This article provides an overview of Verato’s auto-scaling infrastructure and API throughput. The performance of an API is crucial for delivering seamless user experiences and ensuring operational efficiency. And, achieving optimal API throughput relies heavily on the robustness of the underlying infrastructure to ensure the API's scalability, reliability, and performance.
Verato Auto-Scaling Infrastructure
- Verato’s auto-scaling infrastructure automatically adjusts computing resources based on real-time demand and dynamically scales up resources during peak usage and scales down during low traffic periods to maintain service availability.
- The Verato infrastructure can auto-scale and bring up additional application nodes based on traffic and load. It is highly recommended to increase the load gradually to avoid a drastic change in the transaction rate which may result in potential errors. When a stack experiences a huge increase in traffic, then the existing resources try to keep up with traffic increase and may hit their capacity, thus causing potential errors.
- For example, if the current transaction per seconds (TPS) rate is 10, it is not recommended to increase the TPS rate to 50 instantly. Instead, Verato recommends the use multiple API threads or channels to achieve peak rates with a gradual increase (see below).
If you have a future need or event for a high TPS rate, please engage your Account Team or submit a support case for requirements gathering (additional fees may apply).
Verato’s Standard Throughput (Transactions Per Second)
- Standard TPS: Production environments support a standard, monthly avg Transactions per Second rate of 10 TPS in a highly available (HA) state. Non-production environments support a standard, monthly avg TPS rate of 3 in a non-HA state.
- Bulk data uploads will be processed at the standard TPS rate.
- Verato’s standard TPS rates will gradually increase via auto-scaling which occurs automatically.
- Custom TPS: Non-standard, custom TPS rate are available based on customer requirements.
If there’s a need for a one-time event or prolonged, non-standard TPS rate, please engage your Account Team or submit a support case for requirements gathering (additional fees may apply).
Optimizing Throughput (TPS) Using a Multi-Thread (Multi-Channel) Approach
- Customers can employ various channels (5-20 parallel API threads) to satisfy their optimal peak loads on Verato’s Standard TPS rates.
- When sending high volumes through parallel threads, the best practice is to organize the adds and updates to avoid operating against the same native ID/LINK ID simultaneously across multiple threads.