Scalability

Quickwork is engineered to manage billions of transactions per customer, per account, per month—setting it apart from other iPaaS and integration providers that often track performance by task count or individual API requests. At Quickwork, a transaction is defined as the end-to-end execution of a workflow, regardless of the number of tasks, steps, or APIs involved, or the time required to process each transaction. This transaction-focused approach better aligns with the large-scale, high-complexity use cases Quickwork supports, which differ significantly from traditional task-based ecosystems seen in other iPaaS solutions.

To handle such high transaction volumes while optimizing resource usage, energy efficiency, and transaction isolation, Quickwork leverages a robust set of infrastructure strategies, which are also customizable to meet specific enterprise needs:

• Serverless ecosystem: Quickwork’s architecture is built on a serverless framework that dynamically scales individual workloads and connectors from zero to thousands of containers within milliseconds, leveraging real-time telemetry for instant scaling without lag.
• Auto-scaling infrastructure: Quickwork employs a custom-built scaler compatible with various cloud providers, provisioning instances with pre-warmed settings to seamlessly handle unexpected workload surges. Instances are selected based on workload type and configured to maintain minimum availability, ensuring uninterrupted service continuity.
• Event-driven scalers: Using metric-driven servers with zero latency, Quickwork enables precise horizontal scaling for individual workflows or workflow groups. Customers can further customize these scaling settings based on their workload patterns and requirements.
• Batched data workloads: Quickwork allows customers to define batch sizes for data triggers, supporting payloads ranging from single events to hundreds within a batch, enabling customized performance for different operational needs.
• Provisioned concurrency: With configurable lightweight queues, customers can throttle individual workflows to achieve performance isolation between workloads, ensuring resource efficiency and stable performance across different integration flows.
• Multi-threading: Each transaction runs within its own dedicated container by default, though customers have the option to share containers among transactions for specific workflows, maximizing throughput and efficiency, especially during I/O operations.
• Stateless & stateful workloads: Quickwork provides options for stateless and stateful workload configurations, allowing enterprises to balance performance with reliability based on their use case requirements.
• Customizable limits & Quotas: For high-volume use cases, nearly all workflow execution settings in Quickwork are configurable. Customers can set their own limits and quotas, fine-tuning the platform’s performance to align with specific demands.