Rust in Production

In the season premier we talk to none other than Daniel Stenberg! We focus on integrating Rust modules in curl, their benefits, ways in which Rust and Rust crates helped improve curl, but also how curl helped those crates, and where curl is used in the official Rust toolchain. Along the way we also learn about the early history of curl and Rust, which section of your car's owner's-manual you should "re"-read, some weird HTTP edge-cases, and Daniel's experience in open-source maintainership. And don't forget: have fun!

Imagine you're faced with the challenge to build a system that can handle billions of recyclable items per year with the goal of being better than a human at identifying what can be recycled. Material classification is a complex problem that requires a lot of data and a lot of processing power and it is a cutting-edge field of research. Carters Schultz and his colleagues at AMP chose Rust to build the core of this system -- and it worked "shockingly well". In this interview, Carter, charismatic and witty, shares his experience of getting your hands dirty with Rust, and building a cutting-edge production-ready system, which can be now replicated across the world.

This is Rust in Production, a podcast about companies who use Rust to shape the future of infrastructure. We follow their journey in pursuit of more reliable and efficient software as they solve some of the most challenging technical problems in the world. I'm your host, Matthias Endler, and I'm a software engineer at corrode, a consultancy that helps companies make the most of Rust. I've been using Rust since 2015, have been a member of the Rust Cologne meetup since Rust 1.0 and ran a YouTube channel called "Hello Rust". There are plenty of great podcasts about Rust, but I felt that there was a missing piece. I wanted to hear more about how companies who use Rust in production. What are the challenges they face? How do they overcome them? What are the benefits of using Rust? How does the company find and hire Rust developers? And what advice would they give to other companies who want to use Rust. I sit down with decision-makers from companies that bet big on Rust and ask them in-depth questions about what they learned along the way.  New episodes air every two weeks on Thursdays at 4pm UTC. If you don't want to miss out, please subscribe on Spotify, Apple Podcasts, or wherever you listen to podcasts. This helps other people find the show and supports our work. If you want to learn more about the show, please visit corrode.dev/podcast. Stay tuned for the first episode, where I talk to Paul Dix from InfluxData about how they use Rust in the latest version of InfluxDB.

For our very first episode, we welcome a special guest, Paul Dix, the CTO of InfluxData. He starts by giving us an overview of InfluxDB, an open source time series database used by developers to track server and application data. He takes us back to the early days of InfluxDB and explains how it came into existence, starting with the challenges they faced with their initial SaaS application and how they made the decision to repurpose their infrastructure and create this open source database. Paul also sheds light on the popularity of the programming language Go, which had a significant influence on their decision to use it for their project. He takes us through the journey of InfluxDB's development and the improvements that have been made over the years. He emphasizes the enhancements made in versions 0.11 and 1.0 to improve performance and query capabilities. Moreover, he shares their decision to explore using Rust for certain parts of the project and the positive impact it has had. Moving forward, the conversation delves into the challenges of managing high volumes of data in time series databases. Paul talks about the solutions they implemented, such as using BoltDB and developing the time-structured merge tree storage engine. We then dive into the decision to rewrite InfluxDB in Rust and the benefits it offers. He explains the improved performance, concurrency, and error handling that Rust brings to the table. Paul goes on to discuss the development process and how the engineering team has embraced Rust across their projects. As the conversation progresses, we tou...

In this episode, we are joined by Steven, the CTO of PubNub, a company that has developed an edge net messaging network with over a billion connected devices. Steven explains that while message buses like Kafka or RabbitMQ are suitable for smaller scales, PubNub focuses on the challenges of connecting mobile devices and laptops at a web scale. They aim to provide instant signal delivery at a massive scale, prioritizing low latency for a seamless user experience. To achieve this, PubNub has architected their system to be globally distributed, running on AWS with Kubernetes clusters spread across all of Amazon's zones. They utilize GeoDNS to ensure users connect to the closest region for the lowest latency possible. Steven goes on to discuss the challenges they faced in building their system, particularly in terms of memory management and cleanup. They had to deal with issues such as segmentation faults and memory leaks, which caused runtime problems, outages, and potential data loss. PubNub had to invest in additional memory to compensate for these leaks and spend time finding and fixing the problems. While C was efficient, it came with significant engineering costs. As a solution, PubNub started adopting Rust, which helped alleviate some of these challenges. When they replaced a service with Rust, they observed a 5x improvement in memory and performance. Steven also talks about choosing programming languages for their platform and the difficulties in finding and retaining C experts. They didn't consider Java due to its perceived academic nature, and Go didn't make the li...

In this episode, Nicolas, a staff software engineer at Apollo GraphQL, discusses the company's use of GraphQL API technologies. Apollo GraphQL specializes in open-source libraries for both client and server-side applications, with a focus on integrating Rust into their main offerings: the Apollo router and GraphOS cloud. Nicolas explains how the Apollo router consolidates multiple microservices into a single API, efficiently routing requests to appropriate services. He delves into GraphQL's role as an effective query language for APIs, highlighting its ability to provide a comprehensive description of API data and its compatibility with existing data systems. The shift from the JavaScript-based Apollo Gateway to the Rust-based Apollo Router is a key topic, with an emphasis on the performance and safety improvements Rust brings to the table. The conversation covers the use of Rust for the router and GraphQL parser, alongside Kotlin for the management plane and GraphQL for the API. Challenges in stability and reliability are discussed, as well as Rust's advantages in safety and type system consistency. Nicolas shares insights on Async Rust, particularly its impact on productivity and application in CLI tools like Rover. The episode also addresses learning Rust in stages, from basic language concepts to advanced internal mechanisms. It touches on functional patterns in Rust and strategies for effective dependency management. Closing the discussion, Nicolas highlights the inclusive and supportive nature of the Rust community.

In this episode, we have Micah Wylde from Arroyo as our guest. Micah introduces us to Arroyo, a real-time data processing engine that simplifies stream processing for data engineers using Rust. They explain how Arroyo enables users to write SQL queries with Rust user-defined functions on top of streaming data, highlighting the advantages of real-time data processing and discussing the challenges posed by competitors like Apache Flink. Moving on, we dive into the use of Rust in Arroyo and its benefits in terms of performance and memory safety. We explore the complementarity of workflow engines and stream processors and examine Arroyo's approach to real-time SQL and its compatibility with Postgres. Micah delves into memory and lifetime concerns and elaborates on how Arroyo manages them in its storage layer. Shifting gears, we explore the use of the Tokyo framework in the Arroyo system and how it has enhanced speed and efficiency. Micah shares insights into the challenges and advantages of utilizing Rust, drawing from their experiences with Arroyo projects. Looking ahead, we discuss the future of the Rust ecosystem, addressing the current state of the Rust core and standard library, as well as the challenges of interacting with other languages using FFI or dynamically loading code. We touch upon Rust's limitations regarding a stable ABI and explore potential solutions like WebAssembly. We also touch upon industry perceptions of Rust, investor perspectives, and the hiring process for Rust engineers. The conversation takes us through the crates used in the Arroyo system, our ...

Have you ever wondered how computers stay in sync with the time? That is the responsibility of the Network Time Protocol (NTP). Around since 1985, NTP is one of the oldest protocols still in use on the internet and its reference implementation, ntpd, written in C, is still widely used today. That's a problem. C is a language that is not memory safe, and ntpd has had its share of security vulnerabilities. Here is a list of CVEs. As part of Project Pendulum, Folkert de Vries and his colleagues from Tweede Golf have been working on a Rust implementation of NTP. I sit down with Folkert to talk about the project, the challenges they faced, and the benefits of using Rust for this kind of project. Along the way, we learn about funding open source projects, the challenges of establishing a new implementation of a protocol, and all sorts of other interesting things that might or might not be related to NTP.

In an ever-expanding world of microservices, APIs, and devices, maintaining an overview of application states and the myriad errors that can occur is challenging. For years, Sentry has been a go-to choice for developers to monitor their applications and receive notifications about issues within their code. Traditionally, Sentry was predominantly a Python shop, but they became one of the early adopters of Rust in their technology stack. They have been utilizing Rust for a few years now (since at least 2017), starting with sentry-cli, a command-line utility to work with Sentry, and continuing with their source map parsing pipeline, which reduced processing times from 20 seconds to less than 500 milliseconds. More recently, they have been developing two new projects in Rust: relay and symbolicator. Relay acts as a proxy for sending events to Sentry, while symbolicator is a service for handling the symbolication of stack traces. Both projects are open source and available on GitHub. Arpad Borsos (swatinem), Senior Native Platform Engineer at Sentry, sat down with me to discuss their journey with Rust and how it has enabled them to build a cutting-edge monitoring platform for large-scale applications. Our conversation covered topics such as 100x speedups, the Rust-Python interface, and the comparison between actor-based and task-based concurrency.

In this episode of Rust in Production, we reflect on the inaugural season, sharing insights from companies pioneering with Rust to shape the future of infrastructure. Starting with Rust's early days before the stable compiler, we chart the language's evolution and its growing adoption in production by leading companies. The episode explores why companies choose Rust, highlighting its unparalleled safety features, concurrency management, and performance that not only drives down costs but significantly enhances user experience. The narrative then shifts to the practicalities of adopting Rust, including learning resources, the importance of community support, and integrating Rust into existing codebases. We discuss the surprisingly positive experience of hiring Rust developers, thanks to the language's appealing prospects. Moreover, the episode delves into Rust's explicit approach to handling complexity, offering a steeper yet rewarding learning curve. We also address the misconception of Rust's limited adoption, revealing its widespread use across various industries and the consensus around certain Rust patterns and libraries. Despite the challenges in showcasing Rust's adoption due to the reticence of some companies to share their stories, the positive feedback from the podcast's audience underscores a keen interest in Rust's real-world applications. Concluding the season, we extend a heartfelt thanks to all guests and listeners, emphasizing the value of sharing Rust experiences and the potential for further growth in the Rust ecosystem. We share a wishlist for the Rust ...