justinc-yelp

3 ms·

1. 1.▲
   by justinc-yelp 10y ago

   I'm not actually a good resource on scaling Kafka. Our distributed systems teams do a great job of providing reliable infrastructure and scaling it up, so on the application side we are mostly able to treat it like a black box that ju
2. 2.▲
   by justinc-yelp 10y ago

   Most of the stream processing in the Data Pipeline happens inside of an internal project called PaaStorm, which is storm-like. It was built to take advantage of our platform as a service ( http://engineeringblog.yelp.com/201
3. 3.▲
   Billions of Messages a Day – Yelp's Real-Time Data Pipeline(engineeringblog.yelp.com)

   172 points by justinc-yelp 10y ago|47 comments