Editing Operating system (section)

===Concurrency===
{{see also|Computer multitasking|Process management (computing)}}

[[Concurrency (computer science)|Concurrency]] refers to the operating system's ability to carry out multiple tasks simultaneously.{{sfn|Anderson|Dahlin|2014|p=129}} Virtually all modern operating systems support concurrency.{{sfn|Silberschatz et al.|2018|p=159}}

[[Thread (computing)|Thread]]s enable splitting a process' work into multiple parts that can run simultaneously.{{sfn|Anderson|Dahlin|2014|p=130}} The number of threads is not limited by the number of processors available. If there are more threads than processors, the operating system [[kernel (operating system)|kernel]] schedules, suspends, and resumes threads, controlling when each thread runs and how much CPU time it receives.{{sfn|Anderson|Dahlin|2014|p=131}}  During a [[context switch]] a running thread is suspended, its state is saved into the [[thread control block]] and stack, and the state of the new thread is loaded in.{{sfn|Anderson|Dahlin|2014|pp=157, 159}} Historically, on many systems a thread could run until it relinquished control ([[cooperative multitasking]]). Because this model can allow a single thread to monopolize the processor, most operating systems now can [[interrupt]] a thread ([[preemptive multitasking]]).{{sfn|Anderson|Dahlin|2014|p=139}}

Threads have their own thread ID, [[program counter]] (PC), a [[Processor register|register]] set, and a [[Stack-based memory allocation|stack]], but share code, [[heap memory|heap]] data, and other resources with other threads of the same process.{{sfn|Silberschatz et al.|2018|p=160}}{{sfn|Anderson|Dahlin|2014|p=183}} Thus, there is less overhead to create a thread than a new process.{{sfn|Silberschatz et al.|2018|p=162}} On single-CPU systems, concurrency is switching between processes. Many computers have multiple CPUs.{{sfn|Silberschatz et al.|2018|pp=162–163}} [[Parallel computing|Parallelism]] with multiple threads running on different CPUs can speed up a program, depending on how much of it can be executed concurrently.{{sfn|Silberschatz et al.|2018|p=164}}