Because Sametime on Kubernetes has documented monitoring, I'd prefer that. Also, I like the scalability option of Kubernetes. But my service provider says Kubernetes takes more resources.
I'd like to know how much without having to build two environments. ;-)
Anyone have experience how much overhead Kubernetes adds?
Hi Lars,
Normally we suggest Docker for small environment. Please check further details:
Is Docker better than Kubernetes?
Docker and Kubernetes have different goals and outcomes. The decision should depend on your end goal. Docker is about containerization; Kubernetes is about container orchestration. The Docker containers are the building blocks. A small deployment has the bare minimum number of containers to produce a solution that allows users to collaborate, but without scalability, load balancing, etc. If you are deploying for a small group and do not require scalability, Docker is a simple solution. Kubernetes builds on the base containers and allows them to auto scale; load balance across multiple physical or virtual hosts, even bringing up and tearing down new hosts to deal with load variance.
Meetings are sized based on what is happening on the server at any given time. There are many variables including the CPU and memory size of the Docker instance. A large Docker instance can support up to 200 concurrent peak users, but that does not take into consideration how many of the meetings are being recorded. Meeting recordings are CPU intensive. A Docker instance is best suited for a small department or focused team than an enterprise size deployment.
You can overcome these limitations through Kubernetes autoscaling. Autoscaling adds and removes nodes as needed based upon current usage and monitoring. Kubernetes is recommended if you are deploying Sametime meetings for your entire organization. For sizing and deployment related questions, contact HCL.
Platforms
https://help.hcltechsw.com/sametime/12/admin/c_planning_platforms.html
Talk to a Technical Expert
https://volt.myhclsandbox.com/volt-apps/anon/org/app/7fc42e3c-19f4-48df-87fc-09d0ac7c3296/launch/index.html?form=F_Form1
Hi Rafael, especially for smaller environments, being able to have simple dashboarding is big factor in choosing Kubernetes over Docker.
But I repeat: how much overhead does having a Kubernetes implementation have over a Docker implementation, because just a nebulous "it costs more" does not help in deciding if we want to spend on those features or not. How much extra cores and GBs of RAM does the orchestration layer need to function? I'd like to have some idea to be able to put in the proposal I am preparing for my manager.
I'd really like some numbers. We already have two VMs running the old Sametime 11 server, so we can reuse them for a kubernetes master and worker node. But since it is Docker based, no nice reporting on them...
@Lars Berntrop-Bos
to add to the explanation that Rafael has provided
With Docker Install, you will need to have an LDAP server, and a server for Mongo.
and then a machine where you will install Sametime on Docker.
with Kubernetes install, you will need the same LDAP and Mongo servers,
but instead of just one additional server as in Mongo install, you will need at least
4 additional servers
1) for you Controller,
1)for your main
1) for your Video
1) for you recording
Information on installing the Kubernetes is available here:
https://help.hcltechsw.com/sametime/12/admin/installation_sametime_kubernetes.html
AS far as How much extra cores and GBs of RAM does the orchestration layer need to function?
then I would recommend that you follow this link provided
https://volt.myhclsandbox.com/volt-apps/anon/org/app/7fc42e3c-19f4-48df-87fc-09d0ac7c3296/launch/index.html?form=F_Form1
as the answer to that question will depend on information from your needs.
as example, in a PoT environment that I worked on
Each VM required 4 CPU, 32GB RAM, 200GB HDD
and one machine used for control node, needed 500GB HDD