Big Ideas in App Architecture is a weekly podcast for architects and engineers building modern, data-intensive applications and systems. In each weekly episode, Big Ideas host David Joy invites innovators from across the tech universe to share useful insights from their experiences building scalable, maintainable, and above all reliable systems.
To help you start off strong in 2024, here are our top ten most valuable “a-ha” moments, so-crazy-it-just-might-work ideas, and technical takeaways from the past year of Big Ideas in App Architecture.
On the edge with Chick-fil-A
Brian Chambers began his Chick-fil-A journey selling chicken sandwiches to hungry diners. Today he is the Chief Architect at Chick-fil-A Corporate, where his job is still getting chicken sandwiches into the hands of hungry diners via the Chik-fil-A mobile app. And what an app it is: the company’s state-of-the-art application architecture runs a fleet of 2,800 Edge Kubernetes clusters (one in each of the company’s restaurants) to enable highly available, business-critical workloads to run without internet dependency. On Big Ideas in App Architecture Episode 16, Brian describes how edge computing helps Chick-fil-A solve capacity challenges, collect telemetry data, and many other challenges to ensure high-quality experiences for both kinds of customers — the ones ordering lunch at Chick-fil-A restaurants, and the company’s internal engineering teams.
Listen now: Inside Chick-fil-A’s infrastructure recipe for a perfect customer experience
Nobody turns your knobs (but you)
Andy Pavlo, professor in the computer science department at Carnegie Mellon University and the co-founder of OtterTune, is also the originator of the term Databaseology: the study of how you design and optimize software systems to efficiently store, retain, and process queries for data. On Episode 8 of Big Ideas, Andy talked about applying machine learning techniques to optimize database systems:
“People are always trying to squeeze as much out of the database as they can, and a big part of that is knob configuration tuning. Every database system has runtime parameters that control the behavior of the system, and they’re a huge pain. They’re basically buffer pool sizes, caching policies, log file sizes, things you can tune as the end user for how the data system is going to be used by the application…OtterTune customers have told us that they thought Amazon had already tuned their database for them. Well, they haven’t. You are in charge of your own knobs!
Listen now: Database Benchmarking Efficiency with OtterTune’s Andy Pavlo
Believe in angels
Matt Ranney, Principal Engineer at DoorDash, joins Big Ideas in App Architecture for Episode 5. Matt dishes on the pros and cons of microsystems and K8s, and shares DoorDash’s playbook for failures or downtime in microservices. Matt, who holds a fierce “partial failure is the same as complete failure” philosophy, discusses how to prevent and mitigate these failures to greatly improve overall system reliability — and how to find “productivity angels” in smaller teams through more automation.
Listen now: Building reliable systems with DoorDash’s Matt Ranney
Back to the (database) future
In Episode 9 of Big Ideas, Cockroach Labs co-founder and CTO Peter Mattis looks back on building GIMP as a college student, his early days at Google, and lessons learned that went into designing the architecture of CockroachDB (including lessons learned around building Postgres compatibility). Then he turns the telescope forward to peer into the future of databases. Is serverless the inevitable next paradigm? What about AI? Will there ever be one database that can do everything?
Listen now: Early Days at Google & Building CockroachDB, with Peter Mattis
It’s a jungle out there
These days, the digital analytics space is a Darwinian struggle to out-evolve your competitors amid the intense competition to survive. For Big Ideas episode 28, Shadi Rostami, SVP of Engineering at Amplitude, walks listeners through how the organization’s strategically innovative engineering culture, including real-time customer feedback and cutting-edge North Star metrics, keeps them continually evolving their tech stack to stay ahead in a new and rapidly developing market.
Listen now: Enabling a data-driven and innovative engineering culture at Amplitude
Secrets of the data warehouse
For the very first episode of Big Ideas, we asked Ken Pickering, SVP of Engineering at Starburst Data, to unlock the secrets of the data warehousing industry and help listeners stay ahead of the data storage curve. Ken walks users through the data mesh and AI analytics before delving into the intricacies of separating compute from storage, transferring to cloud-based solutions, and keeping up with evolving data storage ecosystems…And the importance of preparing for the coming event horizon of multi-cloud based systems.
Listen now: Decoding Data Warehousing: Insights from Ken Pickering, SVP of Engineering at Starburst Data
Kill it with fire. Or, actually, don’t.
I don’t mind systems that are written in COBOL. I don’t give a damn. Your system is entirely in COBOL? Excellent, great, wonderful. Now let’s talk about your ability to maintain it. As a mainframe/legacy system rescue professional who has worked on some of the most complex, and also gargantuan, computer systems in the world, Marianne Bellotti has seen some $#1+. For Big Ideas episode 25, author of Kill it With Fire: Manage Aging Computer Systems (and Future-Proof Modern Ones) dishes dirt about how, when mainframe systems go down, it’s always due to Java and that what most people believe about “legacy” systems is all wrong. This episode is an hour-long master class in large system rescue, and a must-listen for any engineer, architect or exec contemplating — or currently in the miserable midst of — a complex modernization project.
Listen now: Engineering resilient systems: Rescuing old treasures and unleashing modern capabilities
Terraform: A (multi cloud) love story
You may not instantly recognize Erol Kavas by name, but chances are excellent you’ve heard of his book, Mastering Terraform on AWS. As a multi-cloud expert with 100+ certifications to his name and a director at the global professional services giant PriceWaterhouseCoopers, Erol Kavas has no shortage of insights to offer when it comes to the state of the cloud ecosystem. He joins Big Ideas in App Architecture for an exploration of when and why enterprises should consider adopting multi-cloud, and patterns for implementing cloud infrastructure.
There may also be a fond mention (or three) of Terraform, including reasons why everyone needs to implement every Terraform best practice (spoiler alert: because those best practices apply no matter what CSP(s) you run on. Listen now: Mastering Multi-Cloud with PwC’s Erol Kavas
Joy, play, and platform engineering
As executive director of platform engineering at Estée Lauder, Meg Adams oversees everything needed to support the application, from cloud infrastructure, all the way to global analytics, automated QA, and developer tooling. But her main passion these days is building strong teams through decisions rooted in the concept of joyful work. “We serve the developer as our customer, and we try to make their lives better and easier as much as we can. I deeply believe that work can be a joyful and contributing part of people’s lives… And so, I have always just been attuned to ways of creating joyful environments where we could do really hard things together,” she explains in episode 27 of Big Ideas. Listeners learn about the concepts of neuroleadership — how the human brain literally works better in environments that incorporate play, and how this can measurably boost a company’s bottom line when leaders and managers consciously build teams where people feel satisfied with the work that they’re doing.
Listen now: How Estée Lauder scales strong engineering culture
A Grand Unified Theory of Software Engineering
Joe Lynch likes to say, “Aside from math and physics, there are no laws in software engineering” — a belief developed during an engineering career spanning decades, and at companies like Google and Twilio. Episode 11 of Big Ideas in App Architecture is a rousing discussion of the difference between “laws” and tendencies in software engineering, and how understanding Conway’s law can lead to better organizational structure, better peer review practices, and better software design. Joe introduces his satirical Grand Unified Theory of software engineering, which addresses real-world human issues like the danger that relentless pressure on engineers to get things right can metastasize into endless perfection seeking. He stresses the importance of setting up a “problem space” to identify problems before you start “solutioning”.
Listen now: How to Understand Problems & Build Better Software with Technical Leader Joe Lynch
