PicoMQ

What it does

PicoMQ is a server for durable message streams accessed over HTTP, using S3-compatible object storage as the underlying storage layer. It supports both the Pico Protocol and the Durable Streams Protocol. Streams can be created, appended to, read, long-polled, or accessed via server-sent events. A Postgres command log handles coordination between nodes.

Who it is for

Developers building systems that need persistent message queues but want to avoid managing dedicated infrastructure. Teams using object storage like AWS S3 already. Projects where traditional message brokers feel over-provisioned or where the cost of maintaining separate Kafka or RabbitMQ clusters is a constraint.

Pricing

The site does not show prices.

How it stands out

PicoMQ decouples storage from compute by building on object storage instead of local disk, following patterns used in AutoMQ. Each stream is independently addressable and can scale independently—a team can create separate streams per use case rather than packing all records of a kind into one topic. The architecture claims zero-disk operation and supports HTTP as the primary interface, which differs from Kafka's binary protocol. Using Postgres for coordination is simpler than running Zookeeper or etcd.

What a founder should check

First, understand the switching cost from Kafka or other message brokers. Does PicoMQ require rewriting producer and consumer code, or does it offer Kafka-compatible APIs that let teams plug it in without major refactoring? Check what tooling and client libraries exist beyond HTTP.

Second, verify the true operational overhead. The pitch emphasizes simplicity, but confirming the actual deployment complexity—container management, monitoring, debugging—versus running a managed Kafka service matters for the actual cost story.

Third, test performance under the buyer's real workload. The site claims high throughput and independent scaling, but latency, tail percentiles, and how performance degrades with many concurrent streams need validation. Compare object storage API call patterns and costs against the vendor's assumptions.