All MCQs

The remainder section contains ordinary work outside the protected critical section.
It does not require the same mutual-exclusion restriction for that shared resource.

The exit section unlocks or signals that the process has left the critical section.
This allows another waiting process or thread to enter safely.

The entry section contains the synchronization steps used to acquire access to the critical section.
Only after successful entry may the process execute the protected code.

A critical section contains operations whose interleaving could corrupt shared state.
Synchronization ensures that incompatible executions do not overlap.

Processor partitioning reserves groups of CPUs for selected applications, users, or service classes.
It can provide isolation, predictability, or performance control.

Shared-cache topology can either help communicating threads or create contention among unrelated workloads.
A topology-aware scheduler considers both effects when placing threads.

Non-uniform memory access systems have faster local memory and slower remote memory.
Placing a thread near its data can reduce memory latency and interconnect traffic.

Hard affinity defines an explicit CPU set on which the thread is permitted to run.
The scheduler must respect that restriction when making placement decisions.

Soft affinity is a scheduling preference rather than an unbreakable restriction.
The scheduler may move the task when load balancing or other needs outweigh locality.

Pull migration lets an underloaded or idle processor obtain work from a busier processor.
The receiving processor initiates the load-balancing action.

Push migration actively transfers work away from an overloaded processor.
The balancing mechanism initiates the movement rather than waiting for an idle CPU to request work.

Load balancing moves or assigns tasks so no processor is overloaded while another remains underused.
The policy must also consider migration cost and cache locality.