David G.

165,000 resources across 625 workspaces and 38 AWS accounts, supporting 170 engineers. That's automation with HCP Terraform. 🪄 Watch to learn more about Benchling's journey to a unified infrastructure as code (IaC) approach with HashiCorp: https://lnkd.in/gegr69x5

49

When Steve Huynh was a principal engineer at Amazon, meetings began with a "study hall." Amazon had a "reading culture" even among engineers, Huynh recently told the Pragmatic Engineer podcast, speaking of his time at the tech giant. Employees frequently drafted six-page memos, which they shared with the company to update progress and demonstrate new projects. Huynh, who said the company's embrace of writing and reading the memos was part of its "secret sauce," said employees' writing was often constrained to the format, whether it was a business strategy or press release. Huynh started at Amazon in 2006, only a few years after the company turned its first profit and while Jeff Bezos was at the helm. Read more about how Bezos instilled this culture of memo-writing from the top down on Business Insider: https://lnkd.in/gBqbEf_j (Credit: Daniel Berman) #amazon #reading #jeffbezos

216

2 Comments

In-memory databases are the architectural backbone of real-time AI. They’re a necessary advantage for memory-intensive workloads that rely on instant access to context. Modern systems like Redis blend durability, multi-model flexibility (JSON, vectors, time-series), and tiered storage to deliver the speed of memory with the reliability of disk. The result? Real-time analytics, instant personalization, and AI systems that actually feel intelligent. Here’s how we deliver the speed of memory with the reliability of disk: https://lnkd.in/gSm65kdB

49

Over the past eight years I’ve worked across the open-source networking world—from building residential routers at Comcast, to Amazon Walkout Technology team, and most recently the Telecom Infra Project OpenWiFi initiative. One lesson has paid dividends in every role: really understanding how a packet traverses the Linux networking stack. If you want to build the same picture in your head, start with these two gold-standard resources: 1️⃣ “How packets travel through Linux” – a one-hour deep dive by Jiri Benc 🎥 https://lnkd.in/e4fzxiPb 2️⃣ linux-network-performance-parameters – Leandro Moreira GitHub guide that links every critical /proc, sysctl, and tc knob to the problems it solves 📚 https://lnkd.in/epKnTgCh Bookmark them—you’ll never look at packet flow the same way again.

109

7 Comments

Building AI infrastructure on-prem? Here are foundational principles to keep in mind: Modular architecture: Decouple compute from storage so each can scale independently. This makes it easier to adapt to growing or shifting workloads over time. Future-ready hardware: Select GPUs and CPUs not just for current workloads but with an eye on scalability, including support for newer accelerator types. Scalable design: Ensure the system allows seamless addition of compute nodes or storage without a full redesign. Open ecosystems: Rely on open source tools and open standards to stay agile, reduce long-term costs, and avoid vendor lock-in. This helps you tailor the stack as your needs evolve. With the right foundation, on-prem setups offer deep control, performance tuning, and long-term flexibility. More tips? Full video on YouTube: https://lnkd.in/eXV6_A6r #Canonical #Ubuntu #OpenSource #Cloud #AI

34

2 Comments

AWS SAP Studies Continued: Amazon RDS I’m in the databases section of my AWS Solutions Architect Professional prep. What RDS is: A managed database that lives inside your VPC. Put it in private subnets and connect from your app tier or from your private network (VPN/Direct Connect). You can make it public, but that’s rarely smart for production. Architecture essentials --- Many engines (MySQL/Postgres) can host multiple logical databases on one RDS instance. --- Storage: each instance uses its own EBS. Multi-AZ Cluster style shares a synced cluster volume across AZs. --- Backups: stored in AWS-managed S3 (you won’t see the bucket). High availability (two flavors) --- Multi-AZ Instance: primary + hidden standby in another AZ. Standby isn’t readable; failover takes a short pause. --- Multi-AZ Cluster: one writer + two readable standbys. Faster failovers and built-in read scaling (use the reader endpoint). Backups & restores (your safety net) --- Automated backups (PITR): daily snapshot + transaction logs let you restore to an exact second within your retention (up to 35 days). --- Manual snapshots: you trigger them; first is full, later are incremental; they don’t expire and outlive the DB until you delete them. --- Performance impact: Instance mode usually backs up from the standby; Single-AZ/dev may see brief I/O pauses; Cluster minimizes impact. --- Restores: always create a new DB with a new endpoint—plan the cutover and practice it. Read replicas (scale + DR help) --- Asynchronous copies to offload reads and help cross-Region scenarios. Great for reports/dashboards. Expect lag; don’t use them for strict read-after-write paths. DOCS: https://lnkd.in/eyAR8iVv https://lnkd.in/ekwxi5G7 #AWS #Solutions #Architect #Professional #RDS #Databases #Scaling #Performance #DisasterRecovery #RPO #RTO #Backups #Snapshots

13

Anyscale 🤝GKE Through our continued collaboration with the Google Kubernetes Engine (GKE) team, we’re advancing the experience to scale Ray and Kubernetes in the cloud. This work includes: Label Selectors for improved scheduling: https://lnkd.in/gUZ8iVUW New Features for AI Scaling: https://lnkd.in/gAGSMiir Enhanced TPU Support: https://lnkd.in/gsKDvwpv

65

3 Comments

🚀 Zero-downtime deployments just got simpler with native support for blue/green deployments on Amazon ECS! 👉 https://go.aws/4m3bEXP Gone are the days of manual environment management and complex traffic routing. #AmazonECS now handles blue/green deployments natively, letting developers focus on building great applications. Here's how it transforms your deployment process: • Set up two target groups in your application load balancer • Configure ECS service with blue/green deployment strategy • Define bake time for side-by-side monitoring • Add Lambda lifecycle hooks for automated testing • Let ECS handle traffic shifting and rollback automatically #AWSLambda hooks can respond with SUCCEEDED, FAILED, or IN_PROGRESS - giving complete control over deployment validation. Multiple hooks can be added for different testing stages, from basic health checks to downstream dependency verification. This native integration eliminates the undifferentiated heavy lifting that previously made blue/green deployments complex to implement.

77

2 Comments

As your synthetic test collection grows, it becomes vulnerable to data sprawl and configuration draft. Learn how using Terraform with Datadog Synthetic Monitoring can help you automate test creation, enforce best practices, and simplify maintenance: https://lnkd.in/eGk-EG9k

76

Kubernetes's greatest strength is also its greatest weakness. Kubernetes clusters enable rapid scaling, resiliency, and automation across clouds, but they eat up IP addresses like a plague of locusts. Enterprises find that IP exhaustion and overlap cause significant configuration issues across a network. Aviatrix's Tim McConnaughy breaks down the major challenges of Kubernetes IP exhaustion and overlap, the common workarounds, and how Aviatrix's new Kubernetes Firewall provides a game-changing solution. 🔗 Read here: https://loom.ly/As3OIkk #IPExhaustion #IPOverlap

41

With out-of-the box support for Cribl Cloud as well as native support for platforms such as Amazon Security Lake, Amazon S3, Azure Blob, and Google Cloud Storage, Cribl seamlessly analyzes all data right at its source. Learn how: https://lnkd.in/dHr5AnRx

30

1 Comment

We ran a head-to-head benchmark of AWS Textract against nanonets-ocr-2, our VLM-based doc processing model. Textract breaks reading order on multi-column pages. It skips charts and maps entirely. It struggles with merged cells and nested tables. And it still outputs raw text blocks instead of structured markdown. If your RAG pipeline or agent is producing garbage answers, the problem probably isn't your LLM. It's your parser. Here's what we found across 7 dimensions: - Layout and reading order: Textract gives up on complex pages. We preserve every section in order. - Visual elements: Textract labels them "Figure 1" and moves on. We describe the content. - Complex tables: Merged cells, spanning rows, sub-headers. We handle them. Textract doesn't. - Multilingual: 200+ languages out of the box. No config. - Signatures and stamps: Detected, linked to signees, included in output. - Accuracy on hard docs: +15% better on Docext and OlmOCR-Bench. - LLM-ready output: Clean markdown with doc structure preserved, and confidence scores One model. Structured markdown out. Confidence scores included. Deploys on-prem in a single Docker container. Try it on the documents your current parser can't handle.

21

🤝 The observability world loves a good rivalry — but this isn’t one. Prometheus and OpenTelemetry are evolving together, not apart. Myrle Krantz, Director of Engineering, explains why. Learn more about the state of observability in our new report: https://lnkd.in/g2d4636F

102

3 Comments

From Terraform to Crossplane Kubernetes-native IaC is evolving—are you keeping up? Crossplane is redefining how we manage infrastructure: declarative, secure, and fully integrated with the K8s ecosystem. Think Terraform, but closer to your cloud-native apps. -YAML over DSL -Native RBAC & multi-tenancy -Seamless integration with Helm, Operators & more Explore how Crossplane offers the next-gen control plane—bridging DevOps and Platform Engineering like never before. #InfrastructureAsCode #Crossplane #Kubernetes #CloudNative #DevOps #PlatformEngineering #MSysTechByte #Infraunlocked #MSysTechnologies

13