Speaker
Abstract
As a part of OneSignal’s no-code Journeys system, we knew that we would need a way to store billions of timers. OneSignal delivers over 11 billion notifications per day, and we would potentially need to store a timer event alongside each of these for customer-defined journeys like “send a notification, wait 24h, send another notification if a condition is met.” After searching the paid and open-source worlds for off-the-shelf solutions, we decided that our best option was to build a system that could store and expire timers for ourselves. Ingesting data via gRPC, persisting to Scylla, and expiring to Kafka via a Rust scheduler has proved to be a great solution for us. This talk walks through the design of this system, its performance characteristics, and how we’ve scaled it since it began as a single-node service for the first year and a half of its existence.
Topics
QCon San Francisco 2023 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.
From the same track
Monday 2 October
10:35 Ballroom A Session Graph Databases LIquid: A Large-Scale Relational Graph Database Scott Meyer Distinguished Software Engineer @LinkedIn, Creator of the Graph Database, LIquid, Metaweb/freebase Alum We describe LIquid(1 2), the graph database built to host LinkedIn. 11:45 Seacliff ABC Session Data PRQL: A Simple, Powerful, Pipelined SQL Replacement Aljaž Mur Eržen Compiler Developer @EdgeDB & PRQL Maintainer Most databases use SQL as the interface to access relational data. Because of that, we associate SQL to be the language of relational algebra. But its affinity with the English language and unclear and inconsistent semantics leave a lot of space for improvements. 13:35 Ballroom A Session Stream Processing Streaming Databases: Embracing the Convergence of Stream Processing and Databases Yingjun Wu Founder and CEO @RisingWave Labs, Previously Engineer @AWS Redshift & Researcher @IBM Research Almaden Streaming databases have gained significant attention in recent years. From its name, it is evident that a streaming database combines the power of stream processing and databases. 14:45 Ballroom A Session Distributed Systems Redesigning OLTP for a New Order of Magnitude Joran Greef Founder and CEO @TigerBeetle The world is becoming more transactional. From colocation and server rental to serverless and usage-based billing. From coal to clean energy and smart meters that arbitrage solar prices 1440 times a month instead of monthly. Not to mention FedNow or the tsunami of instant payments. 15:55 Ballroom A Session Data Lakes Incremental Data Processing with Apache Hudi Saketh Chintapalli, Bhavani Sudha Saktheeswaran Incremental Data Processing is an emerging style of data processing gathering attention recently that has the potential to deliver orders of magnitude speed and efficiency over traditional batch processing on data lakes and data warehouses. 17:05 Ballroom A Session Architecture Sleeping at Scale - Delivering 10k Timers per Second per Node with Rust, Tokio, Kafka, and Scylla Lily Mara, Hunter Laine As a part of OneSignal’s no-code Journeys system, we knew that we would need a way to store billions of timers.