JPS5936765B2 - power control system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a power supply control method for each device constituting a data processing device such as a computer system, and in particular, to dynamically reconfigure the system configuration depending on the load status of the data processing device.
This invention relates to a power supply control system that saves power by cutting off unnecessary devices while optimizing the configuration.

Conventionally, data processing equipment has been powered on and off at the same time, except in the case of failure or repair of the equipment that makes up the system, and each device can be used while the power is on. However, electricity was consumed even when the equipment was not actually in use. In addition, in order to solve such problems, depending on the processor processed by the data processing device, it is possible to turn on/off the power of input/output devices with commands so that power is supplied only to the input/output devices necessary for the program. It has also been proposed to turn it off (Japanese Patent Laid-Open No. 50-4928).

However, it only turns on and off the input/output device.
Power on/off of the main system device is not considered at all.
In large computer systems, processing capacity is increased by connecting multiple central processing units etc. in consideration of the maximum load, but when the load is low, it may be necessary to reduce the number of operating units.
This was not possible before. SUMMARY OF THE INVENTION An object of the present invention is to provide a power control system that dynamically reconfigures the system configuration of a data processing device depending on its load condition and enables minimum power consumption.

The present invention is effective for data processing apparatuses (hereinafter referred to as computer systems) having the following operational configurations.

For example, a computer that runs a computer system during normal weekday work hours, starts job input in the morning, finishes in the evening, processes the remaining jobs, and then turns off the power when finished. Since the system does not need an input device after the job input is completed, at this point the program disconnects the input device from the system configuration and turns off the power. In addition, as time passes, the number of connected central processing units and channel devices is dynamically increased or decreased depending on the load level, and when the load decreases near the end of a job, the number of central processing units is reduced, and all jobs are is processed, resulting in a power control system that can power down all devices in the system.

The present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block connection diagram showing one embodiment of the present invention. The computer system 1 in this example is a central processing unit CP.
l, CP2 storage device Ml, M2 channel device CHl, C
The main body system consists of each device consisting of H2, an input/output device group/O, a configuration control device CC, a service processor SP, and a power supply control device PC.

The above main body system devices are connected via the configuration controller CC by signal lines Tl, tl' consisting of control lines and data lines, and the input/output device group 1/0 is connected to the channel devices CHl, CH2 and the signal line T2. It is connected by twisting.

The configuration control device CC has a configuration control register,
This register manages the main body system devices logically connected to the computer system 1. The power supply control device PC powers on and off various devices through the control signal line T3, and commands to power on and off the central processing units CPl and CP2 are sent from the configuration control device CC through the signal line T4 and further to the service processor. SPl signal line T5
The information is transmitted to the power supply control device PC through the power supply controller PC. Here, the load status of the computer system 1 may be constantly monitored by software, or by hardware such as a performance analyzer that knows the actual operating rate under the power-on status of each device. You can monitor it. Now, within the computer system 1, the central processing units CPl and CP
When the load on CPU 2 decreases and it is detected, the central processing unit, for example CP2, to be disconnected is determined, the jobs in that central processing unit CP2 are transferred to another central processing unit CPl, and the CPU should be disconnected. In the central processing unit CP2, when there are no more jobs to process, the configuration controller CC
by changing the configuration control registers in its central processing unit CP.
2 from the system configuration without affecting other devices. Thereafter, the configuration control device CC issues an instruction to power off the disconnected central processing unit CP2 to the power supply control device PC through the service processor SP, and upon receiving the command, the power supply control device PC Power off the device CP2. central processing unit CP
When the load on CP2 increases and the central processing unit CP2 is newly incorporated into the system configuration, the operation opposite to the above is performed.

Also for channel devices CHl and CH2, storage device Ml
, M2 as well.

In addition, the input/output device group 1/O does not have a configuration control register, and the device specified on the table in which the input/output device group/O in the operating system is registered is specified via channel devices CHl and CH2. This is achieved by turning the power supply control device PC off-line and then performing the opposite operation to the power supply control device PC when the power is turned on. Next, the above operation will be explained in detail.

FIG. 2 is a circuit connection diagram showing an outline of the configuration control device.

The same parts as in FIG. 1 are given the same reference numerals.

The configuration control device CC is connected to each main body system device making up the computer system 1 through signal lines Tl, tl', and the output signals of each device are received by the storage section 14 and related through the gate section 16. It is being sent to another device. Although not shown, the outputs of the gate section 16 are connected to each other within the control device CC according to the system configuration, and are distributed to each device again through the signal lines 11 and tV. The gate section 16 can logically connect or disconnect each device to the computer system 1 according to information set in the configuration control register CFR by the service processor SP and the signal line T4. That is, for example, by setting the flip-flop F2 indicating the control bit of the configuration control register CFR to a logical value "1", the output signal of the central processing unit CP1 is set to the flip-flop F2 of the storage section 14.
The data can be transferred to other devices through the flop F1 and the AND gate A1 of the gate section 16. Here, when connecting each device logically, after turning on the power of the connected device, the control bit of the corresponding configuration control register CFR is set to a logical value of 61'. In the case of disconnection, the corresponding control bit is set to a logical value of 10'', and the power is then turned off, thereby protecting other devices from the effects of power on and off.

Incidentally, the turning on/off of the power supply and the setting of the configuration control register CFR are performed by an interrupt caused by a diagnostic command issued by the central processing units CP1 and CP2 to the service processor SP.

Next, we will explain the details of each part that configures the system configuration according to the load situation and performs power on/off control. FIG. 3 shows a service processor and peripheral block connections illustrating one embodiment of the present invention. The same parts as in FIG. 1 are given the same reference numerals. The central processing unit CPl includes an instruction register 2 and a decoder 2a that decodes the contents of the instruction register 2.
When the diagnostic command to instruct the service processor SP to execute is decoded from the output of the decoder 2a and the command register 2, the details of the diagnostic command are set in the interrupt information register 5 by the interrupt information creation circuit 3. The interrupt signal latch flip-flop 4 is set to interrupt the service processor SP. The service processor SP learns details of diagnostic instructions through an interrupt information register 7, an interrupt latch/flip-flop 6, and an internally built-in interrupt analysis routine-in 8.
Here, the diagnostic instruction conveyed by the interrupt information is 6CFR, which instructs data setting to the configuration control register CFR.
If the setting command is 1, the CFR setting routine 9 and the central processing unit start/stop set routine 10 are started and the central processing unit CPl, CFR is activated through the interface circuit 11.
The main system devices such as P2 are stopped, data specified by the command is set in the CFR, and after the system is reset, the central processing units CP1 and CP2 are started. If the diagnostic command is a "power control command", the power control routine 12 is started and an on/off control signal and an address signal of the device to be controlled are sent to the power control device PC through the interface circuit 11a and signal line T5.

The power supply control device PC connects the specified device power supply to the power supply control line T.
3 and transmits a termination signal to the service processor SP through signal line T5. Next, when the load status of the central processing units CPl and CP2 is monitored by the software 1 and the results are known to the central processing unit CPl by the operating system, the central processing unit CP2 connects to the system configuration or I will explain the case where it is cut out.

FIG. 4a shows a case where an increase in load is detected and the central processing unit CP2 is connected from the operation of only the central processing unit CP1.

First, in step 20, the operating system generates a "power control command" as a diagnostic command in the central processing unit CP1, and instructs the central processing unit CP2 to turn on the power.

In steps 21 to 24, the service processor SP stops the central processing unit CPl, and then stops the central processing unit CPI.
Turn on the power of P2 and start up the central processing unit CPl again. In step 25, the central processing unit CP1 generates a diagnostic command ``CFR setting command 1'' and instructs the service processor SP to connect the central processing unit CP2. In steps 26 to 29, the service processor SP connects the central processing unit CP2 to the central processing unit CP1.
Start. After that, the central processing unit CPl performs step 30.
In this step, the central processing unit CP2 is activated. FIG. 4b shows the case where a reduction in load is detected and the central processing unit CP2 is disconnected. First, in step 31, the operating system generates a diagnostic command 6CFR setting command in the central processing unit CPl, and in steps 32 to 34, the service processor SP generates a diagnostic command 6CFR setting command in the central processing unit CPl.
2 is cut off, and in step 35 the central processing unit CP
1, a diagnostic command is generated to instruct the other central processing unit CP2 to turn off its power. In steps 36 to 38, the service processor SP turns off the power to the central processing unit CP2 and starts up the central processing unit CPl. As explained in the above embodiment, the system configuration is dynamically reconfigured according to the system load situation, unnecessary devices are disconnected, and power is cut off, making it possible to save power.

In addition, although this example describes a case where the system automatically monitors the system load status, a system program that understands the operation plan is provided, and the operator can determine the day's operation schedule using a computer. It is also possible to input data into the system and automatically reconfigure the system depending on the time. Furthermore, after turning on the power of the communication line equipment connected to the outside and knowing that an information signal has been sent from the outside, the power of the system is turned on one after another and processing of the transmitted information signal. It is also possible to consider a system in which the communication line equipment is started, and when the processing is completed, the power is turned off and only the original communication line equipment is powered on. As described above, the present invention provides a power control system that saves power in a data processing device.

[Brief explanation of the drawing]

Fig. 1 is a block connection diagram showing one embodiment of the present invention;
Fig. 3 is a circuit connection diagram showing an outline of the constituent climate control device, Fig. 3 is a connection diagram of a service processor and peripheral parts showing an embodiment of the present invention, and Fig. 4 shows a system according to the load status of the central processing unit. FIG. 3 is an operation flow diagram when configuring and connecting or disconnecting. 1...Computer system, CPl, CP
2...Central processing unit, Ml, M2...
Storage device, CHl, CH2...channel device,
I/0...Input/output device group, CC...Configuration control device, CFR...Configuration control register, S
P...Service processor, PC...
・Power control device.

Claims (1)

[Claims] 1. A main body system device including one or more central processing units, channel devices, and storage devices, a configuration control device that controls logical connections between these main body system devices, and a group of input/output devices connected to the channel device. , comprising at least a service processor connected to the central processing unit, and a power control unit connected to the service processor to control power on/off to each of the devices, wherein the central processing unit is configured to control the service processor. A first instruction that generates an interrupt and instructs logical connection and disconnection of an arbitrary main body system device, and a second instruction that also generates an interrupt to the service processor and instructs the power on and off of an arbitrary main body system device. When the service processor receives an interrupt from the first command, the service processor instructs the configuration control device to connect and disconnect the specified main body system device, and also has the second command.
1. A power supply control system comprising: an interrupt processing means for controlling the power supply control device to instruct the power supply control device to turn on and off the power of a specified main body system device when receiving an interrupt by a command.