Scenarios

更新时间:
复制 MD 格式

Storage and computing capacity are coupled with nodes, which means you must add more nodes to increase either capacity. For systems with low traffic, such as log storage and backup systems, purchasing nodes only to increase storage capacity results in wasted resources and unnecessary costs. The Serverless edition of Hbase is designed for these use cases and can significantly reduce costs.

Small to medium-sized and lightweight users

The Serverless edition of Hbase is ideal for small and medium-sized businesses that require a pay-as-you-go solution. You do not need to purchase an entire cluster. Instead, you pay only for what you use. For short-term projects, you can reduce costs to less than 10% of the price of a traditional cluster. Services can be started and stopped in seconds, which provides high flexibility for your applications.

Users with scheduled tasks

The inherent elasticity of the Serverless edition of Hbase makes it ideal for customers with scheduled tasks. For example, you might process requests at a set time each day or run data analytics at night. This service model provides flexible data access while minimizing backend operations and maintenance (O&M) and costs. The following figure shows an example where a user experiences high request volumes on Mondays and Fridays, with stable traffic for the rest of the week. Serverless Hbase is well-suited to handle this pattern, eliminating the need to frequently scale out to manage short-term growth.

Users with fluctuating workloads

Some services, such as E-commerce and notification delivery, experience sudden traffic spikes. For example, you might need to process a high volume of transactions or notifications during a promotion. Preparing nodes or planning capacity in advance for these spikes is time-consuming and costly. The Serverless Hbase model is also an excellent solution for these use cases. The following figure shows an E-commerce customer who receives a massive volume of requests during the Double 11 shopping festival. After the event, traffic returns to normal levels. The Serverless model can reliably support this type of fluctuating workload.

Users who want a fully managed service

The backend of the Serverless edition of Hbase is managed by Alibaba Cloud's experienced O&M team. The team expertly handles all aspects of cluster O&M, including upgrades, performance tuning, and troubleshooting. This provides small and medium-sized companies with a fully managed cluster service, ensuring that the backend service is always efficient and stable.