moonikakiss

6 ms·

1.
Managing AI Coding Costs at Scale(databricks.com)
317 points by moonikakiss 1mo ago|268 comments
2.
Beating GPT-5.6 Sol on retrieval with 100x cheaper open models(neon.com)
437 points by moonikakiss 1mo ago|128 comments
3.
Databricks Acquires Mooncake Labs(databricks.com)
3 points by moonikakiss 1y ago|0 comments
4.
Moonlink: Streaming Ingest for Apache Iceberg(github.com)
1 points by moonikakiss 1y ago|0 comments
5.
A database kernel engineer's take on Apache Iceberg(mooncake.dev)
3 points by moonikakiss 1y ago|4 comments
6.
How to turn Apache Iceberg into a real-time columnstore?(mooncake.dev)
2 points by moonikakiss 1y ago|0 comments
7.
What if made we Apache Iceberg feel like a RDMS?(mooncake.dev)
1 points by moonikakiss 1y ago|0 comments
8.
Pg_mooncake: Columnstore Mirror of Postgres Tables(github.com)
1 points by moonikakiss 1y ago|0 comments
9.
But, my Postgres (+ lakehouse) is free(mooncake.dev)
2 points by moonikakiss 1y ago|1 comments
10.
AI-First Law Firm?(techcrunch.com)
1 points by moonikakiss 1y ago|0 comments
11.
But, my Postgres (+Lakehouse) is free(mooncake.dev)
3 points by moonikakiss 1y ago|0 comments
12.
by moonikakiss 1y ago
I did due-diligence on Builder.AI for a venture firm I was interning at (circa 2019). It was extremely apparent (Glassdoor, talking to any employee) it was complete BS. When I say apparent, it took less than 15 minutes and a couple of googl
13.
ClickStack: A High-Performance OSS Observability Stack on ClickHouse(clickhouse.com)
3 points by moonikakiss 1y ago|0 comments
14.
by moonikakiss 1y ago
totally agreed on 3. You're also missing the challenges of dealing with updates/deletes; and managing the many small files. CDC from OLTP to Iceberg is extremely non-trivial.
15.
HTAP is Dead(mooncake.dev)
159 points by moonikakiss 1y ago|90 comments
16.
Open Data Standards: Postgres, OTEL and Iceberg(supabase.com)
2 points by moonikakiss 1y ago|0 comments
17.
by moonikakiss 1y ago
yep exactly. we can wait for replay LSN. So you're only reading once all writes to OLTP are complete.
18.
by moonikakiss 1y ago
great point. with pg_mooncake v0.2 (launching in ~couple weeks), you'll be able to get a columnar copy of your Postgres that's always synced (<s freshness). Keep your write path unchanged, and keep your Postgres where it is. De
19.
by moonikakiss 1y ago
yes you can. Even if the columnar tables are in the read replica. you'll be able to do joins with your OLTP tables
20.
by moonikakiss 1y ago
exactly. we take the CDC output / logical decoding from your OLTP tables and write into a columnar format with <s freshness. We had to design this columnstore to be 'operational' so it can keep up with changing oltp tables
21.
by moonikakiss 1y ago
great blog. It seems like you might benefit from columnar storage in Postgres for that slow query that took ~20seconds. It's interesting that people typically think of columnstores for strict BI / analytics. But there are so many
22.
HTAP Databases Are Dead(mooncake.dev)
3 points by moonikakiss 1y ago|1 comments
23.
by moonikakiss 1y ago
Even more so after Neon + Databricks.
24.
by moonikakiss 1y ago
it was a couple things. 1. It's really really hard to replace anyone's OLTP. 2. OLTP and OLAP are owned by such different teams. Who do you make your champion? 3. The modern HTAP dream is possible without something like SingleSt
25.
by moonikakiss 1y ago
This is really, really exciting. I see it as the “right” way OLTP and OLAP will converge. The OP and I built an HTAP system at SingleStore. A single database with one copy of data for both OLTP and OLAP workloads. HTAP never took off [0]. W
26.
by moonikakiss 1y ago
With the push towards open table formats (Iceberg) from both Snowflake and Databricks, it's even harder to get your Postgres OLTP tables ready for OLAP. The problem isn't in the CDC / replication tools in the market. The prob
27.
Why are there so many Postgres Analytics startups?(mooncake.dev)
1 points by moonikakiss 1y ago|0 comments
28.
by moonikakiss 1y ago
Neon actually does have a columnstore extension with pg_mooncake today. The key difference with pg_mooncake vs hydra is the bet on open storage formats (Iceberg). (1) https://neon.tech/docs/extensions/pg_mooncake
29.
HTAP Is Dead(mooncake.dev)
7 points by moonikakiss 1y ago|0 comments
30.
by moonikakiss 1y ago
Is streaming into Iceberg / R2 catalog going to become a priority?

More ›