LIVE WEBINAR

Best Practices To Manage Kubernetes Capacity And Performance

August 10, 2020

 

9:00 am PST | 6:00 pm CEST

Capacity Management in cloud-based infrastructure & apps is the science (and sometimes art) of balancing performance, resource allocation & cost management. It can get very tricky when you're cloud-native.

In this webinar, Mohamed Ahmed, Magalix CEO and Founder, will cover the different strategies to manage Kubernetes capacity.

We will cover the top 5 use cases that you need to consider while managing CPU and memory for your pods. We will also dive deep into specific tactics to right-size your VMs, node pools, and pods limits/requests. Finally, we will touch on 3 billing models that you can utilize as you scale your K8s clusters up or down to save on your monthly cloud bill.

Expect to learn:

  • The best ways to model and frame capacity management for Kubernetes clusters and cloud-native apps. 
  • The top 5 scenarios that you should consider while managing the allocation of CPU/memory to different pods and workloads inside your Kubernetes cluster. 
  • The different ways to rightsize your cluster and create effective node pools. 
  • The top 3 billing models that you should consider as you design the scalability of your Kubernetes cluster. 
  • Bonus: The 5 step framework to apply the above concepts easily and quickly inside your Kubernetes clusters

Every attendee will also get a free copy of our Ultimate Guide to Capacity Management eBook.

Mohamed Ahmed, CEO & Co-Founder, Magalix

Mohamed Ahmed is the founder and CEO of Magalix. He accumulated 20 years experience in distributed systems design and performance engineering. He has a Ph.D. in high-performance computing from UCONN. Mohamed worked at Amazon Web Services and shipped two EC2 products in addition to his work on fleet autonomy. He also shipped two products at Microsoft Azure focused on highly scalable cloud-based queuing systems and IoT backend. Mohamed has 13 publications in operating systems architecture and high-performance computing.

Mohamed Ahmed-1