This article is a translated version of a blog originally written in Japanese.
If you notice any English expressions that seem to change the intended meaning, please know that I’m actively reviewing and revising as needed. Your patience is appreciated!

---

TL;DR

At Snowflake Summit 2025, the surprise announcement of Snowflake’s acquisition of Crunchy Data—a major force in the PostgreSQL community—signals a bold move to reshape the future of database platforms.

As a long-time advocate of both Snowflake and PostgreSQL, I see this as more than a product strategy—it’s a cultural shift. If handled well, it could bring enterprise-grade power to Postgres while preserving its open-source roots.

What’s at stake?
The evolution of HTAP, the sustainability of PostgreSQL development, and the birth of a new kind of data platform leader.

Let’s talk about what this could mean for developers, the OSS community, and the future of databases.

---

Hello, I’m Kos (Kosuke Kida) .
As a Senior Data Evangelist at Exture Inc., I’m responsible for driving thought leadership and developing pioneering use cases around modern data platforms. Over the past two years, I’ve been deeply involved with Snowflake, which led to my recognition as a Snowflake Data Superhero.

I’ve also been a longtime PostgreSQL enthusiast—even though I no longer work directly with it—and have served as the chairman of the board of director of the Japan PostgreSQL User Group. While I haven’t contributed to the core source code, I’ve been honored to be named a PostgreSQL Contributor in recognition of over a decade of community leadership in Japan.

It’s from this dual perspective—someone who loves both Snowflake and PostgreSQL—that I want to share my thoughts on the major announcement made at this year’s Snowflake Summit. I’m writing this article not only to reflect on what this could mean for the future of databases, but also to invite everyone who works with data to follow this evolution together.

To be clear: as someone who cares deeply about both technologies, I couldn’t be more excited. I wholeheartedly welcome this announcement and have incredibly high hopes for what’s to come.
Even hours after the keynote, I still find myself telling everyone I meet here on-site—and each time, I get goosebumps all over again. I’m so glad I was here in person to witness this moment!

A Groundbreaking Announcement from the Snowflake Summit Opening Keynote

Day 1 of Snowflake Summit 2025 has officially kicked off!
As is tradition, the first day began with the Opening Keynote—a session less about individual feature releases and more about product vision, strategic direction, and Snowflake’s stance on the most talked-about technologies shaping the future of data. It featured compelling customer stories and conversations with thought leaders from across the industry.

While I’ll leave the full recap of the keynote to other posts, I want to highlight one specific announcement that, as someone with deep ties to the PostgreSQL world, I simply can’t stay silent about.
It was bold. It was unexpected. And it has the potential to shake up the entire database ecosystem.

Let’s dive into that.

A Sudden, Game-Changing Move: Snowflake Acquires Crunchy Data

Just hours before the Opening Keynote, this headline dropped—and it sent shockwaves through the data community:

Snowflake has acquired Crunchy Data, one of the leading voices in the PostgreSQL ecosystem.

Here are the official announcements from both companies:

• Snowflake’s announcement
• Crunchy Data’s announcement

Naturally, this news became a key topic during the keynote itself.
Here’s a summary of the announcement, as distilled by AI-assisted notes:

Snowflake has entered into an agreement to acquire Crunchy Data, a recognized leader in open-source PostgreSQL technologies.
The vision? To combine PostgreSQL’s openness, extensibility, and developer-first principles with Snowflake’s signature strengths in performance, governance, and scalability.
Developers will be able to continue using the familiar Postgres tools they love—while benefiting from Snowflake’s cloud-native power.

This move signals not just a technical expansion, but a cultural one.
And for those of us who’ve spent years in the PostgreSQL world, it opens up a realm of thrilling possibilities.

Crunchy Data and the Legendary Tom Lane

For those familiar with the PostgreSQL world, Crunchy Data needs no introduction.
Far more than just a managed Postgres provider, Crunchy has been a pillar of the PostgreSQL community for over a decade. They’ve not only delivered robust DBaaS offerings and enterprise support, but have also contributed extensively to the PostgreSQL codebase itself—year after year, actively participating in global dev summits and regional PostgreSQL events around the world.

In short, Crunchy isn’t merely using PostgreSQL. They’ve helped shape it.

And of course, no mention of Crunchy Data is complete without highlighting the presence of Tom Lane—the legendary developer universally regarded as one of the greatest minds in PostgreSQL’s history.

Tom is known for his deep understanding of the PostgreSQL codebase down to the finest details, as well as his leadership in technical discussions on the community mailing lists.
He’s so influential, in fact, that it’s become a running joke in some Postgres circles:

“A talk isn’t complete until you’ve discussed how to survive Tom Lane’s code review.”

Which brings us to the real gravity of this acquisition.

No matter how many database companies claim to have “PostgreSQL compatibility,”
there’s only one Crunchy Data—and there’s only one Tom Lane.
Snowflake didn’t just acquire a product. They gained a community, a legacy, and a level of credibility in the open-source world that simply can’t be replicated.

That’s what makes this announcement such a powerful statement of intent.

What Could This Mean for Snowflake’s Feture?

During the keynote, Snowflake emphasized that it plans to enhance its platform while continuing to support and contribute to the open-source PostgreSQL community. That said, specific technical details weren’t shared—yet.

So what follows is partly speculation (and yes, a bit of wishful thinking) on my part. But I hope it sparks ideas.

Enterprise-Grade PostgreSQL?

Having worked extensively with both commercial databases and PostgreSQL, I’ve seen the familiar narrative: PostgreSQL excels in many areas, but enterprise users often point to weaknesses in security, governance, and operational manageability—areas where Snowflake shines.

Now, imagine what would happen if Snowflake’s architecture and engineering muscle were applied to these gaps.
We might witness the emergence of a next-generation PostgreSQL built for the enterprise, without compromising its open-source principles.

At the same time, Snowflake doesn’t currently have a clear, go-to solution for powering real-time operational workloads—despite the introduction of Hybrid Tables. While promising, they haven’t yet reached a level of frictionless adoption.

Both companies seem to genuinely admire one another, based on the tone of their respective announcements.
One lingering concern? Cost.
While proprietary value-add is great, many of us in the open-source community are hoping Snowflake will also contribute enterprise-grade enhancements to the OSS PostgreSQL itself—not just keep them in a paid version.

The HTAP Challenge

With Hybrid Tables, Snowflake has already stepped into the world of HTAP (Hybrid Transactional and Analytical Processing)—offering transactional updates on big data, something traditionally considered incompatible.

But here’s the reality: migrating existing OLTP workloads—or designing new real-time apps directly on Snowflake—remains a stretch for many users.
Why? Performance, cost, and developer familiarity are just a few reasons. Hybrid Tables are impressive, but adoption is cautious.

Now, enter PostgreSQL.

For traditional OLTP systems, or the wave of AI-driven apps that require instant, context-aware data interactions, PostgreSQL remains one of the most practical and proven choices.

But here’s the tension:
Even if Snowflake runs PostgreSQL as part of its product family, if they operate on fundamentally different file formats, Snowflake loses one of its core advantages as an analytical database.
Sure, you could stream data in near real-time between systems—but that conversion carries cost, complexity, and latency.

Sound familiar? That’s the same core challenge Oracle has spent years tackling with In-Memory features—and many other vendors with HTAP claims still struggle with today.

(I’m happy to be corrected by anyone more up to speed on recent HTAP advances—please teach me!)

So Why Would Snowflake Acquire PostgreSQL?

Here’s where my imagination kicks in.

Snowflake stores data in micro-partitions—immutable blocks rich with metadata, allowing it to skip irrelevant data at query time (via pruning). That’s a big part of what makes it performant for analytical workloads.

PostgreSQL, on the other hand, uses row-level indexes to locate and fetch matching records.

Both systems also implement MVCC (Multi-Version Concurrency Control) differently:

• Snowflake rewrites micro-partitions on any update, keeping old versions intact for features like Time Travel.
• PostgreSQL tracks versioning per row, updating just the necessary data.

Now imagine a hybrid model:

What if “yesterday’s data” lived in Snowflake, while “just-in-time, mutable data” stayed in PostgreSQL—
and at query time, both were seamlessly combined?

This would preserve Snowflake’s analytical performance, PostgreSQL’s transactional agility, and sidestep the cost of constantly converting between formats.

Technically complex? Absolutely.
But if any team has the vision, the talent, and the architecture to try it—it’s the team that just acquired Crunchy Data.

What About PostgreSQL and the Broader OSS Ecosystem?

One of the most important aspects of this announcement—beyond product roadmaps—is how Snowflake will engage with the open-source PostgreSQL community.

Whenever a major company acquires an influential open-source player, the same question always arises:

Will they continue contributing to the open-source project, or will things quietly shift toward a proprietary fork?

It’s worth clarifying:
PostgreSQL is not owned by Crunchy Data. Its source code is maintained by the PostgreSQL Global Development Group (PGDG). This decentralized governance ensures that no single company—regardless of how influential—can control or derail the project.

That means:
No matter what happens with Crunchy Data or Snowflake, PostgreSQL will remain open, free, and community-driven.
To all PostgreSQL users reading this: you can breathe easy.

Potential Pitfalls—and Why They Probably Won’t Happen

Of course, there are scenarios we’d rather not see. For example:

• Former Crunchy team members focus exclusively on Snowflake’s proprietary PostgreSQL fork, abandoning upstream contributions.
• Snowflake tries to push vendor-specific changes into PostgreSQL core that benefit its own ecosystem but break compatibility for others.
• Key contributors like Tom Lane—or those who helped shape Crunchy Data—are sidelined or exit the project altogether.

These outcomes would be obvious, damaging, and—frankly—bad business.
Snowflake would gain nothing but backlash. So let’s assume (and hope) they’re smarter than that.

A New Chapter for Databases?

If we stay optimistic, there’s a much more exciting future in store.

Snowflake and Crunchy Data together have the potential to advance PostgreSQL itself—and even broader database technology—in ways we haven’t seen before.
This could be a pivotal moment where the data world welcomes a new kind of leader: one that champions both enterprise-grade performance and open collaboration.

After all, this isn’t Snowflake’s first open-source investment.
They’ve already made meaningful contributions to projects like:

• Streamlit
• Modin
• NiFi
• Terraform integrations

Each has added real value, and Snowflake hasn’t tried to lock them down or warp them into closed platforms.
To me, that track record—combined with the company’s financial resources and cloud-native mindset—is a promising sign, especially for younger OSS projects.

But PostgreSQL is not a young project. It’s a global pillar that’s been evolving for over 30 years.
What we’ll all be watching now is how Snowflake chooses to support and enhance this legacy—and whether they can give back the same technical excellence that brought them to this point.

I, for one, hope they embrace the role of the hero.
Let’s see if Snowflake can prove itself not just as a product powerhouse, but as a force for good in the open data world.

---

Final Thoughts

If you’ve made it this far, thank you.
As someone who’s walked alongside both the PostgreSQL and Snowflake communities for years, I’ve never been more excited—or curious—about what comes next.

This announcement isn’t just about a product.
It’s a signal that the very definition of a database platform may be evolving.
And it’s a reminder that the lines between open-source and enterprise, transactional and analytical, are blurring faster than ever.

Let’s keep watching. Let’s keep contributing.
And let’s hope this is just the beginning.

— Kosuke Kida
Sr. Data Evangelist @ Exture / Snowflake Data Superheroes 2025 / PostgreSQL Contributor