Mastering Observability

insightful as always from Dotan and this time he talks to Charity Majors of honeycomb.io , a great listen.

The Linux Foundation's Open Model Initiative (OMI) and Model Openness Framework (MOF) are bringing cloud-native principles to AI observability. They seem to be tackling proprietary model challenges by promoting transparency through initiatives like the Community Data License Agreement, making it easier for teams to share and monitor model performance. It's worth watching as it could simplify how we handle model observability at scale. #observability #OpenSource #AIops

Google's Code Assist has added third-party tool support, bringing real-time data integration from partners like Sentry.io, GitLab, and GitHub into their coding environment. While the private preview is limited to Google Cloud partners, it's interesting to see how they're tackling the context-switching problem. I'm sceptical about their claims of seamless integration, but the Sentry.io incident tracking could be a game-changer for observability workflows. Their on-premises support might give them an edge over GitHub Copilot Enterprise. #observability #devops #AIcoding

How can AI-powered observability transform DevSecOps while maintaining robust security controls? Kindo's launch of their channel partner program showcases how AI is revolutionizing observability in DevSecOps. Their WhiteRabbitNeo model enables automated security monitoring across infrastructure-as-code, vulnerability assessment, and IAM management. Im interested in Kindo's ability to automatically execute internal runbooks within platforms while maintaining security observability. For me, this stands out as an cool approach to scaling DevSecOps. #observability #DevSecOps #AISecurity

Observability Digest 39: AI Innovations & Cloud Advancements is out. This week Allan M. explores AI-driven observability breakthroughs, alongside insights from Kai Waehner and Priyanka Sharma. From Dash0's 98% AI log analysis accuracy to EDB's hybrid cloud solutions, we're seeing transformative impacts on monitoring and analysis capabilities. Main points, 🔄 Dash0 's AI log analysis hits 98% accuracy 🎯 EDB enhances hybrid cloud observability 🤖 Kai Waehner predicts data streaming evolution 📊 AWS improves Network Firewall monitoring 🔐 US quantum leadership gap highlighted 💡 Datadog 's Arm migration journey Don't miss our CodeCraft section exploring Python-based EC2 monitoring and open-source LLMOps tools. Join the conversation and share your thoughts on AI's role in shaping the future of observability. https://lnkd.in/eFsJvk_n #Observability #AIInnovation #CloudTech #DevOps #DataAnalytics

Do you need a refresher on the OpenTelemetry Collector, if so why not give Allan's quick guide a read, we're loving his simple config examples, pretty clean (way better than juggling multiple agents!). Been there with the PII filtering headaches - good to see it's baked into the processors. #observability #OpenTelemetry #cloudnative

ClickHouse 's new AWS collaboration is live and thier five-year strategic partnership will be focusing on real-time analytics and GenAI integration and with AWS services like Firehose and Bedrock. The numbers are pretty solid - they're processing 2 billion daily queries across 50 petabytes of data. Curious about their ambitious industry solutions roadmap, their 150% AWS Marketplace growth in 2024 speaks volumes. #observability #cloudanalytics #datascience

How do you successfully migrate thousands of containerized services to Arm while maintaining performance and reducing costs? Datadog's massive migration of their Kubernetes fleet is an interesting read. By implementing multi-architecture builds, they moved 70% of AWS workloads to Arm-based instances, achieving a 10% reduction in AWS costs while improving system resilience. What I took away from their article: - Use container registry manifests and performance benchmarking to identify quick-win migration candidates - Leverage BuildKit for efficient multi-architecture builds vs slower emulation - Track migration progress using bespoke metrics - Implement canary deployments with fallback options for production rollouts We like their data-driven approach using Service Catalog and Cloud Cost Management to create some cool tracing views showing detailed visibility into migration status, priorities and savings. #observability #kubernetes #cloudmigration

Some late night readin, give Dmitry Romanoff's piece on EC2 instance info retrieval with Python/Boto3 a go. Having worked extensively with AWS automation, love how clean this approach is. The error handling is particularly solid (this would of saved me from many late-night pages). One thing I might add from experience - using resource tags for better filtering, especially in multi-account setups. Been there when trying to track down ghost instances! #python #aws #observability

Allan latest deep dive into hybrid cloud operations reveals some eye-opening stats - 40% of enterprises blow past their budgets in year one! He's laid out a solid framework for managing these complex environments, though I'm a bit skeptical about those projected cost savings from IaC. His section on unified observability really hits home, especially for teams juggling legacy and cloud systems. What's particularly interesting is his take on zero trust architecture reducing security incidents by 60%. That's no small feat in today's landscape. The sustainability angle is fresh too, though I'd love to see more concrete metrics on energy efficiency gains. #observability #hybridcloud #cloudnative