COSS Community

Cover image for OCS 2020 Breakout: Dhruba Borthakur and Nadine Farah
Joseph (JJ) Jacks for COSS Community

Posted on • Updated on

OCS 2020 Breakout: Dhruba Borthakur and Nadine Farah

Dhruba Borthakur is the CTO and Co-founder of Rockset. He was an engineer on the database team at Facebook, where he was the founding engineer of the RocksDB data store. Earlier at Yahoo, he was one of the founding engineers of the Hadoop Distributed File System.

Relevant Links
LinkedIn - Twitter

Nadine Farah is a senior developer advocate at Rockset leading Rockset’s developer initiatives. Prior to Rockset, she was at Bose working with the Bose AR team. There, she was a senior developer advocate and an engineer helping build Bose's BLE SDK for watchOS.

Relevant Links
LinkedIn - Twitter

Rockset co-founder Dhruba Borthakur and Nadine Farah, senior developer advocate, discuss how to build serverless analytics on DynamoDB with Rockset.

Introductions and presentation topic: Empowering application developers to build serverless analytics with DynamoDB on AWS - 0:00

High-level overview of Rockset. Built for Real-time Analytics in the Cloud - 0:49

What are Real-time Apps and Analytics? - 2:40

Presentation Agenda - 3:54

Why serverless real-time analytics? - 4:22

DynamoDB + Rockset = The perfect PB&J sandwich - 5:05

Real-time Analytics in Gaming. Egogames Tech Stack Overview - 6:28

Serverless storage with RocksDB-Cloud -7:36

Rockset is the serverless realtime indexing database - 8:07

Walking through Aggregator-Leaf-Tailer Architecture (see: https://rockset.com/blog/aggregator-leaf-tailer-an-architecture-for-live-analytics-on-event-streams/) - 8:50

Why RocksDB is used for storing data in the Leaf Nodes (see: https://github.com/rockset/rocksdb-cloud) - 13:50

RocksDB Architecture - 14:38

RocksDB Serverless Storage with Amazon S3. Exploring architecture. - 16:25

Offloading compaction CPU with RocksDB copy clone feature - 19:30

Continuous indexing of data in DynamoDB - 22:47

Delicate Dance between the Scan API and Streams API - 24:54

Step 1: Bulk Load data into Rockset from DynamoDB - 25:55

RCU consumption - 26:47

Step 2: Stream Data into Rockset from DynamoDB - 27:14

Continuous Polling - 27:45

Diving into demo: Using Rockset and DynamoDB - 28:17

Concluding remarks and contacts - 36:17

*** -


Share your questions and comments below!

Discussion (0)