Skip to the content.

AWS re:Invent 2024 - Architecture

Back to all Re:Invent 2024

Table of Contents

Architecture

AWS re:Invent 2024 - Think big, build small: When to scale and when to simplify (ARC331)

The speaker discusses an iterative approach to architectural design, emphasizing the importance of defining success metrics upfront, measuring performance, and integrating cost considerations early in the process. They also highlight the significance of human factors, such as team productivity and incentives, in driving architectural decisions that balance complexity and simplicity.

AWS re:Invent 2024 - Right-sizing your application services: Rethinking microservices (ARC324)

The talk discusses the importance of considering context when decomposing applications into microservices, rather than simply adopting a 'one-size-fits-all' approach. The speaker emphasizes the need to balance factors such as operational complexity, business criticality, and team dynamics to determine the appropriate granularity of services, rather than solely focusing on the number of microservices.

AWS re:Invent 2024 - How not to sabotage your transformation (SEG104)

This presentation discusses common patterns that can sabotage organizational transformation, such as outdated principles, siloed structures, wrong resource allocation, and over-efficiency. The speaker emphasizes the importance of adopting modern organizational principles, fostering cross-functional collaboration, and balancing speed and efficiency to drive successful transformation.

AWS re:Invent 2024 - Learn to create a robust, easy-to-scale architecture with cells (ARC335)

This session explores how AWS customers can create a robust, easy-to-scale architecture using a cell-based approach. Presenters Mahmoud Ismail and Paul Rapa discuss the benefits of cell-based architecture, including reduced blast radius, improved observability, and more efficient deployments, drawing on Slack's experience in migrating their egress platform to this pattern.

AWS re:Invent 2024 - Building resilient applications on AWS with Capital One (ARC334)

The video discusses how AWS and Capital One build resilient applications on AWS, focusing on principles of fault isolation, observability, and recovery. It covers topics like cell-based architectures, deep health checks, and automated recovery mechanisms to ensure mission-critical applications can withstand failures.

AWS re:Invent 2024 - SaaS meets cell-based architecture: A natural multi-tenant fit (SAS315)

The talk explores the intersection between multi-tenancy principles and cell-based architecture, highlighting how cell-based architecture can address challenges in scaling, resilience, and deployment flexibility in SaaS environments. The speaker emphasizes the need for architects to carefully consider the trade-offs and balance of complexity and value when evaluating cell-based architecture as a potential solution for their SaaS deployments.

AWS re:Invent 2024 - Mastering resilience at every layer of the cake (ARC327)

This session discusses advanced resilience characteristics, features, and patterns for building resilient applications on AWS. The presenters share insights and customer case studies on topics like chaos testing, disaster recovery, and building a resilience culture within organizations.

AWS re:Invent 2024 - Architectural best practices throughout the lifecycle (ARC206)

This session discusses AWS architectural best practices throughout the cloud optimization lifecycle. The speakers cover the journey of cloud optimization, including exploration, foundational, intermediate, advanced, and autonomous stages, as well as the use of tools like the Well-Architected Tool, Trusted Advisor, and AWS Health to support continuous improvement.

AWS re:Invent 2024 - Best practices for creating multi-Region architectures on AWS (ARC323)

This session discusses best practices for creating multi-Region architectures on AWS, focusing on failover strategies and decision-making processes. The speakers cover the AWS multi-Region fundamentals, organizational failover strategies, and a case study on Samsung Account's journey to a robust global disaster recovery architecture.

AWS re:Invent 2024 - How Amazon.com uses AWS Regions to improve customer experience (ARC328)

The talk discusses how Amazon.com uses AWS Regions to improve customer experience, highlighting the evolution of their thinking around regions, the challenges they faced, and the strategies and tools they used to navigate the transition to multi-region architecture. The presenters share their learnings and recommendations for other organizations looking to leverage multiple AWS Regions to enhance availability, reduce latency, and meet data localization requirements.

Developer Experience

AWS re:Invent 2024 - Building core capabilities: The secret to innovation at scale (SEG206)

This talk discusses strategies for driving innovation at scale, including the importance of shared customer obsession, insisting on the highest standards, fostering small autonomous teams, and implementing mechanisms to promote ideas from anywhere. The speaker provides practical examples and recommendations for building the necessary platform capabilities to enable this type of innovative culture.

AWS re:Invent 2024 - Supporting federated identity in multi-tenant environments (SAS307)

The video discusses how AWS Cognito can be used to support federated identity in multi-tenant SaaS environments. It covers the challenges and implementation details of integrating corporate identity providers with a SaaS application, and how Cognito can act as a federation hub to simplify the process for SaaS providers.

AWS re:Invent 2024 - Accelerating multi-tenant development with the SaaS Builder Toolkit (SAS406)

The SaaS Builder Toolkit (SBT) is a CDK-based tool that provides a composable and extensible architecture for building multi-tenant SaaS applications on AWS. It offers a control plane, app plane, and libraries to accelerate the development of SaaS solutions, allowing developers to pick and choose the components that best fit their application needs.

AWS re:Invent 2024 - The ideal micro-frontends platform (ARC325)

The presentation discussed the evolution of distributed systems, focusing on the adoption of micro-frontends as a solution to scale frontend applications. The speakers provided a detailed case study on how Formula 1 implemented micro-frontends, resulting in significant improvements in performance, deployment, and developer experience.

AWS re:Invent 2024 - Navigating the journey to becoming a cloud solutions architect (ARC211)

The session explores the role of a solutions architect, highlighting the key skills and personas required to thrive in this position. The speakers provide a comprehensive overview of the solutions architect's responsibilities, the importance of their role in driving business growth, and the tools and frameworks available to help aspiring and experienced solutions architects succeed.

AWS re:Invent 2024 - Fast evolving GraphQL schema with federation (BWP304)

The presentation covers how Amazon's Buy with Prime team manages the evolution of their GraphQL schema using versioning and federation. It also discusses leveraging generative AI to improve the test coverage and maintainability of their API testing.

AWS re:Invent 2024 - Centric Brands: A journey using AWS analytics and Amazon services (BWP303)

Centric Brands, a global lifestyle brand collective, leveraged AWS analytics and Amazon services, including Buy with Prime, to transform its e-commerce business and provide a seamless shopping experience for its customers. The presentation highlights Centric Brands' journey, the challenges they faced, and the solutions they implemented using AWS services to optimize their operations and drive business growth.

AWS re:Invent 2024 - Building and releasing robust generative AI solutions on AWS (BWP302)

This presentation discusses how Amazon is leveraging its fulfillment and supply chain capabilities to build robust generative AI solutions for merchants and shoppers. The speakers share their experience in developing and deploying these solutions, highlighting key requirements, architectural components, and best practices for maximizing the impact of generative AI in e-commerce.

AWS re:Invent 2024 - Leveraging human-AI collaboration to reinvent knowledge management (ARC205)

This session discusses how AWS and Commonwealth Bank of Australia (CBA) leveraged human-AI collaboration to reinvent their knowledge management practices. The speakers highlight the benefits of building expert technical communities, integrating Generative AI tools like AWS re:Post Private, and creating a knowledge flywheel to scale knowledge sharing and improve developer productivity across the organization.

Operations

AWS re:Invent 2024 - Anatomy of an AWS Region (ARC204)

This talk provides an in-depth look at how AWS builds new regions, including the physical and software infrastructure required, the challenges of managing dependencies, and the tools and processes used to coordinate the region build process. The speakers share key lessons and recommendations for designing resilient systems and effectively managing large-scale infrastructure projects.

AWS re:Invent 2024 - Chaos engineering: A proactive approach to system resilience (ARC326)

This talk explores the benefits of chaos engineering, a proactive approach to system resilience. The presenters share insights from their experiences at Amazon and BMW, highlighting the importance of verifying assumptions, practicing recovery skills, and fostering a culture of psychological safety to drive continuous improvement.

SaaS

AWS re:Invent 2024 - SaaS architecture pitfalls: Lessons from the field (SAS305)

The talk discusses the architectural pitfalls and best practices for building a successful SaaS application, including the importance of a robust control plane, identity management, observability, pricing strategies, and comprehensive testing. The speaker also introduces the SaaS Builder Toolkit, an open-source solution that provides a foundation for addressing these challenges.

Scaling

AWS re:Invent 2024 - Try again: The tools and techniques behind resilient systems (ARC403)

The talk discusses techniques and tools for building resilient distributed systems, including adaptive retries, circuit breakers, erasure coding, and simulation. The speaker emphasizes the importance of understanding the behavior of systems under load and the risks of common practices like retries that can lead to metastable failure modes.

AWS re:Invent 2024 - The incident is over: Now what? (ARC207)

This talk explores AWS's approach to incident handling and post-incident analysis, emphasizing the importance of rapid communication, empathy in the review process, and scaling learnings across the organization to prevent recurring issues. The speakers provide insights into AWS's incident response practices, including their focus on mitigation over root cause analysis, and the use of tools and processes to ensure consistent and effective incident management.

AWS re:Invent 2024 - Failing without flailing: Lessons we learned at AWS the hard way (ARC333)

This talk discusses lessons learned about resilience and scaling at AWS, including the importance of embracing failures, managing persistent connections, and measuring bottlenecks. The speakers provide practical advice and techniques for building resilient and scalable systems, such as regularly exercising recovery workflows, optimizing host startup times, and using metrics to proactively identify and mitigate performance issues.

AWS re:Invent 2024 - Scaling Prime Video for peak NFL streaming on AWS (ARC311)

This talk discusses how Prime Video, the video streaming service of Amazon, scaled its infrastructure and operations to handle the peak demand for streaming NFL Thursday Night Football games on AWS. The key strategies involved were multi-region architecture, elastic scaling, and the application of the AWS Well-Architected Framework to ensure reliability and cost optimization.

AWS re:Invent 2024 - Scaling on AWS for the first 10 million users (ARC201)

The presentation covers the architectural patterns and guidance for scaling an application on AWS from the first 10 million users. It discusses the use of managed services like Amplify Hosting, Fargate, and Aurora Serverless to handle the scaling challenges, and the transition to microservices and event-driven architectures as the application grows beyond 10 million users.

AWS re:Invent 2024 - SaaS storage strategies: Scaling, securing & tuning multi-tenant data (SAS306)

This session provides a comprehensive overview of strategies for scaling, securing, and tuning multi-tenant data in a SaaS environment. It covers key considerations such as security, scalability, and efficiency, as well as practical implementation details on access patterns, scaling techniques, and data modeling approaches.

AWS re:Invent 2024 - Generative AI meets multi-tenancy: Inside a working solution (SAS407)

This talk provides a comprehensive overview of how to build robust generative AI architectures that can scale to serve multiple tenants in a SaaS environment. The speakers dive deep into the technical details of implementing tenant isolation, cost monitoring, and throttling to ensure a seamless and secure multi-tenant experience.