AWS Bites

We take a deep dive into Amazon S3 Files, AWS's exciting new managed file system backed by S3!
We kick things off by exploring why S3 isn't a traditional file system, covering everything from the lack of true directories and atomic renames to immutable objects and POSIX access control differences. We then walk through the existing solutions people have used to bridge that gap, like S3FS FUSE, MountPoint for S3, FSx for Lustre, and Storage Gateway.
From there, we get into the heart of the episode: how S3 Files works, how to set it up, and how it uses EFS under the hood as a caching layer. We share our own real-world benchmarking results comparing S3 Files against various EFS configurations across Lambda and Fargate, and we discuss a real customer project where we put S3 Files to the test.
We also cover the important caveats like eventual consistency, the 60-second write-back delay, the lack of cross-account bucket support, and the cost model so you can make an informed decision.
Resources mentioned
Episode 124: S3 Performance
Episode 95: Mounting S3 as a Filesystem
Amazon S3 FAQs: S3 Files
fourTheorem S3 Files demo code on GitHub
Amazon documentation: Understanding how synchronization works
Sponsor
Thanks to fourTheorem for powering AWS Bites. We help teams build cloud systems that are simple, scalable, and cost effective. Visit fourtheorem.com.
Chapters
00:00 Introduction: Why S3 is amazing but not a file system, and what S3 Files promises to solve
01:47 Why S3 is not a file system: no true directories, immutable objects, no atomic renames, expensive listings, and POSIX differences
05:23 Existing solutions for mounting S3 as a file system: S3FS FUSE, Python fsspec, Hadoop S3A, MountPoint, FSx for Lustre, File Cache, and Storage Gateway
07:16 How S3 Files works: NFS-based access, EFS caching layer, streaming from S3, and supported compute services like EC2, ECS, EKS, and Lambda
09:49 Setting up S3 Files: buckets, file system resources, import and expiration rules, mount targets, access points, VPC requirements, and NFS port configuration
13:42 S3 Files performance numbers from AWS documentation: throughput, IOPS, latency figures, and why real-world benchmarking is recommended
15:39 Benchmarking S3 Files vs EFS configurations on Lambda and Fargate: small and large file reads and writes, memory/CPU impact, and key findings
19:48 Downsides and limitations: NFS only, no hard links, no atomic renames, eventual consistency, the 60-second write-back delay, and large-scale rename performance warnings
23:05 Real-world project experience: a SaaS multi-tenant architecture, cross-account bucket limitation discovered, and how the team worked around it
27:52 Cost breakdown: EFS-equivalent cache pricing, S3 storage costs, reads from cache vs. S3 directly, and how S3 access tiers still apply
29:50 Final recap and take: when S3 Files shines, when to be cautious, mixed access pattern warnings, and an invitation to share your own experiences
33:42 Closing
Send us your AWS questions
Do you have any AWS questions you would like us to address? Leave a comment here or connect with us on X/Twitter, Bluesky, or LinkedIn:
Eóin: Bluesky | LinkedIn
Luciano: X/Twitter | Bluesky | LinkedIn

If you’re curious about building with LLMs, but you want to skip the hype and learn what it takes to ship something reliable in production, this episode is for you.We share our real-world experience building AI-powered apps and the gotchas you hit after the demo: tokens and cost, quotas and throttling, IAM and access friction, marketplace subscriptions, and structured outputs that do not break your JSON parser.We focus on Amazon Bedrock as AWS’s managed inference layer: how to get started with the current access model, how to choose models, how pricing works, and what to watch for in production.We also go deep on structured outputs: constrained decoding, schema design that improves output quality, and how to avoid “grammar compilation timed out”.

In this episode, we mentioned the following resources:
fourTheorem: Bedrock structured outputs guide https://fourtheorem.com/amazon-bedrock-structured-outputs/
Amazon Bedrock https://aws.amazon.com/bedrock/
Bedrock docs https://docs.aws.amazon.com/bedrock/latest/userguide/
Bedrock pricing https://aws.amazon.com/bedrock/pricing/
Structured outputs https://docs.aws.amazon.com/bedrock/latest/userguide/structured-outputs.html
Cross-region inference https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference.html
Quotas https://docs.aws.amazon.com/bedrock/latest/userguide/quotas.html
Throttling help https://repost.aws/knowledge-center/bedrock-throttling-error
Prompt caching https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html
Troubleshooting error codes https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html

Do you have any AWS questions you would like us to address?
Leave a comment here or connect with us on X/Twitter, BlueSky or LinkedIn:

- ⁠https://twitter.com/eoins⁠ | ⁠https://bsky.app/profile/eoin.sh⁠ | ⁠https://www.linkedin.com/in/eoins/⁠
- ⁠https://twitter.com/loige⁠ | ⁠https://bsky.app/profile/loige.co⁠ | ⁠https://www.linkedin.com/in/lucianomammino/

AWS Lambda is fantastic for small, stateless code on demand. But when your “function” starts looking like a workflow (retries, backoff, long waits, human approvals, callbacks), classic Lambda patterns can feel like a fight: 15-minute max runtime, no built-in state, and orchestration glue everywhere (Step Functions, queues, schedules, and state you did not want to own). In this episode of AWS Bites, Eoin and Luciano explore AWS Lambda Durable Functions, announced at re:Invent 2025. It’s still Lambda (same runtimes and scaling), but with durable execution superpowers: named steps, automatic checkpointing, and the ability to suspend and resume from a safe point without redoing completed work. We unpack the replay/resume model under the hood, when this approach shines, and the gotchas (determinism, idempotency, replay-aware logging, debugging resumed runs). To make it real, we share how we rebuilt PodWhisperer v2 using Durable Functions to orchestrate a GPU-powered WhisperX pipeline, LLM refinement, speaker naming, and caption generation.

In this episode, we mentioned the following resources:
AWS announcement blog post: https://aws.amazon.com/blogs/aws/build-multi-step-applications-and-ai-workflows-with-aws-lambda-durable-functions/
Durable Functions best practices: https://docs.aws.amazon.com/lambda/latest/dg/durable-best-practices.html
The replay model deep dive (Dev.to): https://dev.to/aws/the-replay-model-how-aws-lambda-durable-functions-actually-work-2a79
Build workflows that last (Dev.to): https://dev.to/aws/aws-lambda-durable-functions-build-workflows-that-last-3ac7
Testing Durable Functions in TypeScript (Dev.to): https://dev.to/aws/testing-aws-lambda-durable-functions-in-typescript-5bj2
Developing Durable Functions with AWS SAM (Dev.to): https://dev.to/aws/developing-aws-lambda-durable-functions-with-aws-sam-ga9
Hands-on notes: https://www.andmore.dev/blog/lambda_durable_functions/
PodWhisperer (open source): https://github.com/fourTheorem/podwhisperer/
WhisperX: https://github.com/m-bain/whisperX


Do you have any AWS questions you would like us to address?
Leave a comment here or connect with us on X/Twitter, BlueSky or LinkedIn:

- https://twitter.com/eoins | https://bsky.app/profile/eoin.sh | https://www.linkedin.com/in/eoins/
- https://twitter.com/loige | https://bsky.app/profile/loige.co | https://www.linkedin.com/in/lucianomammino/

AWS just made Lambda… less serverless. Lambda Managed Instances (Lambda MI) brings managed EC2 capacity into Lambda, and it changes the rules: environments stay warm, a single environment can handle multiple concurrent invocations, and scaling becomes proactive and asynchronous instead of “spin up on demand when traffic hits.”In this episode of AWS Bites, Eoin and Luciano break down what Lambda MI unlocks (and what it costs): fewer traditional cold starts, but a new world of capacity planning, headroom, and potential throttling during fast spikes. We compare it to Default Lambda, explain how the new scaling signals work, and what “ACTIVE” really means when publishing can take minutes on a new capacity provider.To make it real, we built a video-processing playground: an API, a CPU-heavy processor, and a Step Functions workflow that scales up before work and back down after. We share the practical lessons, the rough edges (regions, runtimes, mandatory VPC, minimum 2 GB + 1 vCPU, concurrency pitfalls), and the pricing reality: requests + EC2 cost + a 15% management fee.

In this episode, we mentioned the following resources:
Lambda Managed Instances official docs: https://docs.aws.amazon.com/lambda/latest/dg/lambda-managed-instances.html
Our example repo (video processing playground): https://github.com/fourTheorem/lambda-mi
Concurrency mental model reference (Vercel Fluid Compute): https://vercel.com/fluid
Lambda MI Node.js runtime best practices (concurrency considerations): https://docs.aws.amazon.com/lambda/latest/dg/lambda-managed-instances-nodejs-runtime.html


Do you have any AWS questions you would like us to address?
Leave a comment here or connect with us on X/Twitter, BlueSky or LinkedIn:

- ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/eoins⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://bsky.app/profile/eoin.sh⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/eoins/⁠⁠⁠⁠
⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠- ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/loige⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://bsky.app/profile/loige.co⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/lucianomammino/

Love AWS Fargate, but occasionally hit the “I need more control” wall (GPUs, storage, network bandwidth, instance sizing)? In this episode of AWS Bites, Eoin and Luciano put the brand-new Amazon ECS Managed Instances (ECS MI) under the microscope as the “middle path” between Fargate simplicity and ECS on EC2 flexibility. We unpack what ECS MI actually is and where it fits in the ECS spectrum, especially how it changes the way you think about clusters and capacity providers. From there we get practical: we talk through the pricing model (EC2 pricing with an additional ECS MI fee that can be a bit counterintuitive if you rely heavily on Reserved Instances or Savings Plans), and we share what it feels like to finally get GPU support in an experience that’s much closer to Fargate than to “full EC2 fleet management”. To make it real, we walk through what we built: a GPU-enabled worker that transcribes podcast audio using OpenAI Whisper, including the end-to-end setup in CDK (roles, capacity provider wiring, task definitions, and service configuration). Along the way we call out the rough edges we ran into, like configuration options that look like they might enable Spot-style behavior, and the operational realities you should expect, such as tasks taking roughly 3–4 minutes to start when ECS needs to provision fresh capacity. We close by mapping out the workloads where ECS MI shines (queue-driven GPU jobs, HPC-ish compute, tighter storage/network control) and the scenarios where it’s probably the wrong choice, like when you need custom AMIs, SSH access, or stricter isolation guarantees.

In this episode, we mentioned the following resources:
Amazon ECS Managed Instances: ⁠https://aws.amazon.com/ecs/managed-instances/⁠
ECS Managed Instances documentation: ⁠https://docs.aws.amazon.com/AmazonECS/latest/developerguide/ManagedInstances.html⁠
Amazon Bottlerocket (what it is): ⁠https://aws.amazon.com/bottlerocket/⁠
Our CDK ECS MI template: ⁠https://github.com/fourTheorem/cdk-ecs-mi-template⁠
Ep 42. How do you containerise and run your API with Fargate?: ⁠https://awsbites.com/42-how-do-you-containerise-and-run-your-api-with-fargate/⁠
Ep 72. How do you save cost with ECS?: ⁠https://awsbites.com/72-how-do-you-save-cost-with-ecs/⁠
Ep 10. Lambda or Fargate for containers?: ⁠https://awsbites.com/10-lambda-or-fargate-for-containers/⁠
Ep 38. How do you choose the right compute service on AWS?: ⁠https://awsbites.com/38-how-do-you-choose-the-right-compute-service-on-aws/⁠
Ep 143. Is App Runner better than Fargate?: ⁠https://awsbites.com/143-is-app-runner-better-than-fargate/⁠


Do you have any AWS questions you would like us to address?
Leave a comment here or connect with us on X/Twitter, BlueSky or LinkedIn:

- ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/eoins⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://bsky.app/profile/eoin.sh⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/eoins/⁠⁠⁠
⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠- ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/loige⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://bsky.app/profile/loige.co⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/lucianomammino/