Christoph Rohland; Hugh Dickins; KOSAKI Motohiro > 자유게시판

• 목록
• 답변
• 글쓰기
• 게시판 리스트 옵션
  • 수정
  • 삭제

In pc science, shared memory is memory that may be concurrently accessed by a number of packages with an intent to provide communication among them or keep away from redundant copies. Shared memory is an environment friendly technique of passing information between packages. Depending on context, programs might run on a single processor or on a number of separate processors. Using memory for communication inside a single program, e.g. amongst its multiple threads, can be referred to as shared memory. In laptop hardware, shared memory refers to a (sometimes massive) block of random access memory (RAM) that may be accessed by several completely different central processing units (CPUs) in a multiprocessor laptop system. A shared memory system is relatively straightforward to program since all processors share a single view of information and the communication between processors could be as fast as memory accesses to the identical location. Attempting to access close by memory areas may trigger false sharing. Shared memory computer systems can not scale very well.

Such cache coherence protocols can, once they work nicely, present extremely excessive-performance entry to shared information between multiple processors. Then again, they'll typically turn into overloaded and change into a bottleneck to performance. Applied sciences like crossbar switches, Omega networks, HyperTransport or front-facet bus can be used to dampen the bottleneck-effects. In case of a Heterogeneous System Structure (processor architecture that integrates several types of processors, reminiscent of CPUs and GPUs, with shared memory), the memory management unit (MMU) of the CPU and the input-output memory administration unit (IOMMU) of the GPU have to share certain traits, like a standard tackle area. The alternate options to shared memory are distributed memory and distributed shared memory, every having an analogous set of points. CPUs and the underlying architecture is not cache coherent. IPC by shared memory is used for example to transfer pictures between the application and the X server on Unix techniques, or inside the IStream object returned by CoMarshalInterThreadInterfaceInStream in the COM libraries below Windows.

Dynamic libraries are generally held in memory as soon as and mapped to a number of processes, and solely pages that had to be customized for the person process (because an emblem resolved differently there) are duplicated, usually with a mechanism known as copy-on-write that transparently copies the web page when a write is attempted, and then lets the write succeed on the non-public copy. In comparison with multiple tackle space working methods, memory sharing -- particularly of sharing procedures or pointer-based buildings -- is less complicated in single address space working methods. POSIX offers a standardized API for using shared memory, POSIX Shared Memory. POSIX interprocess communication (a part of the POSIX:XSI Extension) contains the shared-memory functions shmat, shmctl, shmdt and shmget. Unix System V offers an API for shared memory as properly. This uses shmget from sys/shm.h. BSD techniques provide "anonymous mapped memory" which may be utilized by several processes. It stays within the system until explicitly eliminated by a course of.

This has a disadvantage in that if the method crashes and fails to scrub up shared memory it can keep until system shutdown; that limitation isn't present in an Android-specific implementation dubbed ashmem. POSIX also provides the mmap API for mapping information into memory; a mapping can be shared, allowing the file's contents to be used as shared memory. Linux distributions based on the 2.6 kernel and later provide /dev/shm as shared memory in the form of a RAM disk, more specifically as a world-writable directory (a directory wherein every person of the system can create information) that's saved in memory. Both the RedHat and Debian based mostly distributions embrace it by default. Support for the sort of RAM disk is completely optionally available within the kernel configuration file. On Home windows, one can use CreateFileMapping and MapViewOfFile functions to map a area of a file into Memory Wave clarity support in multiple processes. Qt provides the QSharedMemory class. Other programming languages could have their very own ways of utilizing these working facilities for similar impact. For example, PHP gives an API to create shared memory, just like POSIX capabilities. El-Rewini, Hesham; Abd-El-Barr, Mostafa (2005). Advanced Computer Structure and Parallel Processing. Jeffrey S. Chase; Henry M. Levy; Michael J. Feeley; and Edward D. Lazowska. Robbins, Kay A.; Robbins, Steven (2003). Unix systems programming: communication, concurrency, and threads (2 ed.). Prentice Hall PTR. p. Shared memory facility from the one Unix Specification. Christoph Rohland; Hugh Dickins; KOSAKI Motohiro. Creating Named Shared Memory from MSDN. Shared Memory Introduction, Ch. 12 from e-book by Richard Stevens "UNIX Community Programming, Volume 2, Second Edition: Interprocess Communications". SharedHashFile, Memory Wave clarity support An open supply, shared memory hash desk.