AWS provides a Service Level Agreement (SLA) for Amazon Bedrock that outlines commitments for reliability and uptime, alongside support plans that offer technical assistance. The SLA guarantees a certain level of service availability, typically measured as a percentage of uptime over a billing cycle. If AWS fails to meet this commitment, customers may be eligible for service credits. Additionally, AWS offers tiered support plans (e.g., Developer, Business, Enterprise) with varying response times and access to technical resources. These plans ensure developers receive timely help for operational or technical issues affecting Bedrock’s performance.
The SLA for Amazon Bedrock specifies a monthly uptime percentage, often 99.9% or higher, depending on the region and service configuration. For example, if Bedrock’s uptime falls below this threshold in a given month, AWS calculates the service credit as a percentage of the customer’s monthly bill. This credit is applied automatically unless the customer opts out. The SLA excludes downtime caused by factors outside AWS’s control, such as customer misconfigurations or third-party integrations. To monitor uptime, developers can use AWS CloudWatch metrics to track Bedrock’s availability and performance in real time. AWS also publishes historical uptime data through its Service Health Dashboard, enabling teams to verify compliance with SLA terms.
AWS support plans complement the SLA by providing technical assistance to resolve issues that could impact Bedrock’s reliability. For instance, the Business and Enterprise Support tiers include 24/7 access to AWS experts, with response times as fast as 15 minutes for critical outages. These plans also offer proactive guidance, such as architecture reviews or best practices for scaling Bedrock-based applications. For teams requiring deeper integration, AWS Technical Account Managers (TAMs) can assist with optimizing Bedrock workflows or troubleshooting performance bottlenecks. While the Basic Support plan lacks SLA-backed response times, it still provides access to documentation and community forums. Developers should review AWS’s official SLA documentation for Bedrock to understand region-specific commitments and eligibility requirements for credits.
Zilliz Cloud is a managed vector database built on Milvus perfect for building GenAI applications.
Try FreeLike the article? Spread the word