Presentation: The Evolution of Testing Methodology at AWS: From Status Quo To Formal Methods With TLA+
Amazon Web Services runs some of the largest distributed systems in the world. AWS hosts data and applications for a myriad of organizations including large enterprise businesses, government agencies, researchers, and internet startups. The data gets bigger and bigger; the systems grow at astonishing rates, and the software supporting these systems gets increasingly complex.
We take the availability and durability of these data very seriously. It is of the utmost importance to us that data is never lost or corrupted, is highly secure, and is highly available. At the same time we must continue to innovate, offer greater performance and functionality, and scale these systems while maintaining a sustainable operations profile.
To do all of this well means that the software must be bullet-proof when we clear it for deployment to production systems. Algorithm correctness becomes ever more critical. This talk will take a look at how testing strategies have evolved, and continue to evolve within AWS in order to achieve greater reliability at ever more massive scale. We'll start with the conventional methods, and then roll forward through time covering how and why we incorporated more powerful testing methodologies, ultimately leading us to the use of formal methods where TLA+ has become a cornerstone to our overall strategy.
Tracks
Covering innovative topics
Monday, 3 November
-
Architectures You've Always Wondered about
The newest and biggest Internet architectures
-
Real World Functional
Putting functional programming concepts to work in the real world.
-
The Future of Mobile
The future of mobile and performance improvements
-
Continuous Delivery: From Heroics to Becoming Invisible
Continuous Delivery philosophies, cultures, hiccups, and best practices.
-
Unleashing the Power of Streaming Data
This track explores a variety of use-cases, platforms, and techniques for processing and analyzing stream data from the companies deploying them at scale!
-
Sponsored Solutions Track I
Tuesday, 4 November
-
Engineering for Product Success
Architectures that make products more successful
-
Reactive Service Architecture
Reactive, Responsive, Fault Tolerant and More.
-
Modern CS In the Real World
How modern CS tackles problems in the real world.
-
Applied Machine Learning and Data Science
Understand your big big data!
-
Deploying at Scale
Containerizing Applications, Discovering Services, and Deploying to the Grid.
-
Sponsored Solutions Track II
Wednesday, 5 November
-
Beyond Hadoop
Emerging Big Data Frameworks and Technology
-
Scalable Microservice Architectures
This track addresses the ways companies with hundreds of fine-grained web-services (e.g. Netflix, LinkedIn) manage complexity!
-
Java at the Cutting Edge
The latest and greatest in the Java ecosystem
-
Engineering culture
Successes and failures in creating an engineering culture.
-
Next gen HTML5 and JS
How Web Components, the Future of CSS, and more are changing the web.
-
Sponsored Solutions Track III