Advanced settings give you full control to customize your service's resources and timeout settings. Once you make a change to any advanced setting, you will be required to Deploy the changes.
View Advanced Settings
The advanced settings tab is located under the Edit Service slide-out menu.
- Login to your KintoHub account
- Create a new service or click on a Service in your Environment
- Click on the Edit Service button at the top right
- Find the tab called Advanced on the far right
How It Works?
Unique settings will be available depending on your service type.
The amount of memory your application has guaranteed access to at all times. Using Metric tab, you see the actual memory usage of your service and fine-tune the amount of memory you want to allocate to your service over time. The lowest value of memory you may use is 32 Mb and the highest value is 4 Gb.
If you require more memory, you may contact support.
By default and when toggled off, the vCPU of your service is shared between KintoHub community.
Shared vCPU may randomly affect the performance of your service and is only recommended for
When toggled on, you will have the option to specify a guaranteed dedicated amount of vCPU for your service. The lowest amount of vCPU you may use is 0.1 cores which is 10% of a dedicated vCPU. The highest amount of vCPU available is 2 which guarantees 2 dedicated vCPU cores.
If you require more vCPU, you may contact support.
The auto-scaling feature is only available when your service has Dedicated vCPU enabled. When enabled, you will be able to set the Min Instances and Max Instances that your service may scale up to. This is known as Horizontal Scaling and is best used for Stateless services. The maximum number of instances you may scale to is 10.
- Min Instances is the number of instances running at all times. Your environment may require High Availability which can be achieved by setting Min Instances to the number of instances you wish to have running at all times.
- Max Instances is the maximum number of instances you will allow your service to scale to. When your service sustains 70% CPU usage for a short period, KintoHub will spawn another instance for you. When your service no longer is using a high amount of load, KintoHub will scale down your instances back down to your Min Instances value.
An instance is a single running version of your application or service. When multiple instances are running, KintoHub balances the API request load between your service instances.
When enabled, KintoHub will deploy your service instances on discounted vCPU and Memory resources. Discounted instances uptime are not guaranteed, and you may incur some downtime if you do not set up at least 2 min instances for your service. Today, it is not recommended to use Cost Optimization for Production environments, but the team is working hard on building more confidence in providing a cost-effective production version of Cost Optimization in the future. Price differences between Cost Optimization versus Standard pricing are described in Billing.
Catalog related services may optionally enable or disable Persistent Volumes. When Persistent Volumes are disabled and your instance restarts for any reason, your data WILL be lost. It is not recommended to disable persistence unless you plan to do a quick test or use a Catalog service for caching such as Redis.
The minimum value of storage you may select is 1 Gb and the maximum is 200 Gb.
Catalog Services that support Persistent Storage:
It is currently NOT possible to increase your Storage amount. Vote on the feature here.
You may never decrease the value of your Storage amount due to the irreversible damage that may occur if you decrease the disk by too much. If you wish to decrease your storage amount, you must create another service and migrate the data.
Max Monthly Cost Estimation
KintoHub calculates the worse case cost of your service based on the advanced configurations values that you have set. The worse case is calculated by your service running 24/7 for 30 days. The actual cost of your service is billed per second. If you delete your service, you will no longer be charged.
When setting up Auto Scaling the Max Monthly Cost is calculated with the Max Instances you have set for your service. The probability of the Max Monthly Cost occurring is low for Auto Scale related estimations because it would be rare for the Max Instances to be running 24/7 for 30 days.
KintoHub believes it is important to know how much your services cost per month and guarantee that you will never be surprised by an unexpected bill.
The team will continue to build notification systems and clarity throughout the platform to inform your team about its cloud costs and tips on optimizing your cost as your application grows.