Aggregate Stream Optimizations

Jan 29, 2018

We have made an update to both real-time aggregate data streams:

A.* - Per Second Aggregates
AM.* - Per Minute Aggregates


Before, we would broadcast all symbols, every second/minute, regardless if a trade happened in that timespan. This was easily determined by the

"v": 0
meaning there was no volume.

This means it would broadcast ~8000 ticks/second on the per second aggregate channel.

The Change:

We have changed both channels to only broadcast updates on the tickers which have actually had trades in that timespan. Which means the average is ~1000 ticks/second instead of ~8000. This allows better throughput, and less processing messages of tickers that have not changed.

From the blog

See what's happening at

case study fey Feature Image

Case Study: Fey

Discover how Fey, a rising star in fintech, transformed their financial application's capabilities with's API services. Dive into our latest case study to see how we're powering innovation in finance.