Rodrigue Koffi

If you missed our last Twitch episode, be sure to check it out here: https://lnkd.in/e9s2HmZd ! We had a fascinating discussion on how to transform underused, unstructured data into valuable strategic assets using AWS. Our guest speaker Warre Dreesen from dataroots presented a solution using RAG systems to help journalists quickly find and summarize information about recent news topics. Learn how you can apply these techniques to get more value from your own data assets. 🚀 #aws #genAI #RAG

18

Deezer uses Event Loop Utilization as a custom metric for auto-scaling. Kubernetes' default auto-scaling based on CPU and memory often doesn't meet users' needs. And what Deezer is doing is quite common. 🤔💡 #Kubernetes #AutoScaling #CustomMetrics https://lnkd.in/gEM6BxbD

3

STOP THE CRAZY TRAIN BEFORE IT'S TOO LATE! Here's a nice blog post in "Mission Impossible" style, explaining how my French colleagues built a fun and inspiring AI demo with moving parts and cameras. There's also a link to a YouTube video showing how it all works in practice. Thanks to Nicolas Massé Adrien Legros Pauline T. Mourad Ouachani https://lnkd.in/dH72KMHJ #openshiftai #ai #mlops #predictive #redhat #fun #openshift #opensource

41

Thank you, Percona, for MongoDB Operator v1.8.0! 🎉 The new release helps reduce cross-regional network traffic and latency, making your applications faster and more efficient. Each MongoDB node now has region/zone tags. You can choose the preferred primary node using the CRD based on Kubernetes region/zone labels. This is great news for multi-cloud and hybrid-cloud environment use cases. primaryPreferTagSelector: region: us-west-1 zone: us-west-1a #OpenSource #DBA #Kubernetes #HybridCluster https://lnkd.in/dKvYuTGV

18

After his excellent presentation at the AWS Paris Summit, Guillaume Marchand has written a helpful blog post on the power of Continuous Configuration (on top of CI/CD) and feature flags. If you are looking to increase your team's velocity, and improve the resiliency of your software, read this blog post. #aws #awsappconfig #cicdcc https://lnkd.in/edPvu9pp

33

1 commentaire

I learned about microfrontends almost 7 years ago, thanks to Luca Mezzalira who relentlessly evangelized this frontend design approach. I invited him to talk about it in ReInvent 2018, after which we were lucky to see him joining AWS. Last year, Luca partnered with Matteo Figus Alexander Günsche Harun Hasdal to write a long format guidance around the topic from AWS perspective. I had the privilege to review this document before publishing, and I can only recommend any professional responsible of or contributing to a web application to read it. Even if you do not decide to implement a microfrontend architecture, you can still learn from this document: different techniques for rendering, domain driven design, api integration approaches, governance, etc.. That was my case ^^ Please share it in your network and let your friends also learn! https://lnkd.in/d23QgGQp #aws #frontend #mfe #microfrontend #javascript #react #vue

111

5 commentaires

🌟 Is your Kubernetes deployment as efficient as it could be? Discover how OVHcloud's Managed Rancher Service (MRS) is transforming the scalability and flexibility of container orchestration! 🔧 Managed Rancher Services streamline the deployment and management of Kubernetes clusters, making it easier than ever for software developers and DevOps professionals to enhance their workflows. With OVHcloud’s new Local Zones, say goodbye to latency issues and hello to improved service performance! 🖥️ In a recent article, a detailed demo walks you through creating a Kubernetes cluster step-by-step. From setting up Compute Instances in a Local Zone to configuring Rancher for deployment, this guide is a treasure for anyone looking to leverage the power of container orchestration. 🚀 Don't miss the chance to try MRS for free during its beta phase! Dive into the future of seamless application deployments. How do you approach Kubernetes management? Share your strategies and insights in the comments below! Stay Ahead in Tech! Connect with me for cutting-edge insights and knowledge sharing! Want to make your URL shorter and more trackable? Try linksgpt.com #BitIgniter #LinksGPT #Kubernetes #DevOps

2

Join Jamie Beck and Hicham Mourad in this webinar to learn how #RedHat Ansible Automation Platform simplifies managing your cloud infrastructure, streamlining and operationalizing activities across Google Cloud and cloud-native services.

2

Fantastic data points by Ajit Inamdar on Amazon ECS !! Using Amazon ECS from tiny workloads to large production grade deployments is a no-brainer, and you let AWS handle all the heavy lifting of container orchestration and yes, you can also leverage your existing AWS skills to get started. #aws #containers #amazonecs

17

2 commentaires

Managing legacy systems is hard... even for AWS "S3 is showing its age": very nice article by Chris Riccomini on #AWS #S3: https://lnkd.in/e8rPiMdi Object storage is often becoming the primary storage for #data systems such as #WarpStream. AWS S3 once defines the defacto standard API for all object storages and was in the past well in advance in terms of performance/price and features. This is now over and it may become a switching factor in the near future as explained in the article. In the past they managed to overcome the atrocious "4 characters random prefix" to bucketnames that was needed to avoid hotspot scenario and bad performance, but the road ahead is challenging and the competition is real. - No Compare-and-Swap (CAS) Operations: Unlike its competitors, S3 lacks preconditions, forcing developers to use workarounds like DynamoDB. - S3 Express One Zone (S3E1Z): While promising, S3E1Z misses many standard S3 features and comes with high storage costs. - Lack of Multi-Region Buckets: S3 doesn't support dual-region or multi-region buckets, limiting its availability options.

12

📣 Big news for developers! Cloud Run now supports GPUs! 📣 You can quickly host LLMs on serverless now, only on Google Cloud. ✅ Low Latency ✅ Cost-Efficient ✅ Easy to Use Learn more: 👉 https://lnkd.in/dYG6mbNg #CloudRun #Serverless #AI #GPU #GoogleCloud

1

"What would you like to do, dear customer?”. We asked Pierre Bremehr, Google Partner Manager, @Arvato Systems what excites him about our partnership and why our customers benefit from it. More in the video 👇 #GoogleCloud #CloudSummitDE #AI #Networking #DigitalTransformation

2

For what it's worth from my (formerly) insider perspective, these things take time because they are tougher than you think. Kubernetes has it "easy" because every customer gets their own personal control plane. ECS is a massive shared tenancy system. When you think of something like "deploy history" its actually a tremendous amount of data once multiplied across all the customers using ECS. Container orchestration data accumulates FAST, and the team building ECS spent most of it's early years just keeping up with the growth in customer base and usage, and the amount of data this represented. There were countless internal system improvements along the way, which customers never saw as anything other than great uptime and continuous service instead of unexpected downtime. One of the tradeoffs that the team made was continuously deleting historical data over an hour old so that it didn't accumulate and add to the burden. The general philosophy was that historical data was better stored in external, durable state storage like CloudWatch. That's why ECS got the EventBridge integration which you can use like this to sync and store all that history data into CloudWatch: https://lnkd.in/g2YyREMx (And then you can use CloudWatch Log Insights to query this data along any dimension you want) I'm super glad to see the team now able to go from one hour history to 90 day history on service deployments, even if it did take a decade to get there. I'm sure there's a lot more to come!

74

1 commentaire

Effortless Cluster Management Let's be honest, managing Kubernetes clusters can be a real headache. It's complex, time-consuming, and requires specialized expertise that not every team has readily available. But with AWS EKS, you can say goodbye to those infrastructure worries and focus on what truly matters: building and deploying your amazing applications. #Containers #devops #insights #Kubernetes

1

Observability with OpenTelemtry Serie: (Zero-Code instrumentation) with Jaeger in Java #Observability #OpenTelemtry #Java

11

Uber's Reference architecture, key points : - OSS strategy, mostly apache licensed (PS: we have the same strategy, it works! ), - Kafka and HDFS is the foundation (PS: we have the same foundational architecture, it is solid! ), - Kafka federation: Interesting, good idea but how? Using protocol-aware standard solutions (eg: Envoy) or in-house developed solution? This give us some idea of introducing Envoy's Kafka-aware proxy into production system, - Flink + Kafka? Why not, but is it worth of it?

1