Skip to main content
This article covers the most common ways to optimise your Alumio environment and reduce load to guarantee the stability of your integrations.

Reducing the level of detail in log files

Alumio logs all middleware-related actions by default, such as filtered entities, created tasks, and process information. These cannot be disabled. On top of this, you can enable additional logging for the systems you integrate: authentication logs, API logs, database logs, Webhook logs, and more. While these logs are useful for debugging, they can cause significant load when volumes are high, for example, when processing 100k products with an individual API call per product, or when log files are very large. Three logging levels are available: It is recommended to use Simple or Long logging in production environments and reserve Full logging for the testing phase only.

Reducing the number of tasks exported per execution

Most Alumio integrations are asynchronous –– data is synchronised in two steps: filling the queue and processing it. Real-time integrations combine both steps into one, which can cause high load when processing large numbers of complex tasks. Integrations that require additional API calls per task, or that handle large data objects, are particularly sensitive to this. To reduce load, consider:
  • Reducing the number of tasks exported per Scheduler execution via the Route Scheduler settings.
  • Reducing the number of entities sent per Webhook call.
  • Introducing batch processing to spread the workload.

Reducing the number of scheduled jobs run in parallel

Running many scheduled jobs at the same time can cause spikes in environment load. It is recommended to spread Scheduler jobs across the hour rather than triggering them all at the same time.

Upgrading your environment

If the optimisations above do not bring load within acceptable levels, consider upgrading your Alumio environment. Contact the Support team to discuss available options.