OUR PERSPECTIVE ON THINGS

The Buf blog

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
All
May 2, 2025
Team Buf
Bufstream

Cheap Kafka is cool. Schema-driven development with Kafka is cooler.

If you're looking for a modern drop-in replacement for Apache Kafka to save costs and complexity, Bufstream is probably your best bet. However, we've got a bigger mission here. Buf wants to bring schema-driven development across your entire stack, from your network APIs, to your streaming data, to your lakehouse, unified behind one schema language that can do it all.

All
April 29, 2025
Miguel Young de la Sota
Protobuf

Tip of the week #4: Accepting mistakes we can’t fix

Protobuf’s distributed nature introduces evolution risks that make it hard to fix some types of mistakes. Sometimes the best thing to do is to just let it be.

All
April 22, 2025
Miguel Young de la Sota
Protobuf

Tip of the week #3: Enum names need prefixes

Enums inherit some unfortunate behaviors from C++. Avoid this problem by using the Buf lint rules ENUM_VALUE_PREFIX and ENUM_ZERO_VALUE_SUFFIX.

All
April 15, 2025
Miguel Young de la Sota
Protobuf

Tip of the week #2: Compress your Protos!

Compression is everywhere. This pervasiveness means that wire size tradeoffs matter less than they used to twenty years ago, when Protobuf was designed.

All
April 8, 2025
Miguel Young de la Sota
Protobuf

Tip of the week #1: Field names are forever

Don’t rename fields. Even though there are a slim number of cases you can get away with it, it’s rarely worth doing, and is a potential source of bugs.

All
March 7, 2025
Team Buf
Bufstream

Multi-region, active-active Bufstream at 100 GiB/s

Bufstream — Buf’s drop-in replacement for Apache Kafka® — now supports business-critical workloads with multi-region, active-active clusters on GCP.

All
March 5, 2025
Team Buf
Bufstream

Bufstream on Spanner: 100 GiB/s with zero operational overhead

At less than $1M/month for 100 GiB/s of writes and 300 GiB/s of reads, Bufstream using Spanner is 25x cheaper than Confluent Cloud, 1.5x cheaper than WarpStream, and 3x cheaper than self-hosted Apache Kafka®.

Get started on the path to universal schema adoption

Across your entire data stack — from your network APIs to your streaming data to your data lake.

Get in touch
Read the docs