A constraint is a requirement or limitation that narrows which solutions are allowed. Algorithm design balances time, memory, cost, scale, and correctness rather than optimizing everything at once. section 1.2.1
Constraints shape practical decisions in several ways:
- Matching systems use preferences and operating limits to reduce unstable driver–rider assignments, cancellations, and wait times. section 2.2.2
- Resource-allocation problems enforce hard ceilings, such as available memory or vehicle payload, while preserving as much value as possible. section 7.2
- An optimization may improve speed or efficiency while sacrificing another constraint; there is no universally best algorithm. section 1.2.1
An eight-gigabyte deployment limit
A scheduler has 8 GB of burst RAM and four indivisible applications. Selecting applications requiring 1, 2, and 5 GB uses the entire ceiling and delivers priority 11, while deploying the 4 GB and 5 GB applications would exceed capacity and crash the system. section 7.2




























