Serverless ApsaraDB RDS for PostgreSQL instances offer real-time scalability to meet dynamic business needs, enabling rapid and independent scaling of computing resources to handle fluctuating workloads, minimize resource waste, and reduce operational and maintenance (O&M) costs. This topic describes the features, architecture, and usage of serverless RDS instances, which help to lower costs and enhance O&M efficiency.
Introduction
RDS PostgreSQL Serverless instances offer real-time elasticity for CPU and memory, representing a new category of RDS PostgreSQL product within the disk architecture. These instances provide vertical resource isolation for network resources, namespace, and bucket, and feature on-demand billing for computing resources. They are characterized by low resource usage, simplicity, flexibility, and cost-effectiveness, empowering users to scale computing power quickly and independently during business peaks and troughs. This ensures a rapid response to business changes while optimizing costs, further aiding enterprises in cost reduction and efficiency improvement.
The computing resource unit for RDS PostgreSQL Serverless instances is the RCU (RDS Capacity Unit). The performance of 1 RCU is roughly equivalent to 1 core and up to 2 GB of memory. The instance's computing resources will automatically scale within the RCU range you specify.
The maximum connections for RDS PostgreSQL Serverless instances are fixed at 2400. This limit cannot be adjusted and remains constant regardless of RCU.
In scenarios with significant business fluctuations, the resource usage and specification changes of regular instances and Serverless instances are illustrated in the following figure:
From the above figure, it is evident that in scenarios with significant business fluctuations:
Regular instances: Resources are often wasted during low-activity periods, and may be insufficient during peak periods, impacting business operations.
Serverless instances:
Adjust computing resources in real-time based on business demand, minimizing wasted resources, enhancing resource utilization, and reducing costs.
Adequately meet business needs during peak periods, ensuring uninterrupted operations and improved system stability.
Move away from fixed resource payment models, achieving true pay-as-you-go with dynamic matching of load and resources, leading to significant cost savings.
Eliminate the need for manual configuration changes, improving O&M efficiency and reducing management and development costs.
Support automatic start and stop features. If no connections are established, the instance automatically suspends to release computing resources and reduce costs. Upon establishing a connection, the instance automatically starts.
Optimized for scenarios with high-throughput writes and high concurrency, providing elastic scaling capabilities suitable for businesses with large data volumes and typical peak and trough access patterns.
Advantages
Lower cost: Ideal for startups, PostgreSQL Serverless requires no reliance on other infrastructure and related services. It is ready for immediate use and provides stable, efficient data access services. Costs are based solely on resource usage.
Larger storage space: Storage capacity can reach up to 32 TB and automatically expands based on the instance's data volume, effectively preventing business impact due to insufficient cluster storage resources.
Automatic elastic scaling of computing resources: Computing resources for user read and write operations can be elastically scaled, significantly reducing O&M costs and system risks.
Fully managed and maintenance-free: Alibaba Cloud's professional team handles all O&M tasks such as system deployment, scaling, and alert management. Users experience seamless service with continuous availability, free from maintenance concerns.
Scenarios
Development, testing environments, and other scenarios with infrequent database usage
SaaS applications for small and medium-sized enterprise website services
Individual developer users
Educational scenarios such as school teaching and student experiments
IoT, edge computing, and other scenarios with unpredictable loads
Fully managed services or users seeking a completely maintenance-free solution
Users with fluctuating or unpredictable business demands
Scenarios involving intermittent scheduled tasks
Billing description
Usage method
Change resource scaling range (RCU)
You can adjust the computing resource scaling range of the Serverless instance, including the minimum and maximum RCU, based on actual business needs for optimized resource configuration.
Schedule Serverless instance RCU
If you require stability during specific periods, you can schedule the RCU of the Serverless instance in advance through scheduled tasks to ensure instance stability during those times.
Change instance elasticity policy
For Serverless instance elasticity policies, you can choose the default non-mandatory execution policy to prevent potential service interruptions, or opt for the mandatory execution policy when performance needs outweigh continuous availability.
Set instance automatic start and stop
By enabling automatic start and stop, if no connections occur within 10 minutes, the instance will automatically suspend, with RCU at 0, incurring no computing charges. When a connection is made to the suspended instance, it automatically resumes, and computing charges apply.
FAQ
Q: Why do I receive CloudMonitor alerts related to CPU, memory, and other resources for Serverless instances?
A: The computing resources of RDS PostgreSQL Serverless instances scale automatically within the specified range based on business load. Typically, you should not receive alerts for computing resource thresholds. However, if you have previously set alert rules for CPU, memory, and other thresholds for all RDS PostgreSQL resources before creating the Serverless instance, these rules will automatically apply to the new Serverless instance. When monitoring metrics meet the alert conditions, you will receive the corresponding alert notification.
You can log on to the CloudMonitor console, and on the page, use the advanced filter feature to view the active alert rules for the target Serverless instance and manage them.
