Getting Started With Elastigroup Autoscaler

Product Marketing Manager

November 9, 2016

Reading Time: 2 minutes

Amazon Web Services

Reading Time: 2 minutes

Elastigroup’s auto-scaling delivers unprecedented functionality and has been designed specifically for running economical workloads at the maximum performance that scales up and down dynamically based on the application’s load.

The manner by which Spotinst enables this resiliency is through the use of sophisticated algorithms and a highly developed workload management strategy that ensure maximum up-time at the absolute minimal cost.

Some of the features enabled by Elastigroup are:

Accurate workload predictability
Earlier and more advanced termination notifications
Highly effective Spot bidding strategies
Workload optimization feedback by many dimensions such as machine type, A.Z., load, etc…
Dynamic Scaling Thresholds
Safe Application Draining with AWS Elastic Load Balancer and Google HTTP Load Balancer.
Prioritized utilization of existing, unused Reserved Instances
Advanced Termination policies. e.g. [‘expensive/risky’ instances, Low-performance instances, closest to next hour billing]

Figure A’ shows the real time Reserved Instances utilization, where Elastigroup chooses unused reservations to utilize first.

Scaling based on your metrics

Spotinst Autoscaling gives you tremendous scale up/down flexibility with a broad range of standard and custom metrics to configure your automated provisioning.