Electrical computers and digital processing systems: support – Digital data processing system initialization or configuration
Reexamination Certificate
2000-03-15
2003-11-25
Lee, Thomas (Department: 2185)
Electrical computers and digital processing systems: support
Digital data processing system initialization or configuration
C713S002000
Reexamination Certificate
active
06654880
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a restart processing method and apparatus for a duplexed system of an electronic exchange or the like.
In a duplexed system, when the system goes down due to the occurrence of a fault or for some other reason, restart processing is initiated by a hardware reset, and the contents of memory are dumped into another file memory for fault diagnosis, etc. However, since dumping takes time, the system cannot be restarted without delay, and an improvement on this point is needed.
2. Description of the Related Art
FIG. 1
is an example of a duplexed system to which. the present invention is to be applied. Specific applications of such a duplexed system include information processing systems such as telephone exchange systems of which high reliability is demanded. In
FIG. 1
, MM is a main memory, CC is a central controller including a processor, CHC is a channel controller, and FM is an external storage device (file memory) such as a large-capacity magnetic disk device or a magnetic tape device. The numbers #
0
and #
1
designate channel
0
and channel
1
, respectively, in the duplexed system.
Assuming that this prior. art duplexed system is a telephone exchange system, generally either one of the CCs, CC#
0
or CC#
1
, is operated in ACT (active—working) state and the other is placed in SBY (standby) state. Supposing that CC#
0
is operating in ACT state, if MM#
0
and MM#
1
are both in INS state (In Service state—operating online by being switched into the system), CC#
0
writes the same contents to both MM#
0
and MM#
1
and reads the contents from either one of them. If one MM is in OUS state (Out of Service state—switched out of the system), CC#
0
cannot access this MM. Further, CC#
0
is connected to CHC#
0
and CHC#
1
, both in INS state, and accesses either associated FM#
0
or FM#
1
, or both.
The hardware configuration of the duplexed system described above is designed to minimize service interruption time in the event of a system down.
FIG. 2
shows the restart processing flow provided as software according to the prior art.
FIG. 3
shows a time chart. from the time a system down occurs to the time the service is restored.
The restart processing according to the prior art will be described. In the system configuration of
FIG. 1
, when a system failure is detected during online processing with CC#
0
operating in the ACT state, this is regarded as being a system down and the restart processing is initiated by a hardware reset. In the restart processing, before initializing the contents of the main memory, the contents of MM being used when the system down occurred are copied to FM (this processing is called freeze dumping) for the purpose of collecting information to determine the cause of the system down (S
1
in FIG.
2
). This processing is performed for the period from time t
1
to time t
2
in FIG.
3
. Thereafter, initialization is performed on the MM in order to restore the online service (S
2
in FIG.
2
), an online service program is loaded (S
3
), and the network (exchange switch) is initialized (S
4
). Then, in the exchange switch, resource rescue processing is performed to rescue (restart) subscribers who were in the middle of voice communication when the system went down and subscribers and devices that were in the middle of call initiation procedures (S
5
in FIG.
2
). When the state that existed before the system went down has been restored with the above series of processes, the operations are transferred to the online processing, and the service is thus restored (S
6
in FIG.
2
). In
FIG. 3
, the service is restored at time t
3
, which means that the service was interrupted from time t
1
to time t
3
due to the restart processing.
In the prior art restart processing described above, during the freeze dump processing (S
1
in
FIG. 2
) the central controller (CC) transfers data from the main memory (MM) to the slower external storage device (FM) via the channel controller (CHC) by repeating data transfer and acknowledgement waiting operations, spending much of the processing time waiting for the external storage device (FM) to complete an operation. This greatly reduces CPU (Central Processing Unit) utilization in the central controller (CC).
Further, as the software size becomes larger, the capacity of the main memory (MM) increases, and the amount of data to be stored on the external storage device (FM) increases proportionally, causing the processing time to increase further.
Thus, the prior art has had the problem that the freeze dump entails an inherent time loss and increases the service interruption time in the event of a system down.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a restart processing method for a duplexed system that can shorten service interruption time in the event of a system down.
According to the present invention, there is provided a method of restarting a system that includes a central controller, two main memories, one for working and the other for standby, connected to the central controller, and an external storage device connected to the central controller, comprising the steps of:
inhibiting switching of the standby main memory into the system in response to the detection of a system down;
transferring at least part of the contents of the standby main memory to the external storage device after operation of the central controller and the working main memory is resumed from an initial state after having inhibited the switching of the standby main memory; and
releasing the standby main memory from the switching inhibited state after completion of the transfer to the external storage device.
According to the present invention, there is also provided a restart processing apparatus for a system that includes a central controller, two main memories, one for working and the other for standby, connected to the central controller, and an external storage device connected to the central controller, comprising: a preserve processing section for inhibiting switching of the standby main memory into the system in response to the detection of a system down; and a freeze dump processing section for transferring at least part of the contents of the standby main memory to the external storage device after operation of the central controller and the working main memory is resumed from an initial state after having inhibited the switching of the standby main memory, and for releasing thereafter the standby main memory from the switching inhibited state.
REFERENCES:
patent: 5163148 (1992-11-01), Walls
patent: 5506773 (1996-04-01), Takaba et al.
patent: 5964813 (1999-10-01), Ishii et al.
patent: 6145095 (2000-11-01), Tattari
patent: 6389554 (2002-05-01), Jung et al.
Fujitsu Limited
Katten Muchin Zavis & Rosenman
Lee Thomas
Yanchus, III Paul
LandOfFree
Method and apparatus for reducing system down time by... 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 and apparatus for reducing system down time by..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for reducing system down time by... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3172767