1. Purpose

This Fair Use Policy defines acceptable usage of shared infrastructure resources, including network bandwidth and data transfer, to ensure stable, predictable performance for all customers. This Fair Use Policy is incorporated by reference into the Master Hosted Services Agreement (“MHSA”) and applicable Order Forms between Customer and Crate.io.
The goal is not to restrict normal usage, but to prevent patterns that materially degrade service quality for others.

2. Scope

This policy applies to all use of shared resources within the Services, including:

• Data ingress and egress
• Cross-region or cross-cluster data transfer
• API and query-driven data movement
• Replication or synchronization workloads

3. Principles of Fair Use

Customers are expected to use the Services in a manner consistent with typical production workloads for data-intensive applications.

Usage is considered fair when it:

• aligns with the purchased service tier and expected workload profile 

• does not cause sustained resource contention impacting other customers 

• avoids unnecessary or excessive data movement

4. Examples of Non-Fair Use

The following patterns may be considered misuse of shared resources:

• Sustained high-volume data transfer that exceeds typical usage patterns for the subscribed tier 

• Repeated large-scale exports or full dataset transfers where incremental approaches are reasonable 

• Using the Services primarily as a data transport or proxy layer rather than a database system 

• Inefficient query patterns causing excessive data egress (e.g., repeated full scans without filtering) 

• Workloads that generate disproportionate network load relative to compute/storage usage 

• Automated or uncontrolled processes that continuously transfer data without backoff or limits 

These examples are illustrative, not exhaustive.

5. Monitoring and Evaluation

Crate.io monitors aggregate resource usage to ensure system stability and fairness.

Usage is evaluated based on factors such as:

• sustained bandwidth consumption over time 

• impact on multi-tenant system performance

• deviation from typical usage patterns for similar deployments

No single metric is used in isolation; evaluation considers overall behavior and impact.

6. Notification and Remediation

If usage is identified as exceeding fair use:

1.    Notification

Crate.io will notify the Customer with details of the observed pattern and its impact. 

2.    Remediation Period

The Customer will be given a reasonable period to adjust usage, which may include: 

a.    optimizing queries or data transfer patterns 

b.    implementing rate limits or batching 

c.    adjusting architecture 

d.    upgrading to a more appropriate service tier 

Crate.io will work with the Customer in good faith to resolve the issue.

7. Enforcement

If the issue is not resolved within the remediation period, or if immediate action is required to protect system stability, Crate.io may take proportionate measures, including:

• temporary rate limiting of data transfer

• throttling specific workloads

• restricting or suspending affected operations 

Any enforcement will be limited to the minimum scope necessary to restore system stability.

8. Design for High-Throughput Use Cases

Customers with sustained high data transfer requirements are encouraged to:

• engage with Crate.io to design appropriate architectures 

• use dedicated or higher-capacity service tiers where applicable 

This ensures predictable performance without impacting other users.

9. Updates

This Fair Use Policy may be updated from time to time. Updates will not materially reduce the Customer’s ability to use the Services for normal, intended workloads.