Register: [ Ссылка ] | Watch Vaibhav Maheshwari (LinkedIn) present at Real-Time Analytics Summit 2022, our event filmed exclusively in cinematic, high-quality video. Streaming Infrastructure at LinkedIn heavily relies on Apache Kafka and leverages Brooklin Mirror Maker (BMM) to replicate more than seven trillion messages per day between Kafka clusters. Variable throughput rate across the topic partitions contributes to significant challenges in determining the hardware resources required for replication and even distribution of the replication load across all the BMM Instances.

This talk will cover how we have built a Partition Metrics Provider(PMP) service using Apache Pinot to ingest partition-level throughput metrics provided by Cruise Control in real-time. We will dive deeper into Pinot's optimization features that enabled PMP service and how BMM leverages this information to dynamically scale the clusters, perform load-balanced replication, and reduce operating costs. This talk will also cover how we use dashboards powered by Apache Pinot to identify offending topic partitions and simplify the debugging experience.

Join: [ Ссылка ] | Sub: [ Ссылка ] |

LEARN MORE
► Register for free at [ Ссылка ]
► Learn about Apache Pinot: [ Ссылка ]

CONNECT
Subscribe on YouTube: [ Ссылка ]
Community Slack: [ Ссылка ]
Twitter: [ Ссылка ]
Linkedin: [ Ссылка ]
GitHub: [ Ссылка ]
Site: [ Ссылка ]

ABOUT STARTREE
When you hear “decision maker,” it's natural to think, “C-suite,” or “executive.” But these days, we’re all decision-makers. So, yes, the CEO is a crucial decider, but the franchise owner, the student, or the big box shopper all have important choices to make, too. The difference is, those traditional decision-makers have access to relevant data to guide their thinking. All the others? Flying blind.

Our vision is to change that. We believe every decision-maker should have access to fast, fresh, actionable insights. We’re motivated to unlock what’s possible when you put the right information in the right hands at the right time. Like the restaurant owner who’s able to call in an extra cook because she can see the surge in orders coming. Or the holiday shopper who has plenty of time to pick a different gift for his loved one because he sees his original order is delayed at the moment the supplier registers a procurement delay. Or the use-case we can’t even envision yet, because it is sitting in your data, waiting to be set free.

Exposing timely information to real decision makers in intuitive apps that not only inform, but allow you to act on the information being shared. That’s how we’re unleashing the power of user-facing analytics.

#apachepinot #realtimeanalytics #startree

свернуть