Before you purchase and use Alibaba Cloud Edge Node Service (ENS), read the pre-sales instructions in this topic.
Node resources
Edge node storage supports local disks and ultra disks.
All local disks for an ENS instance reside on a single host. This increases single point of failures (SPOFs) and data loss risks. The reliability of data that is stored in a local disk is determined by the reliability of the associated host. If you require high data reliability, we recommend that you implement data redundancy at the application layer.
You can use the node resources provided by ENS within the validity period. You cannot use the node resources that are not provided. If you use unauthorized resources on an edge node, ENS has the right to handle the unauthorized resources and related resources.
Specific edge nodes may be eliminated due to factors such as node vendor changes. You will be notified 15 days before a node is eliminated by using methods such as phone calls, emails, and internal messages. You will also be provided an alternative node of the same Internet service provider (ISP) and the same province. After a node is eliminated, services are stopped. You need to back up data, release resources, and migrate services before the node is eliminated.
Node O&M
ENS provides an uptime of at least 99.9% that is defined in the service level agreement (SLA). The uptime does not include the downtime that is caused by the following reasons: exceptions in core networks of the ISP, system maintenance operations that are performed by Alibaba Cloud with prior notice, exceptions in networks and devices outside Alibaba Cloud, and exceptions that are caused by customers.
Edge nodes may be cutover. In most cases, the cutover frequency is no more than once per month per edge node, and the duration of each cutover is no more than 6 hours.
The ISP network may be interrupted or jittered. ENS monitors the exceptions and reports the exceptions to CloudMonitor in real time 24/7. After you configure CloudMonitor, you can receive alerts by using methods such as emails, phone calls, and API operations to obtain the status of edge nodes in real time.
The reliability of an instance that has local disks depends on the reliability of the host. If a SPOF occurs, ENS migrates data to another instance on the same edge node to recover connectivity at the earliest opportunity. Data reliability is not guaranteed.
If an edge node fails to be connected, ENS migrates data to another instance on another edge node to recover connectivity at the earliest opportunity. The instance ID remains unchanged and the instance IP address changes. Data reliability is not guaranteed.