Postgres backups are at an inflection point. We'll examine tooling changes, gaps in core Postgres, and debate which backup features should live where
The Postgres backup and restore ecosystem sits at a critical juncture these days. For years, we relied on less than a handful of mature tools, but with the recent shifts in tool lifecycles, evolving community needs, and gaps in core Postgres functionality, we are forced to rethink how we protect our data.
In this talk, we'll examine the current state of the Postgres backup landscape—where we are, what's working, and where we can improve. We'll look at recent changes in tooling that have surprised the community. These aren't just technical events; they signal deeper tensions in how Postgres distributes responsibilities between core and tooling.
# The Core Question: What Belongs Where?
One of the central debates we face is: which backup features should live in core Postgres, and which should remain in the external backup tooling? Postgres has some core backup capabilities, but significant gaps remain—features that users expect, but have not landed in Postgres, and so need to be developed in the tooling space.
We'll explore specific examples: What is still missing with respect to incremental backup support in core? How should parallel backup be handled? Where does encryption fit? Should Postgres have capabilities for restoring backups? For each, we'll discuss the trade-offs: maintenance burden, standardization, and user expectations.
# The Ecosystem Perspective
The backup ecosystem around Postgres is rich and diverse. This variety is a strength, as it lets users choose solutions that match their needs. But it also creates fragmentation: incompatible approaches, duplicated effort, and a learning curve for people navigating the different options.
We'll map the ecosystem: who's building what, where they overlap, where they complement each other, and where gaps still exist. We'll also discuss the role of the Postgres core development in thriving these projects.
# Looking Ahead: The Next Few Years
What does the future look like? We'll discuss trends we're already seeing and predict where the ecosystem is heading. Will more backup functionality migrate into core? Will specialized backup tools become more focused and modular? How will cloud-native patterns and managed Postgres services influence backup strategies? What role will emerging technologies like incremental and snapshot backup and recovery play?
We'll also touch on the broader implications: how lifecycle changes in key tools affect operations teams, what users should do to future-proof their backup strategies, and how the community can better coordinate between core development and tooling.
# Takeaways
By the end of this talk, I expect to have delivered a clearer picture of:
* The current state of Postgres backup tooling (core and surrounding) and how recent changes fit into the larger story
* The upright debate about which features belong in core vs. tooling, and why those decisions matter
* A realistic view of where the ecosystem is heading
* Concrete ways to think about backup architecture decisions in your own environment
This talk is for DBAs, architects, and engineers who manage Postgres and want to understand not just the "how" of backups, but the "why" behind the ecosystem's structure and trajectory.
Present and Future of Postgres Backups
Martín Marqués
EnterpriseDB
Postgres backups are at an inflection point. We'll examine tooling changes, gaps in core Postgres, and debate which backup features should live where
The Postgres backup and restore ecosystem sits at a critical juncture these days. For years, we relied on less than a handful of mature tools, but with the recent shifts in tool lifecycles, evolving community needs, and gaps in core Postgres functionality, we are forced to rethink how we protect our data.
In this talk, we'll examine the current state of the Postgres backup landscape—where we are, what's working, and where we can improve. We'll look at recent changes in tooling that have surprised the community. These aren't just technical events; they signal deeper tensions in how Postgres distributes responsibilities between core and tooling.
# The Core Question: What Belongs Where?
One of the central debates we face is: which backup features should live in core Postgres, and which should remain in the external backup tooling? Postgres has some core backup capabilities, but significant gaps remain—features that users expect, but have not landed in Postgres, and so need to be developed in the tooling space.
We'll explore specific examples: What is still missing with respect to incremental backup support in core? How should parallel backup be handled? Where does encryption fit? Should Postgres have capabilities for restoring backups? For each, we'll discuss the trade-offs: maintenance burden, standardization, and user expectations.
# The Ecosystem Perspective
The backup ecosystem around Postgres is rich and diverse. This variety is a strength, as it lets users choose solutions that match their needs. But it also creates fragmentation: incompatible approaches, duplicated effort, and a learning curve for people navigating the different options.
We'll map the ecosystem: who's building what, where they overlap, where they complement each other, and where gaps still exist. We'll also discuss the role of the Postgres core development in thriving these projects.
# Looking Ahead: The Next Few Years
What does the future look like? We'll discuss trends we're already seeing and predict where the ecosystem is heading. Will more backup functionality migrate into core? Will specialized backup tools become more focused and modular? How will cloud-native patterns and managed Postgres services influence backup strategies? What role will emerging technologies like incremental and snapshot backup and recovery play?
We'll also touch on the broader implications: how lifecycle changes in key tools affect operations teams, what users should do to future-proof their backup strategies, and how the community can better coordinate between core development and tooling.
# Takeaways
By the end of this talk, I expect to have delivered a clearer picture of:
* The current state of Postgres backup tooling (core and surrounding) and how recent changes fit into the larger story
* The upright debate about which features belong in core vs. tooling, and why those decisions matter
* A realistic view of where the ecosystem is heading
* Concrete ways to think about backup architecture decisions in your own environment
This talk is for DBAs, architects, and engineers who manage Postgres and want to understand not just the "how" of backups, but the "why" behind the ecosystem's structure and trajectory.