Introduction Recently I worked on a project with the Serverless Framework which I really like. I used it do define a serverless application for a customer that runs their workloads mostly on AWS but uses Azure DevOps to build and deploy resources. This came about because they were already building lots of .Net applications and are therefore used to VSTS, the former name of Azure DevOps.
After recovering from the initial shock of having to use an Azure tool to deploy resources to AWS I gave it a chance (well, I didn’t really have a choice in the matter).
The serverless kingslayer? - Migration from serverless to CDK Last day I have been at a customer and suggested using the CDK for Infrastructure as code. He responded with a huge yes. He has worked his way through the CDK Intro Workshop aws-cdk-examples/typescript at master · aws-samples/aws-cdk-examples · GitHub and wanted to use it.
Then we discussed a lambda function. The customer was using the serverless framework - like most of the people.
Was ist sind Managed Services der nächsten Generation? Die Cloud-Technologie verändert die traditionellen IT-Modelle auf verschiedene Weise - von der IT-Infrastruktur bis zur Entwicklungsmethode - und hat sogar Auswirkungen auf die Unternehmenskultur. Die Einführung von Cloud-Technologie führt somit auch zu einer neuen Art von IT-Serviceanforderungen, diese werden als Managed Services der nächsten Generation bezeichnet.
Traditionelle Managed Services konzentrieren sich normalerweise auf:
Patch Management Change Management Incident Management Business Continuity Security Access Management…etc .
Should we use CDK for the next project?
The AWS CDK (Cloud Development Kit) is great - and still in beta - so the question for the next project would be:
Should we use cdk or not?
This post will look at the state of the cdk. Also we provide our
CDK Templates updated to the CDK Version 0.38 if you want to get started.
Statistical Computing on Your Local Workstation Recently, a costumer told me about his problems to fullfil stastistical computing workloads on his local workstations. At first, you need to know that statistcal computing language like R and Python by default only use one core of your multi core machine. To parallize them, you have a lot of opportunities with regard to CPU-usage, IOPS etc. Second, even if you parallize your workloads, you got an annoying loud machine in your office.