Telephonic communications – Centralized switching system – Switching controlled in response to called station...
Patent
1997-01-13
1999-10-05
Hong, Harry S.
Telephonic communications
Centralized switching system
Switching controlled in response to called station...
364132, 364187, 379 10, 379 15, 379269, 3951821, 39520078, 707204, H04M 300, G06F 1120, G06F 15177, H04Q 3545
Patent
active
059636343
DESCRIPTION:
BRIEF SUMMARY
This application is the national phase of international application PCT/F195/00399, filed Jul. 11, 1995 which designated the U.S.
FIELD OF THE INVENTION
The invention relates to a method for warming up a spare process in parallel with an active process in a replicated real-time system, in particular in a telephone exchange, the system comprising a control means that comprises at least one active control unit having, permanently or when necessary, a replicating hot-standby spare unit so that the spare unit performs the same processes as the active control unit in parallel.
BACKGROUND OF THE INVENTION
Distributed computer systems have recently been implemented in the control systems of public switched telephone exchanges. The control of a telephone exchange is then distributed to several computers, which are connected with a relatively high-speed bus or equivalent transmission means. In telephone exchanges or other switching systems of this type, an effort is made to support system operations by replicating at least part of the distributed control computers. This should enable replication of the control computers in a way that does not unreasonably consume the performance capacity of the control computers, or require costly special equipment. Requirements for the replication of distributed computers in a telephone exchange environment are disclosed in the article "New fault-tolerance design: developments in software system architecture of the Nokia DX 200", Raimo Kantola, Discovery, Volume 22, First Quarter 1991, pp. 32-39.
The following solutions, for example, have previously been employed in the replication procedures of distributed computer systems. In a solution utilizing microsynchronization of computers, for instance, two computers controlled by special equipment execute precisely the same computer instructions at precisely the same moment. The advantage of the microsynchronization method is its transparency to the application software. Its disadvantage is the costliness of the special equipment and the difficulty, even impossibility, of effectively applying the method to N+1 supported computers, particularly when N is at least several dozen. In the N+1 redundancy mode, N similar computers perform similar, yet independent useful tasks using the same software. One computer is a spare unit to be taken into use if one of said N computers fails, or for example, when the operational arrangements of the switching system so require. The advantage of the N+1 equipment redundancy method is its cost-effectiveness and its compatibility with the 2N method, with the difference that the connection of the spare unit to the active unit must always be carried out prior to unit changeover.
Solutions have previously also been implemented in which the redundant execution of the entire control has been left in charge of the application software, so that the state automaton of the program has comprised the necessary state transitions for keeping the spare unit up-to-date with that which is executed by the active computer. The drawback of this replicating method is that the application must solve two problems at a time: its actual task, and the support therefor. This complicates the development of applications. Another drawback is that this replication method does not result in a uniform execution, as a result of which the maintenance of the applications is costly.
Methods shielding the replications from the applications have also been developed previously, and these have aimed at correctness of computation in all performing computers, and they have therefore been heavy and consumed the performance capacity of the computers. These solutions have given the correctness of the performance a priority over the availability of the system. Therefore they are not particularly suitable for a switching environment, such as a telephone exchange, where high availability is more important than absolute correctness of every discrete minor function.
A replication method is known from Finnish Patent application 912669 in which the pro
REFERENCES:
patent: 4371754 (1983-02-01), De et al.
patent: 4466098 (1984-08-01), Southard
patent: 4633039 (1986-12-01), Holden
patent: 5369759 (1994-11-01), Kanbe
patent: 5469503 (1995-11-01), Butensky et al.
Raimo Kantola; New fault-tolerance design: developments in software system architecture of the Nokia DX 200; Discovery; 1991; pp. 32-39 vol. 22.
Hartikainen Eeva
Kantola Raimo
Hong Harry S.
Nokia Telecommunications Oy
LandOfFree
Method for warming up a spare process in a replicated real-time does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for warming up a spare process in a replicated real-time , we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for warming up a spare process in a replicated real-time will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1180555