The KubeAdvisor Framework 1.0 helps teams adopt best practices to accelerate the adoption of Kubernetes and the cloud-native stack, and optimize their existing stack, with machine learning.
The cloud-native stack is quickly becoming very complex. It is hard for teams of different sizes to keep up with such complexity and be production-ready under the typical tight schedules. Magalix built KubeAdvisor for teams focused on making performant applications and production-ready infrastructure, by advising and making it easy to adopt cloud-native best practices and design patterns.
KubeAdvisor focuses on scanning Kubernetes resources, state, and configurations with the goal of making both infrastructure and cloud-native applications reliable, resilient, and properly observable.
KubeAdvisor reports relevant issues for Devs and DevOps to immediately take action on while providing recommendations to reach the minimum in cloud-native standards. It utilizes all of the currently existing Kubernetes APIs and other interaction points to provide a safe and reliable environment to develop and run different advisors.
Some typical scenarios or questions that KubeAdvisor can help users with include:
- Do I have the right observability stack installed and properly configured?
- Do I have the right observability for my containers, micro-services, and applications?
- Do I have any outdated container images?
- Are there any vulnerability issues recently reported that I need to take care of immediately?
- Am I using the right controllers for the right containers?
- Do I have the right resources and limits set on my containers? Am I risking throttling? Am I over-paying for infrastructure?
In the example scenario above, the KubeAdvisor ML algorithm has analyzed the current allocation of nodes and made a safe recommendation that will reduce cloud costs for this one cluster by over $1000 a year, while maintaining safe levels of CPU to prevent container throttling. Finding the right mix of cluster nodes, compute and memory settings to optimize performance and reliability while controlling for costs is hard – we made KubeAdvisor specifically to help devs and DevOps figure out what works for them, quickly and easily.
“We realized that many teams either get stuck or spend a long time in their early stages of the cloud-native journey. “ said Mohamed Ahmed, Magalix CEO and Co-Founder. “Even after going to production, they still miss a lot of points to make it performant, compliant, secure, and reliable. We decided to build a smart companion that works for teams to accelerate their innovation pace and gain 10X confidence in their cloud-native stack.”
KubeAdvisor is now available for FREE through Magalix. You can sign up directly from magalix.com or install Magalix agent from the GCP Marketplace or Azure marketplace.
Magalix is a Seattle-based startup that focuses on Kubernetes optimization. Magalix helps you adopt Kubernetes best practices for reliability, performance, security and cost savings. Using machine learning, over time it proactively makes recommendations to optimize your cluster from various aspects and help you save up to 60% on your infrastructure costs.