November 23, 2019

134 words 1 min read

Scale Kubernetes to Support 50,000 Services [I]

Scale Kubernetes to Support 50,000 Services [I]

Kubernetes currently has two load balancing modes: userspace and IPTables. They both have limitation on scalability and performance. We introduced IPVS as third kube-proxy mode which scales kubernetes …
Talk Title Scale Kubernetes to Support 50,000 Services [I]
Speakers Quinton Hoole (Technical Vice President, Futurewei), Haibin Michael Xie (Principal Architect, Huawei Technologies)
Conference CloudNativeCon + KubeCon Europe
Conf Tag
Location Berlin Congress Center
Date Mar 28-30, 2017
URL Talk Page
Slides Talk Slides
Video

Kubernetes currently has two load balancing modes: userspace and IPTables. They both have limitation on scalability and performance. We introduced IPVS as third kube-proxy mode which scales kubernetes load balancer to support 50,000 services. Beyond that, control plane needs to be optimized in order to deploy 50,000 services. We will introduce alternative solutions and our prototypes with detailed performance data.

performance load balancer control plane kubernetes
comments powered by Disqus
Bringing Kubernetes into Salesforce [B]

Bringing Kubernetes into Salesforce [B]

November 23, 2019

I'll be talking about our journey to using Kubernetes at Salesforce, in the World's Largest Enterprise Cloud Infrastructure. Topics will include: Our motivations for choosing Kubernetes, and how we b …
Kubernetes-Defined Monitoring [I]

Kubernetes-Defined Monitoring [I]

November 23, 2019

Over the past few years weve all learned how Kubernetes can dramatically change the process of deploying an application, improve reliability, and accelerate operations. As Kubernetes matures, I belie …
Success of CRI: Bringing Hypervisor Based Container to Kubernetes [I]

Success of CRI: Bringing Hypervisor Based Container to Kubernetes [I]

November 23, 2019

CRI, aka Container Runtime Interface, is the new mechanism designed to plug any container runtime in Kubernetes. In this presentation, I will introduce how we introduce the hypervisor based container …
Load Testing Kubernetes: How to Optimize Your Cluster Resource Allocation in Production [I]

Load Testing Kubernetes: How to Optimize Your Cluster Resource Allocation in Production [I]

November 22, 2019

So you've carefully crafted your first Kubernetes service, and you're ready to deploy it to production. Well, not quite: there are still some important unknowns to understand before your service will …
Sched.net: A Network-Aware Kubernetes Scheduler [I]

Sched.net: A Network-Aware Kubernetes Scheduler [I]

November 22, 2019

Different workloads have different optimal scheduling requirements. For instance, a video streaming microservice would need higher network bandwidth than a microservice running the codec. Kubernetes e …
Switching From External Load Balancing to consul & ingress [I]

Switching From External Load Balancing to consul & ingress [I]

November 22, 2019

At Concur we integrated our kubernetes clusters to our own internal F5 ecosystem which worked well for internal data center deployments but turned out to be cumbersome when trying to maintain a consis …