In June 2025, Snowflake — the AI Data Cloud company listed on the New York Stock Exchange — announced its intention to acquire Crunchy Data, one of the most trusted names in enterprise-grade open source PostgreSQL. The deal, valued at approximately $250 million, was unveiled at Snowflake's annual Summit conference in San Francisco and represents one of the most significant events in the PostgreSQL ecosystem in recent memory.
Crunchy Data, founded in 2012 and headquartered in Charleston, South Carolina, had built its reputation over more than a decade as a provider of managed PostgreSQL services, Kubernetes-native deployments, and on-premise PostgreSQL solutions for Fortune 500 companies and U.S. government agencies, including the Department of Homeland Security. With approximately 100 employees joining Snowflake as part of the deal, Crunchy Data's team and technology have been absorbed into a very different kind of company — one whose core business is a fully managed cloud data platform.
The result of this acquisition is a new Snowflake product called Snowflake Postgres, which reached General Availability on February 24, 2026. Snowflake's stated ambition is clear: to bring PostgreSQL workloads into its AI Data Cloud".
For organisations that have invested in Crunchy Postgres — whether through Crunchy Bridge managed cloud services or Crunchy's on-premise and Kubernetes offerings — this acquisition has significant implications. Snowflake's strategy is oriented towards cloud consolidation, and the long-term roadmap for on-premise Crunchy deployments has become increasingly uncertain. This article sets out to explain what has changed, what your options are, and why an alternative called PostgresPURE from Splendid Data deserves your serious consideration.
Limited support timelines ? While community and industry discussions have raised concerns about the future of Crunchy's on-premise products since the acquisition, no formal, publicly confirmed end-of-support date has been announced for existing Crunchy on-premise customers at the time of writing. We have heard from a Crunchy customer that December 2026 is the deadline for Support, but we cannot formally verify that applies to all Crunchy customers. What is clear is that Snowflake's product investment and strategic direction is firmly toward the cloud. Organisations dependent on Crunchy's on-premise capabilities are encouraged to begin planning their migration strategy now, rather than waiting for a formal announcement.
To understand the implications of this acquisition, it helps to understand Snowflake's business model and strategic motivations.
Snowflake is not a database company in the traditional sense. It is a cloud-native, fully managed data platform — a Software-as-a-Service (SaaS) offering that operates exclusively within public cloud environments (AWS, Azure, and GCP). Every workload that runs on Snowflake runs in Snowflake's cloud. There is no on-premise version of Snowflake, and there is no hybrid model where you run Snowflake's technology on your own servers.
This is the fundamental tension for existing Crunchy customers who operate on-premise or who have chosen Crunchy specifically to avoid vendor lock-in. Snowflake's acquisition of Crunchy Data was not motivated by a desire to support on-premise PostgreSQL deployments. It was motivated by the desire to capture transactional workloads and bring them into the Snowflake ecosystem alongside analytical workloads. As Snowflake's own SVP of Engineering stated at the time: "We're tackling a massive $350 billion market opportunity and a real need for our customers to bring Postgres to the Snowflake AI Data Cloud."
For regulated industries — including government, finance, and healthcare — this raises important questions about data sovereignty, compliance, and the cost of being locked into a single vendor's cloud platform for mission-critical transactional data.
Before comparing the two products, it is important to establish a foundational distinction: Snowflake Postgres (formerly Crunchy) is a fully managed Database-as-a-Service (DBaaS), while PostgresPURE is an enterprise software distribution that you or your partner install and manage on infrastructure of your choosing. These are not directly equivalent products — they represent fundamentally different philosophies about who controls your data and how.
Snowflake Postgres is a cloud-native managed PostgreSQL service that runs entirely within Snowflake's AI Data Cloud. It is designed for organisations that want a zero-administration PostgreSQL experience tightly integrated with Snowflake's analytics, governance, and AI capabilities.
Key characteristics of Snowflake Postgres include:
Snowflake Postgres is well-suited to organisations that want to consolidate both transactional and analytical workloads in a single managed cloud platform, and for whom Snowflake's pricing, compliance posture, and cloud-only model are acceptable.
PostgresPURE is a product maintained by Splendid Data, a Netherlands-based company specialising in enterprise PostgreSQL solutions. The product is best understood as a curated, tested, enterprise-grade distribution of PostgreSQL and its most critical open source companion tools — all bundled, validated, and supported under a single subscription.
The philosophy behind PostgresPURE is straightforward: PostgreSQL is the world's most advanced open source relational database, but the broader ecosystem of tools required for a true enterprise deployment — backup, replication, connection pooling, auditing, partitioning, monitoring, and more — exists as a fragmented collection of independently maintained open source projects. Each of these projects has its own release cycle, its own version compatibility matrix, and its own potential incompatibilities with specific versions of PostgreSQL.
PostgresPURE solves this problem by doing what Red Hat did for Linux: taking the best open source components, testing them together against each release of PostgreSQL, validating compatibility, and delivering a coherent, supported distribution with a predictable release cycle and 24/7 enterprise support. Just as Red Hat Enterprise Linux transformed community Linux into a platform that regulated industries and Fortune 500 companies could stake their operations on, PostgresPURE does the same for the PostgreSQL ecosystem.
PostgresPURE can be deployed on:
Supported operating systems include Red Hat Enterprise Linux, CentOS/Rocky, Oracle Linux, and SUSE. The distribution follows the PostgreSQL community's release schedule, with new versions of PostgresPURE released st least twice per year.
PostgresPURE is not just PostgreSQL with a label on it. It is a carefully assembled collection of proven open source tools, each selected for its production-grade reliability and tested for compatibility with the bundled PostgreSQL version. In essence, you pay the PostgresPURE subscription for the release management, compatibility testing, and 24/7 support — the underlying components are all open source and free.
The current PostgresPURE distribution includes:
This is the breadth and depth that enterprises need — not just a database, but the full operational toolkit. And critically, every component in this list has been tested against the specific version of PostgreSQL in that release of PostgresPURE, eliminating the compatibility uncertainty that arises when teams independently assemble their own open source stack.
Disaster Recovery in PostgresPURE is anchored by repmgr (Replication Manager), one of the most widely used and trusted open source tools in the PostgreSQL ecosystem.
repmgr is an open source tool that has provided advanced support for PostgreSQL's built-in streaming replication since version 9.0, and the current repmgr 5 series supports all modern PostgreSQL releases.
repmgr operates on a "Shared Nothing" architecture — each node in a cluster has its own dedicated storage, and replication occurs via PostgreSQL's physical streaming replication over the network. This architecture gives database administrators full visibility and control over the replication process. Key capabilities include:
The combination of repmgr with PostgreSQL's native streaming replication delivers a production-grade DR cluster that is entirely transparent and auditable. There are no black boxes — the DBA team can observe replication lag in real time, test failover in staging, and have full confidence in the behaviour of the system under failure conditions.
The table below contrasts the disaster recovery model of Snowflake Postgres with PostgresPURE using repmgr:
For organisations in regulated sectors — finance, government, healthcare, critical infrastructure, or organisations that just want transparency and certainty — the ability to conduct independent DR tests, produce audit evidence, and control the timing of failover events is not optional. It is a compliance requirement. PostgresPURE with repmgr delivers exactly that level of operational control.
To appreciate the value PostgresPURE brings, consider what Red Hat did for the Linux ecosystem in the 1990s and 2000s.
Linux was freely available as a community operating system, but enterprises that needed 10-year lifecycle support, certified configurations, predictable patch cadence, and 24/7 emergency response from a vendor they could hold accountable were underserved by the community model. Red Hat's solution was not to create a proprietary operating system — it was to take the same open source Linux kernel and ecosystem, bundle it into a tested and validated distribution called Red Hat Enterprise Linux (RHEL), commit to a long-term support lifecycle, and provide enterprise-grade support backed by SLAs.
Red Hat did not own Linux. But it owned the expertise, the testing rigour, and the support model — and that was exactly what enterprises needed. The code remained open source; the value-add was the confidence that came with a professionally maintained distribution.
PostgresPURE is doing precisely this for PostgreSQL. The PostgreSQL community produces excellent software, and so does the broader ecosystem of companion tools. But no single enterprise team should be expected to independently track, evaluate, test, and validate the compatibility of fourteen or more open source components across every PostgreSQL major and minor version. Splendid Data does that work, publishes the results as a validated release, and stands behind it with 24/7 support.
This is the model that enterprises — and particularly the decision-makers responsible for database reliability, compliance, and continuity — have always needed from their software vendors. Not lock-in. Not proprietary code. Just a trusted, professionally maintained distribution with real accountability.
The following table provides a consolidated comparison of Snowflake Postgres (based on Crunchy Data technology) and PostgresPURE across the dimensions that matter most to database managers, application managers, and data owners:
One of the most practical and reassuring aspects of this situation is that migrating from Crunchy PostgreSQL to PostgresPURE is entirely straightforward. Because both products are built on community PostgreSQL, your schemas, stored procedures, triggers, and data are fully compatible — there is no proprietary schema format, no binary translation, and no application rewrite required.
The migration process follows standard PostgreSQL migration principles that can be performed by your team or the PostgreSQL team at Pebble IT:
Step 1 — Install PostgresPURE
Provision your target environment — whether that is an on-premise server, a cloud VM on AWS, Azure, or GCP, or a virtualised environment. Install PostgresPURE using Splendid Data's repository and subscription package. PostgresPURE installs cleanly on RHEL, CentOS/Rocky, Oracle Linux, and SUSE.
Step 2 — Export Your Existing Data
Use pg_dump to export your existing Crunchy PostgreSQL database. This standard PostgreSQL utility creates a complete, portable dump of your schema and data:
pg_dump -Fc -h runchy_host> -U <username> -d <database_name> \ -f /backup/mydb_export.dumpBecause both environments use community PostgreSQL under the hood, the dump format is fully portable between them.
Step 3 — Import into PostgresPURE
Use pg_restore to import the exported data into your new PostgresPURE instance:
pg_restore -h <postgrespure_host> -U <username> -d <new_database> \ -j 4 /backup/mydb_export.dumpThe -j 4 flag enables parallel restore across four workers, significantly reducing import time for large databases.
Step 4 — Validate Extensions and Configuration
Review your extension requirements. Crunchy and PostgresPURE may bundle different versions of community tools (e.g., PostGIS, pgAudit), and some Crunchy-specific tooling may not have a direct equivalent. Validate that all required extensions are present and correctly configured in the new environment before cutting over.
Step 5 — Configure repmgr for Disaster Recovery
Once the primary node is operational, configure repmgr to establish your standby server(s) for streaming replication. Pebble IT's team can assist with this configuration as part of a managed migration engagement.
Step 6 — Cut Over
Perform a final incremental sync, redirect your application connection strings to the new PostgresPURE host, and confirm operational status. With a properly configured repmgr cluster, you can also perform a graceful planned switchover to minimise any downtime window.
This process is well understood, low-risk, and entirely within the capabilities of any experienced PostgreSQL DBA team. Pebble IT can provide advisory and hands-on support throughout.
While repmgr provides a robust and highly capable foundation for high availability in PostgresPURE environments, organisations with stringent RTO/RPO requirements — particularly those in regulated industries such as banking, insurance, government, or healthcare — may benefit from adding Dbvisit Standby MultiPlatform (StandbyMP) to their PostgresPURE deployment.
Dbvisit StandbyMP is a dedicated disaster recovery management platform that adds an enterprise-grade control layer on top of PostgreSQL's native streaming replication.
For PostgresPURE deployments, DBvisit StandbyMP provides:
The table below compares Snowflake Postgres (Business Critical tier) with a PostgresPURE + DBvisit StandbyMP configuration:
The key differentiator is not raw performance — both configurations deliver sub-minute recovery for most failure scenarios. The differentiator is control, transparency, and auditability. For organisations that must demonstrate to a regulator or auditor that they have successfully tested their disaster recovery procedures, a platform that does not allow you to observe or initiate a failover may cause issues with regulatory requirements as well as internal audit requirements. PCI DSS 4.0 requirements as of March 2025 includes:
PostgresPURE with DBvisit StandbyMP gives your team the operational evidence trail that compliance demands.
DBvisit is an independent software company with deep expertise in database disaster recovery across Oracle, SQL Server, and now PostgreSQL — giving PostgresPURE deployments access to a purpose-built DR toolset with dedicated vendor support.
For key stakeholders including database managers, application owners, and data stewards evaluating their options after the Crunchy acquisition, PostgresPURE offers a compelling, risk-managed path forward. The following benefits are the most strategically significant:
1. No Vendor Lock-In
Every component in PostgresPURE is 100% open source. There are no proprietary hooks, no closed-source extensions, and no dependency on a single vendor's cloud platform. Your data stays where you deploy it, under your control.
2. Freedom of Deployment
Deploy on-premise, in a private cloud, or on any public cloud of your choosing. Unlike Snowflake Postgres, PostgresPURE works on AWS, Azure, GCP, OCI, IBM Cloud, or on bare metal in your own data centre — you are never forced into a single provider.
3. Predictable, Tested Releases
Splendid Data releases a validated PostgresPURE component stack against it before publishing. The twice per year release cadence gives organisations a reliable, predictable upgrade path without the risk of self-assembled component incompatibilities.
4. Enterprise Support with SLAs
A PostgresPURE subscription provides access to 24/7 SLA-based support from Splendid Data, with assistance available via online ticketing or telephone depending on severity. Pebble IT, as the authorised Australian and Oceania partner, provides locally available expertise for Australian and Asia-Pacific organisations.
5. Full Operational Control
Unlike a managed cloud database, PostgresPURE gives your DBA team complete access to the operating system, PostgreSQL configuration, replication state, and backup processes. You decide when to upgrade, when to patch, and how to configure the system for your specific workload.
6. Data Sovereignty and Compliance
For organisations subject to data residency requirements under the Australian Privacy Act, GDPR, HIPAA, or sector-specific regulations, running PostgresPURE on your own infrastructure means your data never leaves your control boundary. You hold the encryption keys, you define the network perimeter, and you produce the compliance evidence.
7. Cost Predictability
Snowflake's consumption-based pricing scales with every query and every byte processed. For organisations with steady, high-volume transactional workloads, this creates budget unpredictability. A PostgresPURE subscription provides a consistent, predictable cost structure that does not scale linearly with query volume.
8. The Red Hat Model — Tested Together, Supported as One
The core value proposition of PostgresPURE is that you do not have to manage the compatibility matrix yourself. Splendid Data has done that work — tested fourteen-plus open source components together, validated the combination, and published the result as a supported release. This is precisely the model that made Red Hat Enterprise Linux the default choice for enterprise Linux deployments, and it is equally compelling for enterprise PostgreSQL.
PostgresPURE is particularly well-suited to the following types of organisations:
Snowflake Postgres is for organisations that want zero administrative overhead and are comfortable consolidating all of their data workloads into the Snowflake ecosystem. If that is not you, then PostgresPURE is likely to be a preferred choice.
Snowflake to acquire database startup Crunchy Data | TechCrunch - Cloud data platform Snowflake has announced its intent to acquire Crunchy Data
Support for Crunchy Hardened PostgreSQL Ends Soon - Percona - Recent community discussions and rumours suggest that Crunchy Hardened PostgreSQL may reach end of support
PostgresPURE by Splendid Data – Open Source ready for Enterprises
DBVisit Standby for PostgreSQL – Gold Standard Disaster Recovery for PostgreSQL
