D686 Operating Systems for Computer Scientists - Set 2 - Part 3

Correct Answer: Direct hardware access

Type-1 hypervisors run directly on hardware without host OS, providing better performance and isolation than Type-2 running atop host operating systems.
Option a is incorrect because while performance differs, the fundamental distinction is architectural placement relative to hardware and host OS layers structurally.
Option c is incorrect because memory requirements depend on guest workloads; the key difference is whether hypervisor requires host OS or not.
Option d is incorrect because network capabilities can be similar; the defining characteristic is direct hardware control versus host OS mediation fundamentally.

Correct Answer: External fragmentation

External fragmentation creates unusable memory gaps between allocated blocks, preventing large allocations despite sufficient total free memory being available in system.
Option a is incorrect because internal fragmentation wastes space within allocated blocks when request size doesn't match block size exactly provided.
Option c is incorrect because memory leaks involve failing to free allocated memory after use, different from fragmentation of free space availability.
Option d is incorrect because stack overflow occurs from excessive stack usage exceeding limits, unrelated to heap fragmentation issues in memory allocation.

Correct Answer: Heuristic detection

Heuristic detection analyzes behavioral patterns and characteristics to identify potentially malicious activities without requiring predefined signatures for unknown malware variants effectively.
Option a is incorrect because static analysis examines code without execution, not observing runtime behavior patterns for detecting malicious activities dynamically occurring.
Option c is incorrect because signature scanning matches known malware patterns in files, not detecting new threats through behavioral analysis during execution.
Option d is incorrect because checksum verification detects file modifications but doesn't identify malicious behavior patterns during program execution runtime specifically.

Correct Answer: Frequency scaling

Dynamic frequency scaling adjusts CPU speed to match workload requirements, reducing power consumption during light loads while maintaining performance when needed appropriately.
Option a is incorrect because voltage regulation maintains stable power supply but frequency scaling specifically adjusts processing speed for power savings achieved.
Option c is incorrect because core parking disables unused cores entirely rather than adjusting frequency of active cores based on workload dynamically.
Option d is incorrect because thread pooling manages software threads for efficiency, not hardware frequency adjustment for power consumption management directly implemented.

Correct Answer: Redundancy mechanisms

Redundancy mechanisms provide backup components or paths that take over when failures occur, maintaining system availability despite individual component failures happening.
Option a is incorrect because error logging records failures for analysis but doesn't enable continued operation when components fail without redundancy present.
Option c is incorrect because performance monitoring tracks system metrics but doesn't provide failure tolerance without redundant components available for failover switching.
Option d is incorrect because capacity planning ensures adequate resources for load but doesn't address component failure tolerance through redundancy specifically required.

Correct Answer: Extended attributes

Extended attributes store arbitrary metadata including security information, ACLs, and labels beyond basic permissions, enabling richer security policies and functionality.
Option a is incorrect because journal entries record file system changes for recovery, not storing additional security metadata with files permanently attached.
Option c is incorrect because directory indexes optimize file lookups within directories, not storing security metadata or access control information with files.
Option d is incorrect because block pointers map file data locations on disk, not storing security attributes or additional metadata with files.

Correct Answer: Dynamic loading

Loadable modules can be inserted and removed at runtime without recompilation, providing flexibility to add drivers and features as needed dynamically.
Option a is incorrect because modules add slight overhead compared to compiled-in code; the advantage is flexibility not performance improvement over static.
Option c is incorrect because modules complicate debugging with dynamic loading and unloading; monolithic kernels are simpler to debug typically overall.
Option d is incorrect because while unused modules save memory, the infrastructure for module loading adds overhead compared to minimal static kernels.

Correct Answer: Acknowledgment system

Acknowledgment systems confirm packet receipt, triggering retransmission when acknowledgments aren't received within timeout periods, ensuring reliable delivery despite network issues occurring.
Option a is incorrect because flow control manages transmission rates to prevent overwhelming receivers, not ensuring delivery of lost packets through retransmission.
Option c is incorrect because routing tables determine packet paths through networks, not providing reliability through retransmission of lost data packets.
Option d is incorrect because port numbers identify application endpoints for communication, not implementing reliability mechanisms for packet delivery assurance needed.

Correct Answer: Hardware binding

Hardware binding links software licenses to unique hardware identifiers like CPU serial numbers, preventing execution on unauthorized machines after copying occurs.
Option a is incorrect because code obfuscation makes reverse engineering difficult but doesn't prevent copying or execution on different hardware systems specifically.
Option c is incorrect because time stamps track creation or modification dates but don't prevent software from running on different hardware after copying.
Option d is incorrect because digital watermarks identify ownership or origin but don't technically prevent execution on unauthorized hardware after duplication happens.

Correct Answer: High CPU usage

CPU-bound systems show sustained high processor utilization with minimal I/O wait time, indicating computation rather than I/O operations limit performance primarily.
Option a is incorrect because high disk usage indicates I/O-bound behavior where storage operations limit performance rather than CPU processing capability available.
Option c is incorrect because high memory usage alone doesn't determine CPU versus I/O binding; both types can have high memory utilization patterns.
Option d is incorrect because high network usage suggests network I/O bottleneck rather than CPU-bound behavior where processing limits performance primarily observed.