Auto Scaling

Auto Scaling (AS) automatically adjusts service resources based on your service requirements and configured AS policies. In this way, the number of Elastic Cloud Server (ECS) instances changes along with service load, ensuring stable service running.

Free Trial Use

Product Advantages

Automation and Flexibility

AS automatically increases or reduces the number of ECS instances based on scheduled, periodic, or alarm-triggered policies.

An ECS instance newly added to an AS group can be automatically added to a listener.

High Availability

AS automatically checks running statuses of ECS instances in an AS group. After detecting an unhealthy instance, AS automatically removes it from the AS group and enables a new one to ensure proper service running.

AS allows you to use multiple availability zones (AZs) and evenly distributes instances among these AZs, ensuring high service availability in the AS group.

Visibility

AS provides overall monitoring charts and scaling change views for AS groups, helping you predict service load and manage O&M.

Low Cost

You are charged only for used ECS resources, reducing O&M cost and protecting your investment.

Product Architecture

Elastic Cloud Server  Elastic Cloud Server  Elastic Cloud Server

Recommended configurations:

Elastic Cloud Server

Elastic Cloud Server (ECS) is provided by HUAWEI CLOUD for you to obtain cloud server resources at any time. These resources can be scaled elastically to help you build a reliable, secure, flexible, and efficient application environment to ensure the stable and continuous running of your services and improve your IT service capabilities and service agility.

AS

AS allows you to configure scaling policies based on service load and it adjusts the number of ECS instances in an AS group to ensure proper service running.

  • For random service load changes, you can use CES to monitor CPU, memory, and network metrics, based on which scaling actions are triggered.
  • For regular and expected service load changes, you can configure scheduled policies, based on which scheduled or periodic scaling actions are triggered.

Application Scenarios

  • Website Applications

  • Data Processing and Computing

Website Applications

Application scenarios:

enterprise websites, e-Commerce, and mobile applications

Characteristics:

  • Using AS, you can add or remove ECS instances for applications as needed. You do not need to reserve a large number of ECS instances for expected marketing activities or unknown service peaks, thereby minimizing the cost for stable system running.

Recommended configurations:

ECS AS ELB RDS

Fixed Performance

Application Scenarios

Video websites, media transcoding applications, media content backhaul applications, high-traffic content management system, and distributed high-speed cache systems

Deployment Suggestion

  • In scenarios where a high bandwidth is required, real-time online backhaul is required for cloud transmission and acceleration services, and video content is processed based on cloud storage and ECS, AS needs to work with the Elastic Load Balance (ELB) and Object Storage Service (OBS).

Recommended Configuration

Elastic Cloud Server :

Multiple ECSs with 2 cores, 4 GB memory, 5 Mbit/s bandwidth, and 100 GB common I/O EVS disks

Function Description

Elastic Scaling

Dynamic scaling: The number of ECS instances is dynamically increased or decreased based on AS group monitoring data and application running statuses.

Scheduled or periodic scaling: The number of ECS instances is automatically increased or decreased at a configured interval or periodically.

Configuration for the Number of ECS Instances

You can configure the maximum, minimum, and expected numbers of ECS instances based on service characteristics to ensure proper service running.

Automatic Load Balancing

An ECS instance newly added to an AS group is automatically added to a listener. Automatic Notification

Automatic Notification

This allows you to receive a notification when your AS group changes so that you can promptly learn AS group running status.

Registration