Bound fragmentation before adding a more complex allocator
The team proposes size classes after one trace shows poor reuse, but the trace mixes long-lived metadata with short-lived packet records.
- Focused work estimate
- 4h + prerequisites
- Priority in the scenario
- High
- Engineering practice
- Benchmarking · Allocator design · Experimental design
Estimated field mix
- Systems programming60%
- Performance engineering40%
Field percentages are editorial estimates of the ticket's engineering focus. They total 100%; they are not measured time, proficiency scores, or ownership evidence.
Review it, then add it to your workspace.
The board opens an editable draft; nothing is saved until you confirm it. Sign-in and workspace permissions apply, and Demo boards remain ephemeral.
Project context
A fictional telemetry collector copies decoded packet fields into a small native arena. The prototype misaligns wide values and reuses memory while readers still hold views. Create a local Rust crate or C library with generated byte fixtures; no device traffic or production allocator replacement is supplied.
Setup prerequisites
- Pointers and slices
- Integer overflow
- Memory alignment
Preceding work
Complete these dependencies, or supply their agreed outputs before taking this ticket.
Acceptance criteria
- Measure internal and external fragmentation separately
- Replay at least three declared lifetime distributions
- Compare reset-only, free-list, and size-class approaches with correctness held constant
Implementation constraints
- Do not select an allocator from mean throughput alone; retain raw workload seeds and peak memory.
Verification to include
- Reproduce each workload and its fragmentation measurements.
- Change the seed and show the report identifies noncomparable runs.
Deliverables
- Allocator comparison with workload manifest
Rollout and recovery
Adopt additional complexity only if the declared workload crosses an agreed bound.
Value of the work
For the engineer: Practice representation invariants, lifetimes, unsafe-boundary review and memory diagnostics.
For the team: Review a component whose allocation failures and corruption cases are explicit before reuse in latency-sensitive code.
Evidence boundaries
Outcome Evidence: Tests, patches, and runbooks are requested deliverables. They become Outcome Evidence only through a qualified Mission and immutable Evidence IDs.
Ownership Evidence: Independent adaptation must be observed under a declared verification policy and cite immutable Evidence IDs. Completing a planning ticket establishes no Ownership Evidence.