Architecting for High Availability in the Cloud with Cellular Architecture

With the explosion in the usage of cloud computing and microservices, blast radius underpins highly availability systems. At AWS, all new services are embracing a cellular architecture - which gets them to reduce the blast radius, scale gracefully, and deliver the highest availability to its customers. We dive into the merits of cellular architecture, design options with cellularization, and how to effectively isolate at the level of the AWS account. We also cover Infrastructure-As-Code patterns to quickly bring up cells, how to securely connect each cell with foundational services like metrics across VPCs, and how to manage the multitude of AWS accounts you end up with. 

What's the focus of your work these days?

I’ve recently transitioned from working on the backend, architecture, and infrastructure components of Momento’s serverless cache to lead our developer ecosystem. That means that my current day job is providing clean, simple libraries that allow Momento users to start caching with as little friction as humanly possible, in whatever environment they are already building things in. Nowadays, instead of building out our infrastructure, I’m focused on helping users remove things from theirs :)

What's the motivation for your talk at QCon San Francisco 2023?

Having witnessed several different strategies for managing and scaling server-side infrastructure at my previous jobs, I have seen first-hand what a huge difference it makes for a software company (and more importantly, for their customers) to invest a little upfront to build a solid foundation for a cellular architecture. Over time, the relatively small upfront investment pays off in so many ways: protecting customers from outages, allowing the business to react more quickly to changing demands, dramatically reducing operational overhead and improving developer productivity; the list goes on. I’m excited to share some of these learnings in hopes that the audience can adapt them to their own infrastructure.

How would you describe your main persona and target audience for this session?

This talk is intended for engineering folks who are building and operating multi-regional cloud applications and services. We’ll primarily focus on AWS solutions to cellular architecture problems, so some familiarity with AWS account and permissions management will be useful, but the principles can be applied to any major cloud environment.

Is there anything specific that you'd like people to walk away with after watching your session?

Attendees will hear about some of the very tangible benefits that can be achieved by building automation for deploying and managing cellular infrastructure for your service or application. We’ll also give some concrete examples of specific technologies such as AWS SSO, AWS Organizations, and AWS CodePipelines that can be used to realize a scalable cellular solution. I hope to convince you that a small engineering investment upfront will pay for itself many, many times over by improving your customer experience and increasing your engineering velocity.


Chris Price

Software Engineer @Momento, Previously Launched New Video Streaming Services @AWS, 10+ Years in Engineering Leadership at Software Startups

Chris is a Software Engineer at Momento, where he helped design and build Momento’s cellular architecture and the corresponding CI/CD automation.

His interest in infrastructure automation began at Puppet Labs, during the genesis of the DevOps movement. After 5 years in a tech lead role at Puppet, Chris moved on to AWS, where he helped build the foundation for the brand new AWS MediaTailor service, a core component of Amazon’s NFL Thursday Night Football broadcasts and one of the first services inside of AWS to launch with a cellular architecture on Day 1.

Now at Momento, Chris is passionately working to improve the productivity and happiness of all engineers who have ever experienced the “joys” of building and operating a cache cluster, by contributing to Momento’s serverless caching products. When he escapes from the keyboard, you will most likely find him in the garden or on a bike ride in his beautiful home town of Portland, OR.

Read more
Find Chris Price at:


Wednesday Oct 4 / 03:55PM PDT ( 50 minutes )


Ballroom BC


Cellular Architecture Cloud Infrastructure Architecture Scalability


From the same track

Session Architecture

Understanding Architectures for Multi-Region Data Residency

Wednesday Oct 4 / 02:45PM PDT

Making your app multi-cell/multi-region can have a number of benefits in compliance, scaling and even marketability. There are a number of challenges, from cross-region access and trust to defining sources of truth across the globe. We’ll focus on:

Speaker image - Alex Strachan
Alex Strachan

Staff Software Engineer, Performance & Architecture @Rippling

Session Serverless

Lessons from Leading the Serverless First Journey at CapitalOne

Wednesday Oct 4 / 10:35AM PDT

Going serverless? Worried about challenges with performance, cost, and scale? At CapitalOne, we embraced a serverless first mentality across thousands of applications live today. This talk covers our journey into serverless, the best practices we picked up, and the lessons learned along the way.

Speaker image - George Mao
George Mao

Senior Distinguished Engineer @Capital One Leading All Things Serverless, Ex-AWS WW Serverless Tech Lead

Session Cloud

Million Dollar Lines of Code: An Engineering Perspective on Cloud Cost Optimization

Wednesday Oct 4 / 11:45AM PDT

A single line of code can shape an organization's financial future. Drawing inspiration from five real examples of million-dollar lines of code, we will challenge conventional views on engineering's pivotal role in cloud cost optimization.

Speaker image - Erik Peterson
Erik Peterson

CTO & Founder @CloudZero, Engineering-Led Cloud Cost Optimization Pioneer & 23-Year Software Startup Veteran

Session Architecture

Durable Execution for Control Planes: Building Temporal Cloud on Temporal

Wednesday Oct 4 / 01:35PM PDT

A great SAAS service requires a robust Control Plane as its brain. The Control Plane automates infrastructure provisioning, deployments, configuration, monitoring, fleet management, capacity allocation to customers,  and so forth.

Speaker image - Sergey Bykov
Sergey Bykov

SDE @Temporal Technologies