JP2013025343A - Network apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that: in a conventional network apparatus having a plurality of power supply units, it is difficult to maintain high-efficiency operation and high availability of the network apparatus resulting from redundant operation of the power supply units once the number of power supply units to be run is determined, because power consumption increases/decreases according to the amount of information to be processed.SOLUTION: Provided is a network apparatus which determines the number of power supply units to be run on the basis of power consumption of the network apparatus that constantly varies according to an operational state so that the power supply units can be used with high efficiency, and determines the number of power supply units to be run on the basis of a required redundant operation method.

Description

  The present invention relates to a network device having a plurality of power supply units and a unit for controlling the power supply units.

  In recent years, from the viewpoint of global warming countermeasures, it is an important duty in the telecommunications industry to tackle CO2 emission reduction by energy saving and the like. In addition, high efficiency in production, consumption, and industrial activities, alleviation of traffic congestion in the transportation field, and response in the ICT field to alternative transportation can greatly contribute to CO2 emission reduction. However, on the other hand, services that can be realized by these ICT devices tend to increase CO2 emissions by increasing energy and electric power as the scale and communication volume increase. In order to increase the amount of CO2 emission, network equipment manufacturers are required to provide an apparatus that saves energy.

  As one of the power saving means, a technique disclosed in Patent Document 1 is known. Generally, the efficiency of the power supply unit (efficiency η = output power (W) / input power (W)) is low at light load, has a peak where the efficiency is highest at a certain output power, and the efficiency at higher output power There is a tendency to go down. Patent Document 1 increases the output power per operating power supply unit by operating N power supply units when the power consumption of the device is low, and operating M power supply units when the power consumption is high. The unit is designed to prevent light load operation.

JP 2010-15226 A

  There are dozens of types of NIFs (network interfaces) in network devices, and the power consumption per NIF varies greatly depending on the operating status of a plurality of ports in each NIF and the amount of data handled. Also, the cooling unit reduces the fan speed when the ambient temperature is low (low power consumption state) and increases the rotational speed when the ambient temperature is high (high power consumption state) so that the power consumption depends on the ambient temperature. Increase or decrease.

  Thus, even if there is no change in the device configuration, the power consumption greatly increases and decreases. However, the prior art described above is based on the power consumption of the device calculated by multiplying the numerical value that is predicted or measured in advance for the power consumption of the arithmetic unit and the number of operations of the arithmetic unit, that is, the device configuration If there is no change in power consumption, it is assumed that there is no change in power consumption. Therefore, even if there is no change in the device configuration as in a network device, it cannot cope with a large fluctuation in power consumption.

  Furthermore, in recent years, network devices have become important public infrastructures, and in order to increase their availability, it is common to use a power supply configuration of N + 1 redundant operation or N + N redundant operation system, but this is not considered in the prior art. When the number of power supply units is changed (reduced), the redundancy operation method may change from N + N redundancy to N + 1 redundancy or the redundancy operation may become non-redundant operation, which may impair availability. is there.

In order to solve at least a part of the above problems, an apparatus having the following configuration is proposed.
[Application Example 1]
A plurality of cooling units, a plurality of interface units, a control unit for controlling the entire apparatus, and a plurality of power supply units, and each of the cooling unit, the interface unit, and the control unit has a power monitor circuit. The control unit determines the number of operations of the power supply unit based on the power consumption collected from the power monitor circuit and the power supply efficiency of the power supply unit measured in advance. To do. By adopting this configuration, the number of operation of the power supply unit can be appropriately determined according to the actual power consumption collected from the power monitor circuit, and it becomes easy to cope with changes in the device configuration.

[Application Example 2]
In the communication device according to the application example 1, in the case where the control unit determines the number of operations of the power supply unit, a predetermined number set in advance to the number of the power supply units necessary to maximize the power supply efficiency. Proposed is a communication device characterized by a redundant configuration. By adopting this configuration, it is possible to ensure reliability while improving power supply efficiency.

[Application Example 3]
The communication apparatus according to Application Example 2, wherein the control unit obtains the operable number of the power supply units from a difference between the number of mounted units and the number of failed units, and determines the number of operations of the power supply unit with the operable number as an upper limit. A communication apparatus characterized by the above is proposed. By adopting this configuration, it becomes possible to select the optimum number of operations from among the power supply units that can actually be operated, and it becomes easy to maintain high power supply efficiency even when the power supply unit fails.

[Application Example 4]
The communication apparatus according to Application Example 2, wherein the control unit adopts a method of selecting N + 1 units as a redundancy method, where N is the number necessary for maximizing power supply efficiency. Communication device.

[Application Example 5]
The communication apparatus according to Application Example 2, wherein the control unit adopts a method of selecting N + N units as a redundant method, where N is the number necessary for maximizing power supply efficiency. Communication device.

[Application Example 6]
In the communication device according to Application Example 3, in a case where the control unit can take a plurality of redundant configurations and non-redundant configurations, the control unit is configured to be as close as possible to the configuration set in advance from the operable number. A communication apparatus that determines the number of operations of the power supply unit.
By adopting the configuration of application example 4-6, it becomes easier to increase redundancy.

  By adopting the configuration of the present invention, it becomes easy to realize a communication device that is flexible and has high power supply efficiency.

It is a front view of a network device. It is a rear view of a network device. It is an internal block diagram of a network device. It is a figure which shows the characteristic of the efficiency curve of a power supply unit. It is a figure explaining the non-redundant operation system of a power supply unit. It is a figure explaining the N + 1 redundant operation system of a power supply unit. It is a figure explaining the N + N redundant operation system of a power supply unit. It is a basic flowchart which shows operation | movement of a network apparatus. It is a sub-flowchart at the time of N + N redundant operation. It is an N + 1 redundant operation sub-flowchart. It is a figure which shows the redundancy system selection table of a power supply unit. It is a figure explaining the effect by energy saving operation.

  In this embodiment, a network device 10 such as a switch as shown in FIGS. FIG. 1 is a front view of the network device 10, and FIG. 2 is a rear view of the network relay device 10. The network apparatus 10 includes a power supply unit for one system 20, a power supply unit for two systems, a network interface (hereinafter referred to as NIF) 30, a cooling unit 40, a basic unit CSU (Control Switching Unit) 50, and a redundant unit CSU51. Composed. The CSU has a redundant configuration, and the basic unit CSU 50 controls the entire apparatus such as the power supply units 20 and 21, the NIF 30, and the cooling unit 40 during normal operation. The CSU is also simply referred to as a control unit.

  Next, the internal configuration of the network device 10 will be described with reference to FIG. In FIG. 3, only the basic unit CSU 50 is shown, and the redundant unit CSU is omitted. A power monitor circuit 61 (or current and voltage monitor circuit) is provided at the input portion of the power supply bus line 60 of the NIF 30, the cooling unit 40, and the CSU 50, which are units that receive power from the power supply units 20 and 21. The power monitor signal 62, the power supply unit ON / OFF control signal 63, and the mounting / failure information signal 64 indicate input / output of each signal.

  The power supply units 20 and 21 feed power from the four power supply units 20 and 21 to the two cooling units 40, the two CSUs 50 and 51, and the four NIFs 30. Usually, the voltage of the power supply bus line 60 is generally −48V, + 48V, + 12V, or the like. A power monitor circuit (or current / voltage monitor circuit) is provided at the input portion of the power supply bus line 60 of the load side unit NIF 30, cooling unit 40, and CSU 50, and the power consumption of the load side unit is transmitted to the CSU 50 through the power monitor signal 62. Tally. The same effect can be obtained even if the output power of the power supply units 20 and 21 is totaled.

  Further, the CSU 50 constantly monitors the mounting / failure information signal 64 to keep track of the number of power supply units 20 and 21 that can be operated. The CSU 50 can individually turn on / off the power supply units 20 and 21 by a power supply unit ON / OFF control signal 63.

  The CSU 50 determines the optimum number of power supply units to be operated according to the flowchart of FIG. 8 to be described later, based on the aggregated power consumption of the load side unit and the number of operable power supply units.

  The power consumption of the load side unit is aggregated by the CSU 50 through the power monitor signal 62, so that the changing power consumption can be grasped at any time, and the number of power supply units to be operated is determined based on the aggregated power consumption. Therefore, even when there is no change in the device configuration, it is possible to change the number of power supply units to be operated in response to a change in power consumption. In addition, the number of power supply units to be operated is controlled in units of one unit so that the power supply unit has the highest load factor based on the power consumption calculated from the power monitor. Finer power-saving operation is possible than the control method with only M units.

  FIG. 4 shows the characteristics of the efficiency curve of the power supply unit. In order to improve the efficiency of the power supply unit and optimize the redundant operation method, grasp the characteristics of the efficiency curve of the power supply unit in advance. The efficiency curves of the power supply units 20 and 21 to be installed are acquired by power supply unit evaluation. Since the efficiency curves are almost the same for the same type of power supply unit, in most cases, if the representative power supply unit is evaluated as a single unit, it can be considered the same efficiency curve.

  Moreover, in the design of the power supply unit, the efficiency curve generally has a shape in which the efficiency decreases as the load becomes lower and higher with a certain load factor at the top. For example, in the efficiency curves of the power supply units 20 and 21 in FIG. 4 shown here, the efficiency is highest when the load factor is 0.5 (50%), and the efficiency is low in regions where the load factor is lower than that and regions where the load factor is higher. It has become.

  As described above, since the power consumption of the NIF 30 and the cooling unit 40 changes dynamically, the load factor is set to 0.5 (50%) by controlling the power supply units 20 and 21 according to the changing power consumption. ) Control the number of power supply units 20 and 21 so that they are close to each other (so that the power supply efficiency is the highest), and automatically control to maintain the necessary redundant operation.

Next, the operation method of the network device 10 will be described with reference to FIGS.
FIG. 5 shows a non-redundant operation method of the power supply unit 20. When the load (total) of the NIF 30, the cooling unit 40, the CSU 50, etc. is 2 kW or less at the maximum and two 1.0 kW power supply units 20 are mounted, the operation cannot be performed if one power supply unit 20 fails. That is, only the necessary number of power supply units 20 is provided.

  FIG. 6 shows the N + 1 redundant operation system of the power supply unit 20. If the load (total) is a maximum of 2 kW or less and three 1.0 kW power supply units are mounted, it is possible to operate even if one power supply unit fails. That is, the number of power supply units required + 1 is provided.

  FIG. 7 shows the N + N redundant operation system of the power supply units 20 and 21. When the load (total) is 2 kW or less at the maximum and four 1.0 kW power supply units are mounted, it is possible to operate even if two power supply units fail. That is, it is in a state where twice as many power supply units as necessary are provided.

  Next, the operation of the network apparatus 10 will be described using the basic flowchart of FIG. FIG. 9 shows a sub-flowchart during N + N redundant operation, and FIG. 10 shows a sub-flowchart during N + 1 redundant operation.

(1) The basic flowchart is the system P.30. It is started (80) when any one of ON (power-on), schedule setting, and configuration information change, but each trigger can be enabled / disabled by user setting. In the schedule setting, for example, the basic flowchart can be operated with an arbitrary schedule such as setting a predetermined date and time or setting a predetermined time interval.
(2) When the basic flowchart is started, the CSU 50 performs Yes / No determination of energy saving operation (S801). In the case of Yes, go to S802, and in the case of No, as a normal operation, for example, all power supply units are connected to P.I. Set to ON (S808). Whether or not to perform the energy saving operation (Yes / No) is set in advance by the user.
(3) In S802, the CSU 50 determines the operable number of power supply units from the difference between the mounted number and the number of faulty units. Calculated based on “operable number = number of installed units−number of faults”.

  (4) In S803, the CSU 50 determines the redundancy method that can be selected by the user at the time of initial setting from the operable unit number (operating unit) and rated capacity (rated power) of the power supply unit and the power consumption of the load unit according to the following: Display on an external console. If the aggregated power consumption (power consumption of the entire load unit) is true if “power consumption ≦ number of operating units × rated capacity / 2” and “number of operating units ≧ 2” is true, the N + N redundancy or N + 1 redundancy operation of the power supply unit is performed. Judge that it is possible. If “power consumption ≦ (number of operating units−1) × rated capacity” is true, it is determined that N + 1 redundancy of the power supply unit is possible. If any of the above is false, it is determined that the redundant operation of the power supply unit is not possible. Except at the time of initial setting, display on the external console is not performed, and determination of a selectable redundancy method and confirmation of the redundancy method set by the user may be performed.

  (5) In S804, based on the redundancy method set by the user and the determination result of (4), the CSU 50 selects the redundancy method of the network device according to the table of FIG. A warning is displayed when a method different from the user setting is selected and when it is not redundant. For example, if the user sets N + N redundancy and the redundancy method that can be selected in (4) is N + N redundancy, N + N redundancy is selected. However, if the user has set N + N redundancy but only N + 1 redundancy can be selected, N + 1 redundancy is selected and a warning is displayed indicating that there is a redundancy scheme different from the user setting.

  When the N + N redundant operation method is selected, the sub-flowchart of N + N redundant operation of FIG. 9 is performed as S805, and when the N + 1 redundant operation method is selected, the sub-flowchart of N + 1 redundant operation of FIG.

(6) As S805, the CSU 50 performs the P.P. Determine the number of power supply units to be turned on.
(6a) Calculation of n (S901)
n = number of operable power supply units / 2 (n is rounded down after the decimal point).
(6b) The number of power supplies with the best power supply efficiency η is selected (S902).
Lf_aim is the load factor at which the power supply efficiency is most improved, and each load factor when the number N of power supplies is 1 to n is obtained by “power consumption / (N × 2 × rated power)” (N = 1, 2,... N), The absolute value “Dffn = | Lf_aim−power consumption / (N × 2 × rated power) |” (N = 1, 2,... N) of N is the smallest power supply efficiency η The number of power supplies is good.
(6c) Configuration information The position where the power supply units 20 and 21 are mounted is associated with the input system in the configuration information.
(6d) P.I. Number of power supply units to be turned on (S903)
Refer to the configuration information, and connect N power supply units from each input system. Turn on.

(7) In step S806, the CSU 50 executes the P.P. Determine the number of power supply units to be turned on. (7a) Calculate n (S1001)
n = Number of power supply units that can be operated-1 (n is rounded down).
(7b) The number of power supplies with the best power supply efficiency η is selected (S1002).
Lf_aim is a load factor at which the power supply efficiency is most improved, and each load factor when the number of power supplies N is 1 to n is obtained by “power consumption / (N × rated power)” (N = 1, 2,... N), and Lf_aim The absolute value “Dffn = | Lf_aim−power consumption / (N × rated power) |” (N = 1, 2,... N) where N is the smallest and the number of power supplies with the highest power supply efficiency η. N.
(7c) P.I. Number of power supply units to be turned on (S1003)
Arbitrary N + 1 power supplies Turn on.

The operation using the flowchart of FIG. 8 will be described using an example using the network device 10 of FIG.
It is assumed that the rated power of the power supply units 20 and 21 of this embodiment is 1000 W and the efficiency curve is as shown in FIG. It can be seen from FIG. 4 that the power supply unit used in the embodiment has the highest efficiency when the load factor is 0.5.

  Next, the configuration of this network device 10 includes one unit for the basic unit CSU50, one unit for the redundant unit CSU51, four units for the NIF 30, and two units for the cooling unit 40, and further includes a power supply unit 20.21 for supplying power to them. All units are operating normally in a configuration of N + N redundant operation with a total of 4 units of 2 units each.

  During normal operation, power is monitored in real time by the power monitor circuit 61 in the input unit of each unit, and the information is sent to the basic unit CSU 50 and compiled by the CSU 50. The values are 400 W for each of the CSUs 50 and 51. , NIF30 is 250W each and cooling unit is 25W. That is, the total power consumption of the network device 10 during normal operation is “400 W × 2 + 250 W × 4 + 25 W × 2 = 1850 W”.

  The network device 10 according to the present embodiment controls the number of operating power supply units according to the flowchart of FIG. First, if the user has set in advance whether or not to use the network device 10 in the energy saving operation mode and has selected the energy saving operation mode, then the number of operating power supply units is confirmed (S802). Here, since the number of power supply mounted units is 4 units and all are operating normally, “operating units = mounted units−failure units = 4 units−0 units = 4 units”. Next is display of the selectable redundancy mode (S803).

  "Number of power supply units operating x rated capacity / 2 = 4 units x 1000W / 2 = 2000W ≥ 1850W", which is larger than the required power of the device 1850W and "number of power supply operating ≥ 2". Since it is determined that the user setting is N + N redundancy and that N + N redundancy is possible according to the table of 11, the N + N redundancy operation is displayed on the external console of the apparatus.

  If the power consumption of the NIF 30 being used is 350 W, the total power consumption of the device is “400 W × 2 + 350 W × 4 + 25 W × 2 = 2250 W”, “power consumption ≦ operating number × rated capacity / 2 = 2000 W” and “ "Operating number> 2" is false and N + N redundancy is not possible. “Power consumption ≦ (Number of operating units−1) × Rated capacity = 3000 W” is true and it is determined that N + 1 redundancy is possible, and it is determined that N + N redundancy and N + 1 redundancy are possible according to the table of FIG. Is displayed on the external console of the apparatus, and a warning is displayed that it is different from the N + N redundancy that is the user setting. The table in FIG. 11 is an example and does not have to be exactly the same as this table.

  In this embodiment, it is determined that N + N redundancy is possible, and the N + N redundant operation sub-flowchart (S805) of FIG. 9 is executed. In n calculation, “N = number of operating units / 2 = 2”. From the efficiency curve of the power supply unit of the present embodiment, the load factor with the highest efficiency is “Lf_aim = 0.5”. Then, “Dffn = | Lf_aim−power consumption / (N × 2 × rated power) |” is calculated when N = 1 and N = 2. When N = 1, “Dff1 = | 0.5−1850 W / (1 × 2 × 1000 W) | = 0.425”, and when N = 2, “Dff2 = | 0.5−1850 W / (2 × 2 × 1000 W) | = 0.0375 ”, and when N = 2 is minimum, N = 2 is selected. In S807, as the power supply operating status display, it is displayed that the currently operating power supply unit is in N + N redundant operation.

  The configuration information of the network device 10 according to the present embodiment includes the information in FIG. 7, that is, the configuration information in which each power supply unit 20 has one system and the power supply unit 21 has two systems. Two units and two units are selected from the power supply units 21 and set as operation power supply units. In the present embodiment, since the power supply unit 20 and the power supply unit 21 are two at the maximum, P.I. Although there are no power supply units to be turned off, if there are three or more power supply units each, one power supply unit is connected from the CSU to the PC. Instruct OFF. As a result, the network device 10 is operated with the number of power supply units having the best power supply efficiency, and power consumption can be suppressed.

  FIG. 12 is a diagram showing an example of the effect at the time of N + 1 redundancy.

DESCRIPTION OF SYMBOLS 10 ... Network apparatus 20 ... One-system power supply unit 21 ... Two-system power supply unit 30 ... NIF
40 ... Cooling unit 50 ... Basic part CSU
51 ... Redundant part CSU
60 ... Power supply bus line 61 ... Power monitor circuit 62 ... Power monitor signal 63 ... Power supply unit ON / OFF control signal 64 ... Mounting / failure information signal

Claims (6)

Multiple cooling units;
A plurality of interface parts;
A control unit for controlling the entire apparatus;
A plurality of power supply units,
The cooling unit, the interface unit, and the control unit each have a power monitor circuit,
The control unit determines the number of operations of the power supply unit based on the power consumption collected from the power monitor circuit and the power supply efficiency of the power supply unit measured in advance. The communication device according to claim 1,
When determining the number of operations of the power supply unit, the control unit has a redundant configuration in which a predetermined number is added to the number of the power supply units necessary to maximize the power supply efficiency. A communication device. The communication device according to claim 2,
The control unit obtains the operable number of power supply units from the difference between the number of mounted units and the number of faulty units, and determines the number of operations of the power supply unit with the operable number as an upper limit. The communication device according to claim 2,
The communication unit is characterized in that, as a redundancy method, when the number of units necessary for maximizing power supply efficiency is N, N + 1 is selected. The communication device according to claim 2,
The control unit adopts a method of selecting N + N units as a redundancy method, where N is the number necessary for maximizing power supply efficiency. The communication device according to claim 3,
When the control unit can take a plurality of redundant configurations and non-redundant configurations, the control unit determines the number of operations of the power supply unit so that the configuration is as close as possible to the preset configuration from the operable number. A communication device.

Images