abuse in case a large c
22. 232.▲
    by philippemnoel 3y ago

    One of the ParadeDB authors here, hey! Thanks for pointing this out, you're completely right. That's an oversight on our end. We'll update the benchmarks and re-run them to correct this :)
23. 233.▲
    Pg_bm25: Elastic-Quality Full Text Search Inside Postgres(docs.paradedb.com)

    7 points by philippemnoel 3y ago|0 comments
24. 234.▲
    by philippemnoel 3y ago

    Yes! That will be coming very soon :)
25. 235.▲
    Pg_bm25: Rust-based full text search engine in Postgres(github.com)

    12 points by philippemnoel 3y ago|4 comments
26. 236.▲
    Show HN: Retake – Open-Source Hybrid Search for Postgres(github.com)

    88 points by philippemnoel 3y ago|23 comments