PAAS - Kubernetes

Kubernetes (aka k8s) is an Open Source system for automating deployment, scaling and management of containerised applications. It was originally developed by Google, who donated it to the Cloud Native Computing Foundation (CNCF). The software aims to provide a “platform for deploying, scaling and maintaining application containers on distributed hosts”. It supports a range of container tools, including Docker.

Orchestration using Kubernetes is supported by many leading cloud platforms, including Microsoft Azure, IBM Bluemix, Red Hat OpenShift, Amazon AWS and Oracle OCI.
Kubernetes uses so-called “pods” as its smallest usable unit for orchestration. Pods are worker processes that run on “nodes” (i.e. physical or virtual machines within a cluster). Each pod contains at least one container. If there is more than one container in a pod, the containers share container runtime and the allocated resources between them.
The cluster, including all its nodes, is controlled via a dedicated machine known as the “Kubernetes master”, which uses these running “kubelets” to communicate with the individual nodes. An instance of etcd, the central key value database for all the information required to manage the cluster, runs on the Kubernetes master, as do the automated controller processes and a “scheduler,” which assigns newly-generated pods to nodes.
The controllers monitor and control the cluster, along with all its component parts. If a node goes down, for instance, the controller can replace it with an identical one.

Source: Translated from the German version of Wikipedia

 


CARGO-MANAGED KUBERNETES CLUSTER BY SWISSTXT


A managed cluster in the SWISS TXT Cloud, based on Cloudstack and Kubernetes, with a guaranteed capacity of 128GB RAM and 2.5 GHz - 36 CPUs.
The cluster is provided as a managed service, including continuous maintenance, patches and change-management work, carried out in close consultation with the customer.
We recommend the use of our high-performance storage system (Performance Storage Fibre Channel 15K disks or SSD). Additional storage systems are available in the following specifications purely for data delivery purposes:


  • Medium storage
  • ECO storage
  • S3 object storage


Overview of additional features included in the managed cargo cluster package. As these features are part of the service included with the cluster, they will consume some system resources (such as computing power, memory and storage).

 

 

Cargo : Features hosted in Cluster (cargo-system)

 

 

Logs (stored in Elasticsearch)

- Access Logs (Traefik)
- Container Logs

 

 

Storage

- Integrated NAS Storage (Persistant Volumes)
- S3

 

 

Traffic Distribution (Ingress Controller/LB)

- Traefik Ingress Controller (HTTPs)
- Cloudstack TCP Loadbalancer

 

 

Security

- Predefined RBAC Access Control
- Automatic Let's Encrypt

 

 

Monitoring

- Prometheus
- Pod Autoscaling on Metrics

 


MASTER NODES

The master nodes are scaled to 3 nodes. They are provided and maintained in the cloud by SWISS TXT.
The service fee includes:

Components : 

  • Etcd
  • Kubernetes API Server
  • Kubernetes Controller Manager
  • Kubernetes Scheduler
  • Flannel (Container Netzwerk)
  • Kube-Proxy (Service Kommunikation)
  • CCM (Kommunikation zu Cloud)

System Size:

4 CPU
16 GB Memory

Scale:

3 Nodes


WORKER NODES 


Worker nodes are provided and maintained in the cloud by SWISS TXT.
Billing is according to the scale of the cluster, and can be adjusted upon request.

Components : 

  • Docker (Container Runtime)
  • Kube-Proxy (Service Kommunikation)
  • Kubelet (Container Controller)
  • Flannel (Container Netzwerk)

System Size:

4 CPU
16 GB Memory

Scale:

X Nodes

 


SUMMARY

Kubernetes is a system for orchestrating software containers. It automates container control and management. This means that Kubernetes takes over responsibility for starting, stopping, monitoring and updating each individual container on the basis of a detailed description of exactly how many of each container are required. Everything else happens automatically, so developers hardly need to interact with the hardware or servers.


 

YOUR CONTACT

Bettina Manta

Bettina Manta

Sales & Service Manager
+41 58 136 41 03


INTERESTED?

Your e-mail address will not be communicated to third parties.
Data Privacy