Use elastic scaling to manage the back-end server cluster of the website, and the cluster provides external services through load balancing. When the visit pressure of the website fluctuates greatly, the elastic scaling can automatically increase the cluster nodes when the server load is too heavy, and release idle nodes in time when the load drops, so that the website performance is guaranteed and the cost is maximized.
Auto Scaling
It offers a flexible and economic cloud resources management mode. Based on the server load and network traffic and other metrics, you can perform auto scaling of cloud resources and auto association of load balance to optimize the cloud service performance and cost.
- Overview
- Features
- Application Scenarios
- Advantages
- Customer case
- User's Guide
Overview
You can manage your BCC resources through auto scaling. By defining the scaling rules, you can create or release a certain quantity of BCC instances where there is a change in the scheduled time points or cloud resource monitoring metrics. Thus, it has sufficient computing capability when the volume of business increases suddenly, and releases resources when the volume of business decreases. Meanwhile, the auto scaling allows you to remove the exceptional instance and automatically associate the load balance. It effectively helps you better manage the cloud resources and minimizes the human resources and operations cost.
Features
On-schedule Scaling
It allows you to scale up or down the resources at the specified time. It supports you scaling up or down the number of specified instance or adjusting the number of instances to a fixed quantity. With this feature, you can prepare sufficient resources in advance to cope with the peak access volume.
Alarm-triggered Scaling
When any cloud resource monitoring metric exceeds the threshold, you can perform the scaling. This feature supports you performing aggregation computing of monitoring metrics for a group of BCC instances, and also monitoring a single BCC or EIP and other resources.
Auto Removal of an Exceptional Instance
This feature can automatically scan the status of the instances in the auto-scaling group. When a node gets exceptional, it releases resources timely. When there are fewer instances, you can perform scaling in time to maintain the computing ability.
Auto Association with the BLB
This feature allows you to specify the associated load balance for the auto scaling group. During scaling-up, the new instance gets added automatically to the backend pool to ensure that the new instance responds to the external request in time.
Application Scenarios
Build website service
Build task processing cluster
Advantages
Automated Operations
You need to set the rules only. No manual intervention is required for auto scaling of the BCC and for association with the BLB.
Reduce Resources Cost
Maximize the resources utilization, and timely release redundant resources and exceptional instances to significantly reduce the cost.
Improve Service Availability
Auto scaling takes place based on the load. It guarantees that the service performance increases with the business volume, and enhances the business continuity.
Customer Case
Qi Yi Technology
Qijia.com was established in 2007. Its business covers decoration platform, self-operated decoration business, Qijia franchise network, supply chain and many other sub-ecologies. It is the preferred platform for customers seeking home improvement services on the Internet. Elastic scaling AS can dynamically adjust cluster nodes regularly or according to server load, thereby ensuring website performance, increasing service availability, and helping users greatly reduce costs.
Butterfly interaction
Butterfly Interactive is a company dedicated to using mobile Internet to change the world. It is a professional company focusing on mobile Internet applications, mini games, and H5 game development and distribution. At present, Butterfly Interactive has become a leader in the field of H5 and mini games. Elastic scaling monitors the status of the task source. When the business peaks, it automatically expands the cluster to speed up the task processing speed. When the business peaks, it automatically releases resources to save resources, ensure the availability of the system and minimize costs.