TWIL: September 11, 2022
When you change jobs, you go through intensive learning days, and that was the case for me too, for the past two weeks. However, most of what I’ve been learning is either confidential or of no interest to anyone outside the company. Still, I’d like to highlight a set of podcasts about Geo-distributed Cloud Apps, AI Scaling though mixture of expert models, the migration of Azure App Service to Kestrel and Microservice Architectures. Also, checkout an interesting video on Azure Landing Zones, and a set of cloud architecture mind maps which I found very useful to understand the whole Azure landscape. Enjoy!
Episode 856: Building and Scaling Geo-Distributed Cloud Apps with Denis Magda
Denis started at Sun Microsystems and Oracle where he worked on JVM/JDK and led one of the Java development groups. After learning Java from the inside, he joined the world of distributed systems and databases, where he’s remained ever since. He talks to Scott about how to build large horizontal cloud apps that are geo-distributed and truly global. Cloud availability zones and regions are not immune to outages. The zones go down regularly, and regions become unavailable during natural disasters or human-caused incidents. If an availability zone or a larger area goes down, so does your application…unless the application functions across multiple geographic locations. We’ll discuss availability and reliability patterns used by architects whose apps managed to withstand major cloud outages.
The Stack Overflow Podcast
Episode 478: What companies lose when they track worker productivity
The home team gathers for a conversation about workplace productivity monitoring: Does it motivate employees to get more done, or does it lead to stress that takes away from deep, focused work and replaces it with busywork instead?
Towards Data Science
AI scaling with mixture of expert models
What if there was a flexible way to scale AI — one that allowed us to decouple model size from compute budgets, so that we can track a more compute-efficient course to scale? That’s the promise of so-called mixture of experts models, or MoEs. Unlike more traditional transformers, MoEs don’t update all of their parameters on every training pass. Instead, they route inputs intelligently to sub-models called experts, which can each specialize in different tasks. On a given training pass, only those experts have their parameters updated. The result is a sparse model, a more compute-efficient training process, and a new potential path to scale.
The Azure Podcast
Episode 436: App Services on Kestrel
The team catches up with Byron Tardif from the App Services team to learn about App Services’ migration to Kestrel+YARP (Yet Another Reverse Proxy).
Episode 1809: Microservices Architectures with Shawn Wildermuth
What’s wrong with microservices? Carl and Richard talk to Shawn Wildermuth about his rant about microservices. Shawn talks about the intent of microservices in the first place, to try and break down the giant service balls of goo that get built over time. But is it necessary? The conversation explores the optimization problem, where having services together is efficient right up until it isn’t – when you have a service that changes more often than others or needs to scale more. Only then does it make sense to carve it out. Lots of fun conversation!
Episode 1810: Testing Angular Forms with Martine Dowden
How do you test Angular forms? While at CodePaLOUsa in Louisville, Carl and Richard talked to Martine Dowden about her approach to building tests that are maintainable, and are best automated because they are tedious to test manually – like forms validation. Martine talks about a mix of automated unit testing and eyes-on manual smoke tests being the most efficient way to have a well-tested web application.
Azure Landing Zones: Architectural Blueprint, Tooling & Best Practices
As you build new Cloud-based services and migrate or modernize, check out the new tools to assess where you are in your Cloud journey and how to automate the build out of recommended foundational services to host your workloads in Azure. We take a look at updates to the Azure Landing Zone guidance, that gives you proven best practices, architectural blueprint and tooling.
The Azure Cloud Native Architecture Mapbook
Azure offers a wide range of services, providing a million ways to architect your solution. This Azure Architecture Mapbook uses original maps and expert analysis to help you to explore the capabilities of the Microsoft Cloud platform and choose the best solutions for your unique requirements. It was released in 2020 so some of the diagrams are missing new services or capabilities, but it’s still an awesome reference.
- Azure Solution Architect Map
- Azure Infrastructure Architect Map
- AKS Architecture Map
- Azure Application Architecture Map
- Azure Data Architecture Map
- Azure Security Architecture Map
Front Matter CMS
Front Matter is a CMS that runs within Visual Studio Code/GitPod/… It gives you the power and control of a full-blown CMS while also providing you the flexibility and speed of the static site generator of your choice. Jump right into editing and creating content with Front Matter and be able to preview it straight in VS Code.
Hugo is a general-purpose website framework. Technically speaking, Hugo is a static site generator. Unlike systems that dynamically build a page with each visitor request, Hugo builds pages when you create or update your content. Since websites are viewed far more often than they are edited, Hugo is designed to provide an optimal viewing experience for your website’s end users and an ideal writing experience for website authors.
Jekyll is a static site generator. It takes text written in your favorite markup language and uses layouts to create a static website. You can tweak the site’s look and feel, URLs, the data displayed on the page, and more.
Project Natick seeks to understand the benefits and difficulties in deploying subsea datacenters worldwide. Phase two extends the research we accomplished in phase one by deploying a full-scale datacenter module in the North Sea, powered by renewable energy. This is the continuing story of Project Natick.
Have an awesome week!
Photo by Johny vino on Unsplash