Jeremy C.

🎯 Running inference workloads on Kubernetes? I just wrote a deep dive on Gateway API Inference Extension and how it works. The Gateway API Inference Extension introduces two new Custom Resource Definitions (CRDs): InferenceModel and InferencePool. By using these two new resources, along with the endpoint selector, your L7 routing infrastructure becomes an “Inference Gateway” enabling you to self-host GenAI/LLMs with a “model-as-a-service” mindset. https://lnkd.in/g_eFqwUd #ai #llm #genai #inference

57

2 Comments

People often ask me about the differences in architecture between Amazon Dynamo (the 2007 SOSP paper), DynamoDB (the AWS serverless NoSQL database), and Aurora DSQL (the AWS serverless SQL databases). I captured the key points on durability, high availability, programming model, and trade-offs in my blog: https://lnkd.in/gkMG_tVD TL;DR: Despite the shared name, Dynamo and DynamoDB have little in common architecturally, and make very different trade-off decisions.

632

8 Comments

Amazon MSK now supports Apache Kafka version 3.9, bringing new flexibility for managing tiered storage and enabling consumer applications to continue reading historical data even after tiered storage is disabled at the topic level. This release also includes a range of bug fixes and improvements, making it easier than ever to run and scale real-time streaming workloads on a fully managed service. With compatibility across all AWS regions where MSK is available, teams can confidently migrate or upgrade their Kafka clusters and focus on building innovative streaming applications, while AWS handles the operational heavy lifting. #ApacheKafka #AmazonMSK https://lnkd.in/gSMJFsRh

21

AWS CodeCommit is back and open for fresh sign-ups now. We're bringing the band back together and will be adding new features and expanding to even more #AWS Regions in 2026. Read all about it at https://lnkd.in/guauiRBB

1,182

75 Comments

Automating Multi-Environment ECS Infrastructure with Terraform Modules One of the biggest challenges organizations face when scaling AWS ECS workloads isn’t just running containers—it’s managing multiple environments reliably. Dev, staging, production, and even region-specific deployments often drift apart when they’re built manually. Over time, this leads to fragile systems, unpredictable performance, and security gaps. As Solution Architects, our job is to prevent this drift and design automation patterns that balance flexibility with governance. Terraform modules are the foundation of this strategy. 🔍 The Multi-Environment Challenge Without automation, here’s what usually happens: Each environment is set up manually → differences creep in Developers copy-paste Terraform code between environments → duplication grows Fixes applied in one environment don’t always make it to others Scaling across regions or business units becomes a nightmare The result: wasted time, inconsistent infrastructure, and high risk when promoting workloads to production. 🚀 The Terraform Module Approach Terraform modules solve this by creating reusable building blocks that encapsulate best practices while exposing configurable variables. Here’s how I design ECS infrastructure modules: 1️⃣ Networking Module VPC, subnets, routing, NAT gateways Parameterized for CIDR ranges and AZ counts 2️⃣ ECS Cluster Module Cluster definition (Fargate or EC2 launch type) IAM roles for execution and tasks Logging and monitoring integration 3️⃣ Service Module Task definitions with container images, CPU/memory limits Load balancer integration via target groups Autoscaling policies tied to CloudWatch metrics 4️⃣ Security Module Security groups and IAM roles with least privilege Secrets Manager or SSM Parameter Store integration 📦 Environment-Specific Flexibility By combining these modules, you can spin up dev, staging, and prod clusters with minimal changes: Variables define region, scaling thresholds, or instance size The same codebase serves multiple environments, reducing drift Each environment benefits from identical architecture and security posture This design also makes it easy to replicate environments across new regions, business units, or customer accounts. 💡 Value Delivered ✔️ Consistency: Every environment is identical in structure ✔️ Velocity: New environments spun up in minutes, not weeks ✔️ Governance: Security and compliance controls embedded in modules ✔️ Cost Efficiency: Scaling policies tuned per environment without rewriting code ✔️ Auditability: Git history provides a single source of truth 🎯 Architect’s Leadership Role Terraform modules are not just code—they’re organizational standards. As architects, our role is to: Define the baseline modules that encode best practices Ensure teams don’t “reinvent” infrastructure patterns in silos Mentor engineers on module composition and variable design

5

From Kubernetes diffs and ECS weighted traffic shifting to Terraform auto-approvals and GitOps enhancements, we packed Q1 with powerful CD platform upgrades. Highlights include: ✅ Namespace enforcement for Kubernetes & Helm. ✅ Blue-Green weighted traffic for ECS. ✅ Git-triggered pipelines in GitLab. ✅ Environment-type-based freeze windows. ✅ API-driven observability. 👉 Dive into what’s new: https://lnkd.in/dKGG3h95

24

Still using static secrets to pull images? There’s a better way. 🚀 The JFrog Kubelet Credential Provider now supports all major cloud providers (EKS, GKE, AKS) and the latest KEP-4412 enhancements. Forget the headache of imagePullSecrets and move to automated, identity-based authentication for your #Kubernetes clusters: https://bit.ly/4rPkLjc

32

Two of AWS’ Sr. Principal Engineers, Nicholas Matsakis and Marc Bowes, take us inside the development of Aurora DSQL in a fascinating new blog. The post shows how the team scaled write operations without two-phase commit, overcame garbage collection hurdles, and ultimately embraced Rust for both data and control planes. What I find particularly interesting is how the team's journey with Rust evolved. They started cautiously, using it for a single component, and ended up rewriting almost everything in Rust. It's a reminder that sometimes the hard choices - like adopting a new language - can unlock tremendous value. The full post is well worth your time. It's not just about database internals, but also about team dynamics, a culture of learning (and questioning established approaches), and making difficult architectural decisions. Read it here: https://lnkd.in/eMhxBu9e Now, go build!

1,045

16 Comments