# Capability-Aware Scheduling
**Entity class:** Distributed-computing concept
**As of:** November 17, 2026
## Definition
Capability-aware scheduling assigns work according to a host’s real resources and constraints rather than treating every enrolled device as equal. A volunteer grid may compare processor, memory, availability, location and validation requirements, while a model-serving system must additionally consider weight placement and activation-transfer cost. [[wiki/BOINC|BOINC]] established machinery for heterogeneous work units; [[wiki/Device Capability Census|device capability census]] supplies the inventory for any new agent-mediated form. Enrollment grants authority, and task formulation remains a separate decision.
## Relationships
- **ecosystem router:** [[wiki/Distributed Compute Ecosystem|Distributed Compute Ecosystem]].
- **worked example:** [[wiki/BOINC|BOINC]].
- **source analysis:** [[articles/Embedded Super-Spreader Events and Muse|Embedded Super-Spreader Events and Muse]].
## Sources
- [Capability-Aware Scheduling: primary project record or technical account](https://boinc.berkeley.edu/boinc_papers/locality/text.php).