Apache Pinot Virtual Meetup with Uber

Apache Pinot is an open source, distributed columnar database for real-time analytics, widely used in Uber, LinkedIn and many other companies to power realtime business usecases. Please join us for a virtual meetup hosted by Uber and Apache Pinot community. We will kick off with an update on Apache Pinot 0.5.0 release followed by the interesting talks by the speakers from Uber, Confluera and City Storage Systems.

Agenda
5:00 PM - Welcome/ Intro
5:05 PM - What’s new in Apache 0.5.0 ?
5:10 PM - Pinot Realtime Ingestion with Cloud-based Deep Storage
5:35 PM - Powering analytics and search at Confluera using Pinot
6:00 PM - Pinot at City Storage Systems
6:25 PM - Closing Remarks

Talks:

1. Pinot Realtime Ingestion with Cloud-based Deep Storage- Ting Chen (Uber)

In this talk, we will share Uber's experience on operating and improving Pinot's latest real-time ingestion method: the Low Level Consumer (LLC) over the past two years. After a quick review of the method, the talk will first present how we added a deepstore based on HDFS to resolve the Pinot controller space limit issue. Next to satisfy Uber's needs for non-stopping real-time data ingestion, we will discuss a redesign of the segment completion distributed protocol in LLC so that the real-time ingestion can still proceed even if the deep store is not available.

Speaker Bio: Ting Chen is a software engineer in Uber's Data team. He is a tech lead on Realtime Analytics team whose mission is to provide fast and reliable real-time insights to Uber products and customers. Ting is an Apache Pinot committer.

2. Powering analytics and search at Confluera using Pinot- Pradeep G.V. (Confluera)

In this talk, we will give a brief overview of how Pinot serves the
Search and Analytics needs at Confluera, our deployment experience
with confluera and how pinot has led to rapid experimentation with
data in coming up with new features and further aided in security
Investigations.

Speaker Bio: Pradeep is a member of technical staff at Confluera, where he takes care of most things related to data such as search, analytics, storage, anomaly detection. Prior to this he worked at Google on recommendation systems & ads infrastructure. His primary interests span around distributed systems, but at the same time finds most things in large software systems fascinating.

3. Pinot at City Storage Systems- Elon Azoulay & Alex Filipchik (City Storage Systems)

Pinot is an emerging technology that is built to provide real-time insights at internet scale. While jumping into something new is always exciting, some might wonder: how hard is it to run Pinot in production, does it need a large team to support it, how does it fit into existing architecture? Join us to learn how we run Pinot on Kubernetes in the Cloud. We will also talk about how other parts of the stack suck as Presto, HUDI, Flink, and Istio are orchestrated to get the most out of it.

Speaker Bios:
Elon:I grew up on the east coast, graduated from Rutgers with a CS degree, worked in New York City as a DBA, always wanting to be a part of creating a better database. In 2014 I got that opportunity, working at Facebook on the core Presto Team. I joined City Storage Systems to help build out the data infrastructure. This is where I discovered Pinot, which we quickly fell in love with. Besides working on databases, I enjoy biking, going to the beach with my family, and playing classical guitar.

Alex: I spent the last 10 years tinkering with distributed systems at scale and even participated in the launch of a gaming console. Throughout the years I've slowly moved from working on products that users touch to building data infrastructure that powers those products. When I'm not thinking about the data world you can find me snowboarding, enjoying water sports, or hiking with my family.

https://uber.zoom.us/j/92749891719?pwd=a2daZUs0cjRXZmJOcnFPc1RXamFlZz09
Passcode: 541532