This feature lets you to configure your system in a most dynamic way to automatically size your Cameyo environment for the actual need.
This feature is only available on BYO-cloud environments!
- Prepare a master server
Set up a master server with all applications you need, as you would do for a single Cameyo server
once you are happy with the settings and configured apps,
- go to your servers page and under DETAILS > COMMANDS click [MAKE ELASTIC]
- then an elasticity cluster will be created.
Specify the basic settings for all dynamic created servers within that cluster:
In general, you should use the same instance size for the cluster than what you've been using for the master server. This can be changed later as needed.
you will now be redirected to the elastic cluster page
- Rename your cluster to a meaningful name
- Configure the needed instance features, like SSL, SSD and GPU according to your needs
Be aware that features like SSD and GPU will have an impact on the cost of your cloud provider!
If you configure GPU, make sure the selected region supports GPU otherwise the creation of elastic servers will fail!
Google Cloud Regions/Zones supported by Cameyo
When configuring SSL, each server will get a random name with the pattern: https://<companyname>-<random>.cameyo.net and the certificates will be automatically be deployed.
Now we come to the most important parts of the configuration:
- USAGE MANAGEMENT
Configure working days and times according to the time zone set on the left-hand side in GENERAL, followed by the shutdown timeouts for the server within and outside your work hours. Your elastic servers will be automatically shut down if no user is connected to the server for the specified amount of time.
Tick the Auto power-on checkbox if you want your servers automatically power-on at the beginning of your workday (make sure you configure it a bit earlier than when users start to work, to get the expected effect).
In this part you have to configure how many server you want to pre-create and whether you want to have a certain amount of servers pre-powered-on within work hours.
The most important part is to choose the right baseline snapshot: Always use Dynamic snapshot!
The number of instances (elastic servers) should be self-explanatory
Maximum instance capacity defines the maximum capacity at which instances still accept sessions. In most cases it should remain at 100%.
The Rollover capacity threshold defines when the next server should start to spin up (by either reaching 80% of CPU or RAM). E.g. when your server has 8GB RAM and the threshold is set to 75% and the overall used RAM of this server goes beyond 6GB RAM it starts up the next elastic server.
- Click [SAVE] once you have configured everything
- Elastic servers will be automatically created based on the last dynamic snapshot of the master server
- Elastic servers will be automatically shut down according to the configuration above and will eventually be deleted when no longer needed.
- If you need to make changes or updates to your environment, do not make changes on an elastic clone.
- For changes to be reflected on all elastic clones, make them on the master server and then after the changes click on [RESNAPSHOT]. This will create a new dynamic snapshot for the elastic servers
Beware that taking the snapshot takes some time and that elastic servers with active sessions won't be affected. Once we detect an elastic server has no active sessions, it will be deleted and replaced with a server based on the new dynamic snapshot in consideration of defined Instances available.