Scaling and reliability

​

Scaling

BYOC is designed for enterprise-scale workloads. The usage metering service can sustain up to 1,000,000 usage events per second. Scaling is handled by Stigg through the IaC deployment. There is no manual intervention required on your side — when your traffic profile requires more capacity, Stigg updates the IaC configuration and applies it to your environment.

​

Horizontal scaling

All stateless BYOC services scale horizontally. The metering services and sync agent are designed to run as multiple replicas behind a load balancer. Adding capacity does not require restarts or configuration changes to your application.

​

Storage scaling

Persistent storage (databases and event buffers) scales independently from compute. Storage capacity is monitored by Stigg and expanded proactively before saturation is reached.

​

Traffic spikes

Usage events are buffered locally within the metering service before aggregation. This buffer absorbs short-term traffic spikes and protects the downstream database from being overwhelmed during bursts.

​

Reliability

​

High availability deployment

BYOC components are deployed across multiple availability zones within your cloud region. If a single availability zone fails, the remaining zones continue serving traffic without interruption.

​

Failure isolation

A failure in the BYOC environment does not affect your ability to access Stigg Cloud directly. Your SDK can be configured to fall back to Stigg Cloud if the BYOC endpoint becomes unreachable. This mirrors the fallback strategy available with the Sidecar.

​

Durability

Usage events reported to the BYOC metering service are persisted to durable storage in your cloud before acknowledgment. Events are not lost if the processing layer encounters an error — they are retained and retried.

​

SLA

Stigg targets 99.9% uptime for the BYOC usage metering service. The SLA and specific terms are defined in your enterprise agreement. Contact us to discuss SLA requirements.