Inter-process communication
From Wikipedia, the free encyclopedia
Inter-Process Communication (IPC) is a set of techniques for the exchange of data between two or more processes. Processes may be running on one or more computers connected by a network. IPC techniques are divided into methods for message passing, synchronization, shared memory, and remote procedure calls (RPC). The method of IPC used may vary based on the bandwidth and latency of communication between the processes, and the type of data being communicated.
IPC may also be referred to as inter-application communication.
Contents |
[edit] Implementations
There are a number of APIs which may be used for IPC. A number of platform independent APIs include the following:
- Anonymous pipes and named pipes
- Common Object Request Broker Architecture (CORBA)
- Distributed Computing Environment (DCE)
- Message Bus (MBUS) (specified in RFC 3259)
- ONC RPC
- Sockets
- XML XML-RPC or SOAP
- ZeroC's Internet Communications Engine (ICE)
The following are platform specific APIs:
- Apple Computer's Apple events (previously known as Interapplication Communications (IAC)).
- Freedesktop.org's D-Bus
- KDE's Desktop Communications Protocol (DCOP)
- Libt2n for C++ only, handles complex objects and exceptions
- The Mach kernel's Mach Ports
- Microsoft's ActiveX, Component Object Model (COM), Distributed Component Object Model (DCOM), Dynamic Data Exchange (DDE), Object Linking and Embedding (OLE), anonymous pipes, named pipes, Local Procedure Call
- Novell's SPX
- POSIX mmap, message queues, semaphores, and shared memory
- RISC OS's messages
- Solaris's Doors
- System V's message queues, semaphores, and shared memory
Table of IPC Methods:
| Method | Provided by (Operating systems or other environments) |
|---|---|
| File | All operating systems. |
| Signal | Most operating systems; some systems, such as Windows, only implement signals in the C run-time library and do not actually provide support for their use as an IPC technique. |
| Socket | Most operating systems. |
| Pipe | All POSIX systems. |
| Named pipe | All POSIX systems. |
| Semaphore | All POSIX systems. |
| Shared memory | All POSIX systems. |
| Message passing (shared nothing) |
Used in MPI paradigm, Java RMI, CORBA and others. |
| Memory map | All POSIX systems; may carry race condition risk if a temporary file is used. Windows also supports this technique but the APIs used are platform specific. |
| Message queue | Most operating systems. |
| Mailbox | Some operating systems. |
[edit] See also
- Computer network programming
- Communicating Sequential Processes (CSP paradigm)
- .NET Remoting
- Microkernel
- Nanokernel
[edit] References
- J. Liedtke, K. Elphinstone, S. Schiinberg, H. Hartig, G. Heiser, N. Islam, T Jaeger (1997). "Achieved IPC Performance". Proceedings of the 6th Workshop on Hot Topics in Operating Systems (HotOS-VI): 28. ISBN 0-8186-7834-8. .
- Jochen Liedtke. Improving IPC by kernel design, ACM Press 1994, ISBN 0-89791-632-8
- Stevens, Richard. UNIX Network Programming, Volume 2, Second Edition: Interprocess Communications. Prentice Hall, 1999. ISBN 0-13-081081-9
- Nenad Marovac. "On interprocess interaction in distributed architectures", ACM SIGARCH Computer Architecture News, 11(4), 1983
[edit] External links
 |
This computer-related article is a stub. You can help Wikipedia by expanding it. |
