RU2749754C1 - Redundant server device - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: invention relates to the field of information technology, namely to computer technology. A redundant server device containing a processor module, a disk memory module, a power supply, and also includes an installation platform to which at least one disk memory module and at least two processor modules are connected by means of connectors. modules of disk memory, while one of the processor modules has the status of the main one, and the other processor modules have the status of reserve ones, and the status of the processor module changes by means of the software installed on the processor module, and the processor modules are made with the ability to back up each other and transfer between themselves from a given frequency an informational message about its status and its performance.

EFFECT: increased reliability of the server.

13 cl, 4 dwg

Description

FIELD OF TECHNOLOGY

The claimed invention relates to computer technology and can be used in the construction of fault-tolerant server devices for various purposes in the structure of a distributed computing complex, integrated into a local area network, for automation systems, dispatching and other systems.

LEVEL OF TECHNOLOGY

In various dispatching and automation systems, server redundancy is used to increase system reliability. In such schemes, servers can work independently of each other, conduct independent polling of the lower level, each server has its own database. While the system is operating, its built-in mechanisms control the availability of communication with controllers, I / O modules and auxiliary systems. In the event of a server failure, backup facilities allow you to switch to the second server. In this case, client workstations connected to the failed server are also automatically transferred to the working one. On the one hand, this method of organizing work may seem the most reliable, but it has certain technical features, in particular, all lower-level equipment should allow polling by several servers at the same time. The servers conduct independent polling, so data from different machines can enter the system with different time stamps, which greatly complicates the work. The procedure for synchronizing a database (DB) can lead to duplication, which means a distortion of the real picture, therefore, it requires special verification methods (decision-making). Clients become more complex because it becomes unclear which peer server they should contact.

To solve these features, the servers are assigned the status of primary and backup. They constantly exchange data and monitor each other's statuses. If the backup server does not find a server with the primary status on the network, then it changes its status to the primary one. The server with the status of the main one polls the controllers, input / output systems and other equipment and transmits this data to the backup one with the same time stamps as it has itself. As a result, both servers have identical data with the same timestamps. In addition, the main server transmits data to the backup server that are generated in the process of control, calculations, or entered by operators. They must be synchronized between servers for bumpless failover in the event of a primary server failure. Clients over the network automatically connect to the main server. If such a server breaks down, then the clients after switching the backup server to the status of the main one connect to it.

The time of switching from the main to the backup server in case of failure of the first is essential for increasing the reliability of the control system. During this period of time, operators lose control over the object of control and over the operation of the entire system, and also the controllers and input / output systems are not polled and the polled data is not archived, which can be critical if events that can lead to the development of an accident are missed. at the managed object.

The presence of two servers with the main status in the system is unacceptable for the reasons that were indicated earlier. Therefore, the backup server must reliably identify the failure of the primary server before switching its status. If these are two separate computers on the network, then reliable identification of the failure of the primary server can take a significant amount of time (for all available network adapters, requests must be sent and timed out). At the same time, a long period of time for determining the failure of the main server can affect the bumpless switching and cause failures in the operation of the entire control system and the controlled object. To ensure complete bumplessness, server switching times should be less than or comparable to the polling cycle of controllers and I / O modules.

In addition, a very important parameter from the point of view of the reliability of the control system is the speed of restoring the operability of a failed server, since at this time the working server does not have a reserve and when it fails, the entire system stops working. If the server is a separate computer installed in a rack, then it will take quite a long time to remove and repair it or replace it with another one with the configuration and installation of the necessary software.

From the prior art known backup memory, described in US patent No. 10,255,149, G06F 11/20, publ. 04/09/2019. It includes the first port; a second port different from the first port; a first storage device connected to a first port; and a second storage device connected to a second port. The first memory changes the operating mode of the second memory from standby to active using internal communication. A server is made on the basis of this redundant storage device.

The disadvantage of such a device is the absence of a redundant server, which reduces the reliability of the device.

A redundant server device is also known from the prior art according to RF patent No. 117199, G06F 15/16, publ. 06/20/2012 (analogue of the proposed solution).

The redundant server contains a shared memory block and two backup channels, each of which includes a computer, while the shared memory block is made in the form of two redundant disk memory modules, and the computers are interconnected through dedicated redundant machine-to-machine communication channels and are connected via control buses and corresponding interfaces to each disk module.

The disadvantage of such a server is the presence of separate computers for each server device, as well as necessarily a common disk memory for both computers, which limits the reliability of the device.

To overcome the disadvantages of systems and devices known from the prior art, there is a need for reliable fault-tolerant equipment, such as a redundant server, which can be implemented using at least two hot-swappable processor modules, at least one disk device. hot-swappable memory and at least one power supply unit made as a single device.

The technical result of the claimed invention is to provide a reliable server that ensures fault tolerance of systems for various purposes created with its use, including automation and dispatching systems.

The technical result is achieved through a redundant server device containing a processor module, a disk memory module, a power supply, while it includes an installation platform to which at least one disk memory module and at least two processor modules are connected via connectors , made with the ability to back up each other with the transfer of information about their performance, and connected to each other and disk memory modules through the connectors of the installation platform. Where one of the processor modules has the status of the main, and the other processor modules have the status of reserve, while the status of the processor module is changed by means of the software installed on the processor module.

In one particular case of the device, each processor module can interact with at least one disk memory module to which other processor modules do not have access. In another particular case of the device, the processor module can be combined with the disk memory module into a single module. At least one communication module can be installed on the installation platform by means of connectors. In this case, the processor module and the communication module can be provided with connectors for connecting external devices. In addition, external devices can be connected to the installation platform via connectors. To interact with external devices, the processor and communication modules can be equipped with radio interfaces: wi-fi, Bluetooth, etc.

The disk memory module contains at least one drive. All modules installed on the installation platform, as well as drives, can be hot-swappable.

To demonstrate its operability, each processor module generates at least one analog, discrete, frequency or other physical signal, which is controlled by other processor modules, or each processor module, via a dedicated interface, proactively or upon request, transmits an information message to the rest of the processor modules ...

The power supply can be installed on a docking platform via a connector or combined with a docking platform. The power supply can be hot-swappable.

BRIEF DESCRIPTION OF DRAWINGS

The claimed technical solution can be illustrated by the following figures of the drawings, where

in fig. 1 is a general diagram of a preferred embodiment of a redundant server device;

in fig. 2 shows a diagram of a particular case of a redundant server device;

in fig. 3 is a diagram of a particular case of an example of implementing a redundant server device;

in fig. 4 shows a diagram of another particular case of an example of implementing a redundant server device.

The following designations are adopted in the drawings:

1 - installation platform (UP);

2 - processor modules (PM);

3 - disk memory modules (DM);

4 - power supplies (PSU);

5 - connectors for communication interfaces with external devices;

6 - connectors for power supply;

7 - installation socket for processor module 2 (PM);

8 - installation connector for disk memory module 3 (DM);

9 - installation connector for power supply 4 (PSU);

10 - communication modules (KM);

11 - installation connector for communication module 10 (KM);

12, 13, 14 - monitor, computer keyboard, computer mouse;

15 - controller;

16 - client's computer;

17 - input / output module;

18, 19, 20 - connectors for connecting external devices.

SUBSTANCE OF THE STATED INVENTION

The specified technical result is achieved due to the development of a redundant server device, which has several special cases of implementation.

So, in a preferred embodiment, the redundant server device (Fig. 1) contains an installation platform 1 (UE), on which at least two processor modules 2 (PM) are installed by means of connectors 7, at least two processor modules (PM) are installed by means of connectors 8. one disk memory module 3 (DM) and through connectors 9 installed at least one power supply 4 (PSU). The installation platform 1 also provides connectors 5 for connecting external devices such as client computers, controllers, I / O modules and systems, monitors, computer keyboards, manipulators (for example, a computer mouse), printers, etc., via wired interfaces, such such as Ethernet, FOCL, USB, HDMI, RS-232, RS-485 and others. Data from the wired interfaces through connectors 5 of the installation platform 1 (UE) are transmitted to the processor modules 2 (PM) through connectors 7.

There is at least one power connector 6.

The installation platform 1 (UE) can have a subrack design.

One of the installed processor modules 2 (PM) has the main status, which, together with other modules included in the server device, performs the computing, communication, archiving and other functions assigned to the server device. The remaining processor modules 2 (PM) are in hot standby and have a reserve status, with the ability to monitor the operability and status of the remaining processor modules 2 (PM), synchronize data with the main processor module 2 (PM) and the ability to switch to the main status when a failure is detected main processor module 2 (PM). If the operation of the main processor module 2 (PM) fails, one of the backup processor modules 2 (PM) switches to the status of the main one. The principles of choosing a backup processor module 2 (PM) to designate it as the main one when using three or more processor modules 2 (PM) in one server device may be different. For example, the processor module 2 (PM) installed in one or another slot 7 of the installation platform 1 (UE) may have priority for designation as the main one. In case of replacement of a failed processor module 2 (main or backup) (PM) while the server device is running, the software of the newly installed processor module 2 (PM) can request from the main processor module 2 (PM) all the files necessary to start the control task, data and settings, start the control task and only then set the backup status for this newly installed processor module 2 (PM).

Processor modules 2 (PM) may differ from each other in their characteristics, for example, the type and frequency of the processor, the amount of RAM, etc., but at the same time they must be able to perform the set of functions assigned to them. That is, in the status of the main one, each of them could perform a set of functions assigned to this redundant server, and in the status of the backup one could perform the functions of monitoring the operability of the main processor module and synchronizing data with it.

The decision to assign or change the status of processor modules 2 (PM) is made by the software installed in them.

To identify the operability, the processor modules 2 (PM) can transmit information to each other in the form of physical signals, for example, pulses with a given frequency (for example, 1 kHz). The absence of pulses for a specified time (for example, 100 ms) means a failure of this processor module 2 (PM). To generate physical signals, each processor module 2 (PM) must have a discrete, analog or other output, depending on which the type of signal is generated. In addition, to control the physical signals generated by other processor modules 2 (PM), each processor module 2 (PM) must have physical inputs corresponding to the type of output. The number of inputs must be one less than the number of processor modules 2 (PM) installed on the installation platform 1 (PM). Physical signals from the output of one processor module 2 (PM) to the inputs of the remaining processor modules 2 (PM) can be transmitted, for example, through the contacts on connectors 7 of the installation platform 1 (UP), intended for connecting processor modules 2 (PM).

Another way to transfer information about the health of processor modules (PM) can be the use of interfaces, for example, Ethernet. In this case, each processor module 2 (PM) can, for example, send to other processor modules 2 (PM) with a given frequency an information message about its operability and its status. The absence of several such messages in a row can be perceived as a failure of this processor module 2. It is important that the interface used to transmit messages about the operability of processor modules 2 (PM) is used only for redundancy functions of processor modules 2 (PM) and does not participate in data exchange with external devices so that the operation of external devices can never affect the redundancy. This allows you to significantly reduce the time for making a decision on the operability of processor modules and ensure shockless switching, thereby ensuring reliable operation of the device and the entire system with which it interacts.

At least one disk memory module 3 (DM) can be connected to the installation platform 1 (UE) by means of connectors 8. It is intended for storing archives of technological parameter values received from controllers, other computers, input / output modules, other external equipment or entered by the operator. In addition, this disk memory module 3 (DM) can store event and information security logs, reports and other documents and databases that can be generated or used by a control system based on the claimed redundant server device. From this disk memory module 3 (DM) processor modules 2 (PM) can read the necessary archive, reference and other data for their subsequent processing or transmission to the client computers connected to the server.

Disk storage modules 3 (DM) can be created on the basis of different data storage devices, for example, HDD magnetic disks, solid state disks (SSD) or any other storage devices. Each disk memory module 3 (DM) must contain at least one drive. When two or more drives are used in the disk memory module 3 (DM), their memory can be used either as separate disks, or as a common space for storing data, or one of the drives is used for backup storage of data from other drives, for example, disk memory module 3 (DM) writes all the data it receives in parallel to two drives, which prevents data loss in the event of a failure of one of the drives.

Disk memory modules 3 (DM) can be hot-swappable, as well as hot-swappable for their drives, thereby ensuring reliable operation of the claimed device and the entire system with which they interact.

If only one disk memory module 3 (DP) is installed on the installation platform 1 (UE), then it is accessible from all processor modules 2 (PM) installed on the installation platform 1 (UE).

In the particular case of a redundant server device for each processor module 2 (PM), at least one disk memory module 3 (DM) is installed, available only to this processor module 2 (PM). In this case, the design of the installation platform 1 (UE) provides that each connector 7 for the processor module 2 (PM) is connected to one specific (specific) connector (s) 8 for the disk memory module 3 (DM). Processor module 2 (PM) with the status of the main one saves data to its disk memory module 3 (DM) and transfers the same data to other processor modules 2 (PM) so that they save this data to the corresponding disk memory modules 3 (DM).

In another particular case of a redundant server device, there can be a combination of processor module 2 (PM) and disk memory module 3 (DM) into a single module. This approach has its advantages, since the design of the installation platform 1 (PC) is simplified, but it does not allow independent replacement of processor module 2 (PM) and disk memory module 3 (DM) in case of failure of any of them.

In yet another particular case of a redundant server device (see Fig. 2), at least one communication module 10 (KM) can be installed on the installation platform 1 (UE) via connectors 11, providing additional opportunities for communication with external equipment and redundancy of these connections when installing two or more communication modules 10 (KM). In this case, the communication module 10 (KM) can be connected to external devices, both through standard wired interfaces through the connectors 5 installed on it, and through wireless interfaces. As wired interfaces, such as, for example, Ethernet, FOCL, USB, HDMI, RS-232, RS-485 and others can be used, and as wireless - Wi-fi, Bluetooth and others. For data exchange between processor modules 2 (PM) and communication modules 10 (KM), any interface (for example, Ethernet) can be used, providing the necessary bandwidth for data exchange with all external devices connected to the communication module 10 (KM). The connection of processor modules 2 (PM) with communication modules 10 (KM) is performed through their installation connectors 7 and 11, respectively.

Processor Modules 2 (PM) can also be equipped with wired and wireless interfaces for communication with external equipment. For communication with external equipment via wired interfaces, processor modules 2 (PM) can be equipped with connectors 5.

In the case when each processor module 2 (PM) is assigned one disk memory module 3 (DM) or when combining processor module 2 (PM) with a disk memory module 3 (DM) to synchronize data on disk memory modules 3 (DM) either the internal interface for providing redundancy can be used, through which signs of the state of the processor module 2 (PM) are transmitted, or the external wired interface through connectors 5 on the installation platform 1 (UP), processor modules 2 (PM) or communication modules 10 (KM), or wireless interface of processor modules 2 (PM).

The redundant server device includes at least one power supply unit 4 (PSU). It converts the supply voltage, for example, 220 volts, into the voltage levels necessary for the operation of all modules installed on the installation platform 1 (UE), and providing each of them with the necessary power. These voltages are supplied to processor modules 2 (PM), disk memory modules 3 (DM) and communication modules 10 (KM) through connectors 7, 8, and 11, respectively, through which they are connected to the installation platform 1 (UE).

Power supply 4 (PSU) can be made either built into the installation platform 1 (UE), or connected to the installation platform 1 (UE) by means of connectors 9. Removable power supplies 4 (PSU) are made in such a way that in the presence of a backup power supply 4 (BP) the main power supply unit 4 (BP) could be disconnected from the installation platform 1 (UP) without stopping the operation of the server device, thereby ensuring their hot swap.

To connect the supply voltage to the installation platform 1 (UE), at least one connector 6 must be provided. To ensure redundancy of the supplied voltage, more than one power connector can be provided on the installation platform 1 (UE) 6. The voltage from these connectors is supplied simultaneously to all power supplies 4 (PSU) built into the installation platform (UP) or installed on it.

The above features provide various options for implementing a reliable redundant server device that provides fault tolerance for systems for various purposes created with its use, including automation and dispatching systems.

The operation of the claimed redundant server device can be considered on non-exhaustive examples.

Example 1

FIG. 3 shows a schematic representation of a redundant server device consisting of an installation platform 1 (UP), two processor modules 2 (PM), two disk memory modules 3 (DM) and two power supplies 4 (PSU). All of the above elements of the device are mounted on the installation platform 1 (UE) by means of the corresponding installation connectors. Processor modules 2 (PM), disk memory modules 3 (DM) and power supplies 4 (PS) are hot-swappable.

Each processor module 2 (PM) is connected through installation connectors to one disk memory module 3 (DM).

On the installation platform 1 (UE) there are four connectors 5 for connecting to the Ethernet network. Each processor module 2 (PM) through the installation connector 7 is connected to two Ethernet connectors 5, while one of the connectors 5 is used to connect to the network of controllers 15, and the second connector 5 is used to access it from the client computer 16.

In addition, on the installation platform 1 (UE) there is a connector 20 of the RS-485 interface, which is connected to both processor modules 2 (PM) through their installation connectors 7 and is used to communicate with the I / O modules 17. Polling via RS-485 can Maintain only processor module 2 (PM) with the status of the main one.

Each processor module 2 (PM) has one HDMI connector 18 for connecting a monitor 12 and at least three USB connectors 19 for connecting a computer keyboard 13, a computer mouse 14 and other external devices.

Also on the installation platform 1 (UE) there is a connector 6 for connecting to a 220 volt AC network. Voltage from this connector 6 is supplied simultaneously to both power supplies 4 (PSU) through the installation connectors.

Each of the two power supply units 4 (PSU) operates independently and can provide power to all elements of the redundant server device through their installation connectors when the second power supply unit 4 (PSU) is not working.

Processor modules 2 (PM) can be in the statuses of primary and secondary. Moreover, the statuses of processor modules 2 (PM) may not be the same. Processor module 2 (PM) with the status of the main one receives data from controllers 15 and input / output modules 17 via Ethernet and RS-485 interfaces, saves the received data to the database on the disk memory module 3 (DM) connected to it through the installation connector, transmits the received data and data stored in the database of the disk memory module 3 (DM) to client computers 16, connected to the processor module 2 (PM) via the Ethernet interface, and to the processor module 2 (PM) with the reserve status. Processor module 2 (PM) with the standby status receives data from processor module 2 (PM) with the status of primary and saves them to the database on the disk memory module 3 (DM) connected to it. An Ethernet network is used to transfer data between processor modules 2 (PM).

To transmit information about its operability, each processor module 2 (PM) generates one discrete output signal of the TTL type, which is transmitted through the installation connectors to another processor module 2 (PM) and is controlled by another processor module 2 (PM) via a discrete input. The software of each processor module 2 (PM) generates pulses with a frequency of 1 kHz on the output discrete signal and monitors the presence of pulses at the discrete input of processor module 2 (PM). The absence of pulses for 100 ms is considered a sign of failure of the processor module 2 (PM), which should generate them. If processor module 2 (PM) with the main status fails, then processor module 2 (PM) with the backup status switches its status from the backup to the main one. If the processor module 2 (PM) with the redundant status fails, then no switching is performed. The failed processor module 2 (PM) is removed from the installation platform 1 (PM) and replaced with a serviceable processor module 2 (PM), and information about the failure of processor module 2 (PM) is transmitted to client computers 16 and is recorded in the disk module database. memory 3 (DM) of the processor module 2 (PM) with the status of the main one.

Example 2

FIG. 4 shows a schematic representation of a redundant server device consisting of an installation platform 1 (UP) made in the form of a crate, two processor modules 2 (PM), each of which is combined with a disk memory module 3 (DM) into a single module, two communication modules 10 (KM) and two power supplies 4 (BP). All the listed elements of the device are mounted in crate 1 using connectors. Processor modules 2 (PM), combined with a disk memory module 3 (DM) into a single module, communication modules 10 (KM) and power supplies 4 (PS) are hot-swappable.

On the installation platform (rack) 1 (UE) there are four connectors 5 for connecting to the Ethernet network. Communication modules 10 (KM) interact with processor modules 2 (PM) through their installation connectors via the Ethernet interface and provide four independent Ethernet ports for connection to connectors 5. Thanks to this, all four Ethernet connectors 5 are equivalent in terms of interaction with processor modules 2 (PM). Two of the four slots 5 are used to provide redundant communication with controllers 15, and the other two are used for redundant communication with client computers 16.

Communication modules 10 (KM) reserve each other's work. The connection to slots 5 is received by the communication module that is in the main state. The state of the communication modules 10 (CM) is controlled by the processor module 2 (PM), which is in the main status. To do this, it generates two discrete signals, one for each of the communication modules 10 (KM). These signals are used to switch the state of the communication modules 10 (KM).

In addition, there is a connector 20 of the RS-485 interface on the installation platform (rack) 1 (UE). This connector is connected to both processor modules 2 (PM). They provide communication with I / O modules 17. Polling via RS-485 can only be carried out by processor module 2 (PM) with the status of the main one.

Each processor module 2 (PM) has one HDMI connector 18 for connecting a monitor 12 and three USB connectors 19 for connecting a computer keyboard 13, a computer mouse 14 and other external devices.

Also on the installation platform 1 (UP) there are two connectors 6 for connecting to a 220 volt AC network. The voltage from connectors 6 is supplied to both power supplies 4 (PSU) through the installation connectors. This allows the server device to be backed up by supplying voltage to connectors 6 from different feeders.

Each of the two power supply units 4 (PSU) operates independently and can provide power to all elements of the server device through their installation connectors when the second power supply unit 4 (PSU) is not working.

Each processor module 2 (PM) has two Ethernet interfaces. One of them is used to communicate with communication modules 10 (KM), and the other is used to interact with the second processor module 2 (PM). Connection to these interfaces is made through the installation connectors.

Processor modules 2 (PM) can be in the statuses of primary and secondary. Moreover, the statuses of processor modules 2 (PM) cannot be the same. Processor module 2 with the main status receives data from controllers 15 and I / O modules 17 via Ethernet and RS-485 interfaces, saves them to the database on the built-in disk memory module 3 (DM), transmits the received data from controllers 15 and modules input / output 17 and data from the database, stored on the disk memory module 3 (DM), to client computers 16 and the processor module 2 (PM) with the status of a reserve on the Ethernet interface dedicated for the interaction of processor modules 2 (PM). Processor module 2 (PM) with the backup status receives data from processor module 2 (PM) with the status of primary and saves them to the database on the built-in disk memory module 3 (DM).

To transmit information about its operability, each processor module 2 (PM) via the Ethernet interface intended for interaction with the second processor module 2 (PM) transmits information messages with a frequency of 30 ms. The software of each processor module 2 (PM) monitors the presence of these messages. The absence of three such messages in a row is considered a sign of a processor module failure. If processor module 2 (PM) with the main status fails, then processor module 2 (PM) with the backup status switches its status from the backup to the main one. If the processor module 2 (PM) fails in the standby status, then no switching is performed. The failed processor module 2 (PM) is removed from the installation platform 1 (UE) and replaced with a serviceable processor module 2 (PM), and information about the failure is transmitted to client computers 16 and written to the database on the disk memory module 3 (DM), combined with the processor module 2 (PM) with the status of the main one.

The above examples are not exhaustive, but they show that the claimed redundant server device is reliable, providing fault tolerance of systems for various purposes created with its use, including automation and dispatching systems.

Claims (13)

1. A redundant server device containing a processor module, a disk memory module, a power supply unit, characterized in that it includes an installation platform to which at least one disk memory module and at least two processor modules are connected by means of connectors, connected to each other and to the disk memory modules through the connectors of the installation platform, where one of the processor modules has the status of the main one, and the other processor modules have the status of reserve ones, while the status of the processor module is changed by the software installed on the processor module, and the processor modules are made with the ability to reserve each other and transmit between themselves at a given frequency an information message about its status and about its performance in the form of an output physical signal or through a dedicated interface, which is controlled by other processor modules. 2. The device according to claim. 1, characterized in that the processor module can interact with at least one disk memory module to which other processor modules do not have access. 3. The device according to claim. 1, characterized in that the processor module can be combined with at least one disk memory module into a single module. 4. The device according to claim 1, characterized in that at least one communication module is installed on the installation platform by means of connectors. 5. The device according to claim 4, characterized in that the processor module and the communication module are provided with connectors for connecting external devices. 6. The device according to claim 1, characterized in that the external devices are connected to the installation platform by means of connectors. 7. A device according to claim 4, characterized in that, for interaction with external devices, the processor and communication modules can be equipped with radio interfaces: wi-fi, Bluetooth, etc. 8. The device according to claim. 1, characterized in that the disk memory module contains at least one drive. 9. The device according to claim 8, wherein the drive is hot-swappable. 10. The device according to claim 1, characterized in that the modules installed on the installation platform are hot-swappable. 11. The device according to claim 1, characterized in that the physical signal that each processor module generates to demonstrate its operability is at least one analog, discrete, frequency or other physical signal that is controlled by other processor modules. 12. The device according to claim. 1, characterized in that the power supply unit can be installed on the installation platform by means of a connector or combined with the installation platform. 13. The device according to claim 1, wherein the power supply unit is hot-swappable.

Images