Auto scale
Existing autoscale solutions fall short
Traditional autoscaling or scaling groups offered by cloud providers are triggered by machine parameters like CPU, memory, and HDD, rather than by actual core requirements. Moreover, these solutions require oversight and management – getting machines up and running, handling issues, and maintaining DevOps and SecOps, etc. Because Incredibuild is both aware of your workloads and connected to your cloud, we take autoscaling to the next level.
Right out of the box, Incredibuild automatically monitors the amount of cores needed at any given moment. This means that Incredibuild can dynamically and optimally allocate or de-allocate spot or on-demand instances based on your predefined preferences, workload demands and core availability. And unlike scheduling or cloud provider autoscaling, Incredibuild automatically spins up new instances and transfers cache – ensuring seamless processing continuity.
Auto fallback
A fallback strategy made easy
The fallback mechanism allows for up to three desirable instance types that will be requested and deployed over different Availability Zones. If the primary instance type cannot be provisioned across all defined Availability Zones, Incredibuild will work within a defined policy to provision a machine based on a secondary instance type choice.
If this instance type is also not available, then it will attempt to provision a machine based on a third choice.
If Spot instances were selected and none are available based on the fallback policy, Incredibuild will spin up an on-demand machine instead.
Behind the scenes, Incredibuild will keep trying to spin up a spot instance and once one becomes available, it will terminate the on-demand machine and switch to use a spot instance.
Download Data SheetWhy Incredibuild for Cloud?
-
Beyond cloud
provider offeringsUnlike built-in autoscaling solutions that trigger work based on machine parameters, Incredibuild
automatically scales based on the unique core requirements of your workloads at any given time. -
Seamless
orchestrationIncredibuild dynamically and optimally allocates and de-allocates the best and most cost-effective mix of spot or on-demand instances based on your preferences and resource availability.
-
Keep your spot
With Incredibuild, handling and recovery is automatic, straight out of the box. Incredibuild automatically spins up a new instance and transfers the cache to ensure seamless continuity. No more “cold” machines – Incredibuild ensures you get the optimal machine type, every time.
-
Lower bill shock
Incredibuild delivers up to 35% decrease in cloud cost while optimizing performance speeds. For example, instead of taking an expensive 64-core machine for 1 hour, Incredibuild enables you to leverage multiple readily-available cheaper machines, using 640 cores for 6 minutes, instead.
What our clients say
Invrsion eliminates reservation time, guarantees fallback
“During the COVID-19 lockdown, it has taken as much as 3 days to allocate a single g3.8xlarge EC2 instance. With Incredibuild instead of having to use 64 cores (G type), we can take 16 core from a cheaper instance type (C or M), which is cheaper and more available. If a machine is not available, the built-in fallback mechanism automatically finds us the best machine.“
Michele AntoliniCTO
Minitab accelerates builds, tests, optimizes costs on Azure
“Incredibuild Cloud offers unmatched agility. The ability to automatically scale up and down in the (Azure) cloud, while accelerating our development, makes everything so easy… we are optimizing our resources according to our exact needs, so nothing goes to waste”
Anthony BianucciGlobal Business Information Systems
Proletariat accelerates builds on UE, optimizes cost on AWS
“Employees’ home bandwidth could not adequately handle our usual data load traveling back and forth to our in-office resources. Massive amounts of data and VPN saturation were making it very hard to improve everybody’s workload all across the board. We needed a solution that would solve this challenge to ensure we met our launch deadline”
Avida MichaudEngineering Lead
