Docs

Elastic Compute Service

Overview

Features overview

tip

The "Beta" products or features are not open to all users. To facilitate understanding of products, some documents for "Beta" products or features have been made available to all users. Note that certain documents for "Beta" products or features are only accessible to users with access permission to such products. If you need to try Beta products or features, please contact your account manager to apply.

Category	Feature	Description	Related documents
Billing rules	Pay-as-you-go	Pay-as-you-go: Without paying for a large amount of resources in advance, you can provision and release resources on-demand at any time.	Pay-as-you-go
		Subscription	Pay in advance to maximize cost savings.	Subscription
		Spot instance	A unique Pay-as-you-go method that saves up to 80% of costs compared with standard Pay-as-you-go. However, the price fluctuates based on market supply and demand in real time, and the resources purchased by this billing method may be reclaimed due to price or inventory considerations.	Spot instance
		RI	A type of voucher with deduction and reservation capabilities, primarily used to offset the bill of computing resources (vCPU and memory) for Pay-as-you-go instances. By using it, you can significantly reduce the cost of using Pay-as-you-go instances.	Offsetting resource overview
		Elastic Scheduled Instances	A type of Pay-as-you-go instance that can be reserved in advance. It ensures stable supply of resources and saves costs in elastic scenarios.	ESI overview
Instances and images	Instance	An instance is a virtual machine that consists of essential computing components such as CPU, memory, operating system, network, and disk. It is a secure, stable, elastic, and scalable IaaS cloud computing service that helps users lower IT costs and increase O&M efficiency.	Instance overview
		Image	An image is a special file that contains the basic operating system and application data required for an ECS instance.	Overview
		HPC cluster	RDMA network is added to the original instance specifications, achieving logical isolation for HPC instances.	HPC cluster overview
Storage	Volume	A highly available, highly reliable, high-performance, elastic, and scalable block storage device. You can use it as scalable disk storage for ECS and Vital Kubernetes Engine services.	Elastic block storage overview
Storage		Snapshot	A backup method where each snapshot is a backup file of volume data at a specific point in time. When data is lost or system fails, you can roll back volume data by using the snapshot feature to recover the lost data.	Snapshot overview
Network and security	Virtual Private Cloud	Virtual Private Cloud (VPC) builds an isolated, self-configured, and self-managed virtual network environment for cloud resources. Networks of VPCs are isolated from each other.	VPC overview
		Network interface	A network interface is a virtual network interface (VNI). It is a network interface mounted on an instance (such as an ECS instance) in a VPC. By default, a private IP address is allocated from the VPC.	Network interface overview
		IP address	After an IP address is assigned to a network interface, other resources in the network can use this IP address to accurately locate the ECS instance that mounts this network interface. Then, they can communicate with it.	IP address overview
		Security group	A security group consists of multiple security group rules and is designed to control the outbound and inbound traffic of network interfaces. Security groups are designed based on the allowlist principle. By default, security groups deny the inbound and outbound traffic of network interfaces. You can add security group rules as needed to allow traffic.	Security group overview
		EIP	Elastic IP (EIP) is an IP connectivity service that can be separately purchased and held for cloud resources. After an ECS instance is associated with an EIP, the ECS instance can communicate with the internet.	EIP overview
		Key pair	SSH key pairs are a secure authentication method, including a public key and a private key. This feature is supported for Linux instances only.	Key pair overview
Elasticity and deployment	Launch template	A collection of instance configuration information used for purchase of instances. In the template, you can store all configuration information except passwords, including key pair, compute specifications, volume, network, etc. If all configuration information is valid, you can purchase an instance with one click and do not need to reconfigure it.	Overview of instance launch templates
Elasticity and deployment		Deployment set	A strategy that controls the distribution of instances across underlying hardware. It enables physical security isolation. By creating deployment sets, you can distribute instances across different physical servers. This ensures underlying disaster recovery capabilities and high availability.	Deployment set overview
Maintenance and monitoring	Task management	This feature enables centralized management and viewing of the progress and details of custom image creation/replication/import/export tasks in a region.	Managing custom image tasks
		Instance system logs and screenshots	When the instance fails, users can view system logs and screenshots without logging in to the instance. This makes fault diagnosis easy.	Obtaining instance system logs and screenshots
		Cloud assistant	Cloud assistant is a native tool used for automated O&M. With the cloud assistant, you can perform operation commands on one or multiple objects. You do not need to input passwords, log in, or use a jump server.	Cloud assistant overview
		operation and maintenance orchestration	An automated O&M service for cloud products. You can orchestrate task flows based on your needs and execute them automatically. You can also centrally manage orchestrated tasks and view the execution progress and results.	Orchestration overview
		Metric monitoring	Provides monitoring data for the CPU, memory, inbound and outbound network bandwidth, and disk I/O bandwidth of your ECS instances.	Viewing monitoring data of instance
		Event monitoring	When an instance or task triggers a system event (for example, a system failure causes an instance to be redeployed), you can view event details and take action.	Event monitoring overview
		For details, refer to Viewing instance health status.	Instance health monitoring enables automatic detection of software and hardware running and collection of instance health data. This feature helps you check the health status of your instances on the console, making it easy for you to check and address issues.	For details, refer to Viewing instance health status.
		Instance diagnosis	Used to detect performance, network, disk, and other statuses of an instance. This feature helps you identify and understand common issues with ECS instances to ensure normal operation of instances.	Instance diagnosis overview
Account and resource management	IAM	A free service provided by VolcanoEngine. It is a permission management system used to control access to cloud resources for different identities. This service offers access management for resources and helps you improve management efficiency and reduce information leakage.	IAM overview
		Tag	By adding different tags to cloud resources of different types and purposes, you can label and categorize cloud resource instances. This helps you search for instances based on certain purposes more quickly.	Tag overview
		Project	A resource management approach that allows you to group and manage resources based on businesses or projects. This helps you maintain resource independence, data security, and cost efficiency.	Managing projects

Last updated: 2026.06.18 16:37:45