Incremental Update 15

We’re happy to announce the release of feldera v0.35! Today we have another, smaller release since our last release just happened last week, but we include some exciting changes nevertheless.

Big Records

We fixed problems with ad-hoc queries in the web-console and fda when programs had individual rows which got really big (think megabytes). The fix for the ad-hoc queries in the text format is that we no longer display the full content for very large strings. But don't worry you can still retrieve the full content using json as the format.

Switching from YAML to JSON

Speaking of JSON, with this release we also changed the manager to use JSON instead of YAML as our internal config representation. This made sense since everything user-facing was already JSON and the serde-yaml crate is no longer maintained. It also makes it easier to guarantee that future upgrades of the platform remain frictionless.

What's next

We're working behind the scenes on some exciting upgrades to the Feldera platform, so stay tuned for future releases:

• Performance improvements: We're currently working on interning large objects (variant, maps, and arrays) which will make working with them much more efficient
• S3 support: We'll soon have the ability leverage object stores for pipeline storage. We will make an announcement about it in the next few weeks.