PostgreSQL and pgvector: Now Faster than Pinecone, 75% cheaper, 100% open-source.
Introducing pgvectorscale, an open-source PostgreSQL extension that builds on pgvector, enabling greater performance and scalability.
Here’s how pgvectorscale helps pgvector outperform specialized
🍿 NEW VIDEO: How to Combine Vector Search and SQL for RAG.
Learn how to build a “chat with your database” RAG app and answer questions like “Find robotics companies in CA and last week’s news for them.”
Stack: GPT4,
@pgvector
, Instructor by
@jxnlco
Link below 🔗👇
Introducing ✨pgai✨
Pgai is a PostgreSQL extension that brings more AI workflows to PostgreSQL, like embedding creation and model completion.
Pgai helps more PostgreSQL developers gain AI engineering skills, making it easier for them to build search, and retrieval-augmented
Have you ever wondered why your
@PrometheusIO
queries perform as they do? 👀
In our latest blog post, we break down how storage, indexing, and querying work in Prometheus, giving you practical tips on how to make your PromQL queries faster 🔥
Why juggle multiple databases, each with its own set of rules and languages? 🤹♂️
Just use
#PostgreSQL
for everything, giving you versatility, reliability, and saving time for what really matters: innovation. 🚀
🎉 Massive CONGRATULATIONS to
@michaelfreedman
- Timescale CTO & co-founder - who was just named the Robert E. Kahn Professor of Computer Science
@Princeton
🎉
Remember Insights from yesterday? We used Timescale to build it: collecting +10 billion records a day, up to +350 TB in a single database & counting.
This is our company’s largest dogfooding effort ever.
#PostgreSQL
can scale.
Learn how we did it:
Ollama now in pgai: Use open-source LLMs in PostgreSQL 🦙
You asked and we delivered…PostgreSQL developers can now access Ollama models right inside the database, with the pgai extension. ✨
pgai is an open-source PostgreSQL extension that brings AI models closer to your data.
Creating embeddings for your data in SQL
With pgai, a new open-source PostgreSQL extension bringing AI workflows to your database, you can create OpenAI embeddings for your data with a simple SQL query (see the code snippet for more)
See this and other examples of how you can
He helped design TimescaleDB’s continuous aggregates. 🔧 Now,
@HarlemCavalier
shares how this feature was influenced by PostgreSQL views and materialized views.
Have no idea what these are? That’s what this post 👇 is for.
Today, we’re launching Dynamic PostgreSQL. 🚀
Instantaneously scale your database compute within a min/max range according to load. Buy the base, rent the peak. Usage-based storage pricing.
Top performance. Less expensive. Worry-Free. 100%
#PostgreSQL
.
Dynamic Infra Launch Week kicks off today. Continue to level up your database game with Timescale. 🚀
Dynamic PostgreSQL, usage-based storage, additional extensions, and more.
“More complexity means less time to build. Complex architectures are more brittle, require more complex application logic, offer less time for development, and slow down development.”
#PostgresForEverything
🚀 TimescaleDB 2.0 is now GA 🥳
✅ Multi-node, petabyte scale for the most demanding time-series workloads
✅ 💯 free features
✅ Even better user experience, new functionality & more
Read full post for details 👇 1/
Big day for us
@TimescaleDB
. After months of hard work, we’ve surpassed 1 million downloads and just released version 1.0!
Blog has more, including quotes from some of our production users at Comcast, Bloomberg, Cray, and more:
🐯🚀!!
Timescale +
@popsql
= a developer’s dream come true! 🥳
🚀 Building the Best PostgreSQL GUI for everyone, everywhere 🚀
We’re excited to announce that the PopSQL team is joining Timescale to help build the best PostgreSQL developer experience for the cloud era. ☁️
👋 Calling all
#PostgreSQL
users – the 2023 State of PostgreSQL survey is now open! 🎉
Share your feedback, favorite extensions, community experiences, and more 👇
The survey is open until Aug 31 @ 11:59 p.m. PT ✨
When you’re scaling a PostgreSQL database in AWS, what you want is to make PostgreSQL compatible with Amazon S3 so that you can store large volumes of data in the DB for less $$$, right? 😉
Guess what? We’ve built this, so you don’t have to 😎
🌟🌟 We hit 12,000
@github
stars this week! 🌟🌟
Biggest thank you to everyone who’s starred us, submitted issues and helped make TimescaleDB better. 🙏
Haven’t checked out our repo yet? You can find it👇
1/ Promscale 0.8 is out 🚀
This version includes new features that make it simple to manage your
@opentelemetry
tracing data 🙌, improved CLI flags, new extension packages for Linux, additional instrumentation, and more!
(More in 🧵)
Should you use a standalone vector database for your
#AI
app? 🤔
With pgvectorscale, PostgreSQL and pgvector now rivals specialized vector databases like Pinecone in performance at a fraction of the cost. By incorporating specialized data structure and algorithms, pgvectorscale
Big news 💥 Timescale Cloud now includes a bottomless, consumption-based object store built on Amazon S3!
📈Bottomless scalability
💰Cost-effective
💻Query S3 (+ all your data) directly with SQL
Supercharged storage for PostgreSQL, one command away.
We ❤️
@Cloudflare
, so we were super excited when they launched
#hyperdrive
to speed up edge
#postgresql
connections with Timescale supported!
Read about how we contributed a full tutorial to spin up a serverless, globally distributed time-series API
💪 How we made PostgreSQL faster than Pinecone
With all the excitement around pgvectorscale this week, we’ve received many good questions from developers curious about how pgvectorscale speeds up PostgreSQL beyond standalone vector databases like Pinecone.
@cevianNY
wrote this
At
#StrataData
London, Erik Nordström speaking now about using TimescaleDB as long-term storage for
@PrometheusIO
.
Leverage full
#SQL
to analyze your Prometheus metrics, including JOINs with other data.
#tsdb
We are excited to announce the 1st release candidate for
@TimescaleDB
1.0! This milestone signifies TimescaleDB’s maturity & readiness as an enterprise time-series database designed to support full
#SQL
& scale.
🥳 BIG NEWS for us, our community members & developers everywhere.
✅ We just raised $40M to help developers measure everything that matters
✅ 2M+ active databases
✅ 10+ launches over the next month
✅ 60+ team members, 6 continents, and hiring!
Read more 👇1/
We took the time to write up a detailed primer on
#PostgreSQL
streaming replication, complete with performance measurements.
It's a quick, 14 minute read ;)
Happy Lunar New Year to everyone who celebrates! We’re excited that it’s the year of the Tiger. 🐅
At Timescale, our mascot, Eon, is a fun-loving data-driven tiger.
To celebrate, we’re sharing our favorite facts about tigers for the next 15 days. 🎉
🥳 We just announced some big news about our software license (more rights for users & all paid features are now free!).
👉 Check out
@acoustik
thread to learn how we're building a self-sustaining business in the cloud era & why it matters for the software industry as a whole.
💸💸💸 Today we’re announcing that we’re going all-in on cloud, making all of our enterprise features available for free, updating our software license to provide greater rights to our users, and reinforcing our commitment to the
@TimescaleDB
community. 1/
Help us build the next great database at Timescale – we're
#hiring
! We have open roles across all departments right now! Check out our open positions and apply today:
When building AI applications, developers often wonder if they need a standalone vector database or if they can use a general-purpose database like PostgreSQL. 🤔
While dedicated vector databases seem to promise better performance, pgvectorscale challenges this by bringing
Pgvector vs Pinecone: Vector Database Performance and Cost Comparison
🤔Do you need a separate vector database?
When building an AI application, do you need a standalone vector database like Pinecone? Or can you use PostgreSQL, a database you might already be familiar with, know
🎊TimescaleDB 1.7 is here! Our latest release brings PG12 support, new real-time aggregate functionality, additional data management features in our Community edition, and more.
Read our latest blog to see what's new and how to get started:
📆Today marks a major milestone, two years in the making.
Big 🙏 and 👏 to everyone who's played a role in bringing TimescaleDB 2.0 to life - we couldn't do it without you 💛!
Read on for details 🔽
🥳 Today’s the day! TimescaleDB 2.0 RC has arrived 🚀
Multi-node. Petabyte-scale. Relational. Purpose-built for time-series. 💯 free.
Get started and learn what's new👇
1/
We unabashedly support everyone’s right to make decisions about their own reproductive health, and are deeply disappointed in the US Supreme Court’s decision to overturn Roe v. Wade.
We are committed to supporting our people and creating an inclusive environment for all.
Hey
#PostgreSQL
users! 👋 Check out this handy cheat sheet that will save you time by providing easy access to all the essential PostgreSQL commands you need to know. -->
👋 Calling all
#PostgreSQL
community members - the 2021 State of Postgres survey is now open 🎉!
Share your feedback, favorite extensions, community experiences, and more 👇
Survey closes April 6, 2021 @ 11:59pm PT
1/ Wishing a very special happy birthday to
@PostgreSQL
! 🎉 🎂 🎈🐘
We're celebrating the release of PostgreSQL by rewinding it back with some of our favorite posts.
Our friends at
@DNSFilter
wrote a piece on why they replaced InfluxDB with TimescaleDB. Their results: 10x better resource utilization, even with 30% more requests. Thanks
@MikeSchroll
. RT then read it here:
ICYMI: yes, multi-node TimescaleDB will indeed be free 🥳.
Check out Ajay's post for details, why we made our decision & how to get started (+ even more ways we're investing in our community)
...and big 🙏 to our community members everywhere!
#PostgreSQL
#timeseries
We just made a big announcement: multi-node TimescaleDB, our elastic, petabyte-scale, distributed time-series database, will be available for free.
And that post is now
#1
on HackerNews 🔥
What are the differences between
@PrometheusIO
and
@opentelemetry
metrics? When to use one or the other? And how to convert between both standards? 🔥🧐
@jamessewell
tells you everything 👇
1/
#AlwaysBeLaunching
continues! Today we're sharing new performance improvements in TimescaleDB 2.7, including 400x faster queries using the now( ) PostgreSQL function ⚡️
Learn more about this and other performance improvements in 🧵 (below)
Did someone say CHEAP STORAGE FOR POSTGRES? 👍
You can now run ‘data_tiering_policy’ in your Timescale Cloud hypertables and tier data to a low-cost storage layer built on Amazon S3. Your data remains fully queryable, with no extra charge per query!
READ:
@sql_johnpruitt
has been using SQL for decades.
His tips for learning the programming language? Do not approach it from an object-oriented point of view, and read
@ModernSQL
.
Check out more advice in this interview for our June SQL series 👇
PostgreSQL is in pole position to be the foundational database for AI
With the rise of AI applications leveraging LLMs, developers need more from their databases.
The good news is PostgreSQL is evolving to meet these needs through its rich ecosystem and community 💪.
This
🧑💻
#DevTakeover
Week 2 🧑💻
Meet Sebastian Webber (
@sebawebber
), Senior Software Engineer.
Read his experience working with Timescale and some other fun facts👇
Introducing the Time Series Benchmark Suite: New framework by
@TimescaleDB
engineers for comparing the performance of databases storing time-series workloads.
#TSBS
#tsdb
#PostgreSQL
What is SQL used for?
Read how
@samokhvalov
uses SQL to help devs find a place to experiment with their ideas (and took matters into his own hands to make it happen).
New Features We're Looking Forward to in PostgreSQL 17 (scheduled for September)! 👇
In PostgreSQL 17, we're seeing some exciting changes that promise to modernize and enhance the database's performance and usability.
Notable features include:
👥pg_createsubscriber
💪 Support
Real-time data aggregation is now optimized in
#PostgreSQL
, achieving a 50,000% performance increase.
The update in TimescaleDB 2.13.0 enhances continuous aggregates by integrating real-time updates and significantly reducing query planning time. ⤵️
Really enjoyed this detailed and insightful analysis on why
#PostgreSQL
continues to be a go-to for developers, check out this great read from
@starbuxman
: 🔥
👋 Calling all
#PostgreSQL
users – the 2022 State of PostgreSQL survey is now open!
Share your favorite extensions, preferred frameworks, community experiences, and more.
The survey is open until June 30th @ 11:59 p.m. PT ✨
Today
@zabbix
4.2.0alpha3 was released. Among many new features and improvements is official support for TimescaleDB! Zabbix users now have a scalable trouble-free solution for storage of time series data.
More on the release here:
“Just Use Postgres For Everything” is popping up everywhere these days. We’ve seen it used as both a rallying cry and a mic drop, but what does it actually mean? 😏👇
✨ TimescaleDB docs just got some 💯 updates:
✅ Faster access to what you need
✅ Better getting started materials
✅ Easier contribution process (our docs are open-source!)
Check them out:
1/
A new version of
@JaegerTracing
is out - including a contribution by
@cevianNY
! 🤩
The support for a remote gRPC storage plugin in Jaeger will make it easier to use Promscale and other observability solutions 🔥
Thanks a lot to the Jaeger team!
It's "Scaling Postgres" Launch Week.
Announcing Tiered Storage on Timescale💥
Store your older data in a low-cost, bottomless storage tier while still being able to query it via normal SQL from within
#PostgreSQL
. Now in general availability.
📢 Our CTO and co-founder,
@michaelfreedman
, joined
@schwiepit
of
@TwoSigmaVC
for an enlightening firechat at the
@StartupGrind
Product + AI Summit.
Mike shared insights on "Product Development in the Age of the Cloud" - a definite must-watch
We love solving hard database problems. Building Vectorized Query Execution is definitely one!
Read about this new capability built for our columnar storage, available in the latest version of TimescaleDB.
#GoFast
🔥