TWI770038B - Remote menagement system and remote menagement method thereof for server - Google Patents

Abstract

A remote management system and a remote management method thereof for server are provided. The remote management system includes at least one server, a gate apparatus and a remote monitoring equipment. The server is placed in sever room, monitoring operating performance itself (such as resource usage rate), and senses sensing data (such as power consumption, temperature, fan speed, etc.) through at least one sensor and even temperature of hot and cold channels. The gateway apparatus is coupled with the server, and forwards the operating performance and sensing data. The remote monitoring equipment is connected to the gateway apparatus, and integrates the operating performance and sensing data to provide a corresponding monitoring operation (such as alarm reporting, graphical interface presenting, air conditioner control, loading transferring, etc.). Accordingly, the maintenance person can know statuses of the server and the server room immediately and solve hot point problem of partial rack effectively.

Description

Remote management system of server and remote management method thereof

The present invention relates to a remote monitoring technology, and more particularly, to a remote management system of a server and a remote management method thereof.

For large companies, information security companies or telecom operators, they need to plan a computer room to place a large number of servers to facilitate centralized control. Since maintenance personnel are not always in the computer room, they usually cannot immediately grasp the situation in the computer room. In the prior art, it has been proposed to monitor the conditions in the equipment room by means of remote monitoring. However, the existing technologies are only aimed at monitoring a single part, and cannot provide a comprehensive monitoring solution for the server and the computer room as a whole.

In view of this, the present invention provides a remote management system of a server and a remote management method thereof, which integrates internal operation performance, sensing data and external environment status, thereby providing a suitable adjustment mechanism.

The remote management system of the server of the present invention includes a server, a gateway device and a remote monitoring device. The server is installed in the computer room, monitors its own operation performance, and senses the sensing data through the sensing element. The gateway device is coupled to the server and transmits operation performance and sensing data. The remote monitoring equipment is connected to the gateway device, and integrates operation performance and sensing data to provide corresponding monitoring operations.

On the other hand, the remote management method of the server of the present invention is suitable for remote monitoring equipment to monitor the server. The distal cannulation method includes the following steps. Monitor the performance of the server itself. Sensing data of the computer room where the sensing server is located. Integrate operational performance and sensing data to provide corresponding monitoring operations.

Based on the above, by integrating the operation performance and external sensing data at the same time, the embodiment of the present invention can further provide a more suitable adjustment mechanism, so as to facilitate the management of maintenance personnel and quickly eliminate obstacles.

In order to make the above-mentioned features and advantages of the present invention more obvious and easy to understand, the following embodiments are given and described in detail with the accompanying drawings as follows.

1 is a schematic structural diagram of a remote management system 1 according to an embodiment of the present invention. The remote management system 1 includes one or more servers 100 and one or more sensing elements 130 (external or built-in in Server 100 ), air conditioner temperature controller 150 , computer room cold and hot aisle sensor 170 , gateway device 200 , remote monitoring equipment 300 and user equipment 400 .

These servers 100 are respectively, partially or fully installed in a computer room (not shown in the figure), and the servers 100 can be any type of servers such as mobile data, network, file, database, mail, domain name, proxy, application, etc. . The server 100 is loaded with a server management software 101 to sense its internal central processing unit (CPU), memory, and input/output (I/O) utilization rate through a system basic input output system (BIOS) 103 . The sensing element 130 may be a sensing element for sensing the temperature, power consumption, fan speed and other sensing values of the server 100 . The management software 101 of the server 100 can also obtain the sensing data of the aforementioned sensing value through the sensing element 130 .

An air conditioner temperature controller 150 is arranged in the computer room, and a cold and hot aisle sensor 170 is arranged in the cold and hot aisles of the computer room. The air conditioner temperature controller 150 is a specific circuit, chip or microprocessor to directly control functions such as temperature, wind direction, air volume, and operation of the air conditioner. The computer room hot and cold aisle sensor 170 is used for sensing temperature and other sensing data in the computer room. It should be noted that other types of sensing elements can be installed in the computer room, which can be adjusted according to the needs of the user.

The gateway device 200 is installed at the computer room end, and is integrated with the management software 101 of the server 100, the air conditioner temperature controller 150 and the cold and hot aisle sensor 170 of the computer room through a web service (web service) application programming interface (API). The device 200 can obtain the operation performance and sensing data collected by the server management software 101 and transfer it to other networks (eg, the Internet, etc.) outside the server 100 .

The remote monitoring device 300 can be an electronic device such as a cloud server, a personal computer, a computer workstation, etc., and can be connected to the gateway device 200 to send and receive data.

The user equipment 400 can be an electronic device such as a smart phone, a tablet computer, a desktop computer, a notebook computer, etc., and can be connected to the remote monitoring device 300 to send and receive data. The user equipment 400 is loaded with a browser or a monitoring program to provide a user interface for convenient operation by maintenance personnel.

The above is the description of each device in the remote management system 1 , and then the operation process of the embodiment of the present invention will be described with the aforementioned devices. Each process of the method can be adjusted according to the implementation situation, and is not limited to this.

Please refer to FIG. 2 which is a flowchart of a remote management method of the server 100 according to an embodiment of the present invention. Since the CPU, memory, and input/output device are the three core components of the server 100 , the performance data (eg, utilization rate (%)) can be used to evaluate the overall operation performance of the server 100 . The server 100 monitors the utilization rate of CPU, memory and input/output device through its management software 101, and integrates the aforementioned data to calculate the operation performance such as the resource utilization rate of the server 100 (step S21). A larger value of the resource utilization ratio indicates that the server 100 is busy.

On the other hand, the management software 101 of the server 100 also senses sensing data such as power consumption, temperature, and fan speed of the server 100 through the sensing element 130 (step S23 ), and the computer room cold and hot aisle sensor 170 senses the sensing data of the server 100 . The temperature of the hot and cold aisles can also obtain other external environmental conditions such as the temperature of the incoming and outgoing air. The gateway device 200 then transmits the operation performance, sensing data and external environment status to the remote monitoring device 300 .

After the remote monitoring device 300 obtains the aforementioned data, it can integrate data such as operation performance, sensing data, and external environment status to provide corresponding monitoring operations (step S25 ). There are many embodiments of the monitoring operation. Several examples are listed below to illustrate, and the user can adjust it according to the needs.

Monitoring operation - interface rendering operation:

The user equipment 400 can run a browser and a monitoring program to present sensing data such as power consumption, temperature, fan speed and other sensing data of the server 100 , resource utilization and even the temperature of the hot and cold aisles through a graphical user interface. By obtaining the resource utilization rate of the server 100, the current operating performance of the server 100 can be understood. For example, the low, medium and high workload groups of the server 100 can be classified based on the value of the data utilization rate, and different colors can be used to indicate the difference. group, so that the maintenance personnel can understand the current operation performance of the server 100 in different colors in the user interface. As for the temperature information of the computer room and the temperature of the hot and cold aisles, the remote monitoring device 300 classifies the temperature of the server 100 and the hot and cold aisles into different colors, and uses different colors to create a two-dimensional (2D)/three-dimensional (3D) temperature field of the cabinet, so that the user equipment A 3D temperature field of the cabinet is established on the user interface of the 400, so as to effectively grasp the hot spots and surrounding conditions of the computer room. The remote monitoring device 300 can also use the power consumption information of the server 100 to establish a level 3 index of energy use efficiency (PUE) (PUE Level 1, 2, and 3 represent the different measurement positions of the electricity meter, and Level 1 refers to the amount of The switchboard side is measured, Level 2 means the uninterruptible power system (UPS) side is measured, and Level 3 means the power consumption side of the server 100 is measured), and the PUE is displayed on the user interface of the user equipment 400 Level 3 indicators to understand the power efficiency of the computer room.

Monitoring operation - alarm notification operation:

When the operation performance and the sensing data meet the notification conditions, the remote monitoring device 300 performs an alarm notification operation on the user equipment 400 . For example, for the abnormal air dispersion alarm, the remote monitoring device 300 sets the server 100's high temperature threshold (>25°C), the upper limit value of resource utilization (>50%), and the lower limit value of the fan speed (<10cfm). When the temperature is too high, the resource utilization rate is too low, and the fan speed is too slow, the remote monitoring device 300 automatically sends a fan abnormality alarm to the user equipment 400, so that the user interface of the user equipment 400 displays the fan abnormality alarm message. For hotspot alarms, the remote monitoring device 300 sets the IT server temperature threshold (>25°C) and the upper limit value of device resource utilization (>50%). When the temperature and resource utilization are too high, the remote monitoring device 300 will A hotspot alarm is automatically issued to the user equipment 400, so that the user interface of the user equipment 400 presents a hotspot alarm message.

Monitoring Operations - Offloading Operations:

Assuming that there are more than two sets of servers 100, when the operating performance of one set of servers 100 exceeds an operating threshold, the remote monitoring device 300 performs a load transfer operation to transfer the workload of this server 100 to another server device 100. For example, the remote monitoring device 300 distinguishes the resource utilization rate range, assuming that the resource utilization rate (<25%) is a low utilization rate, the resource utilization rate (between 25% and 50%) is a medium utilization rate, and the resource utilization rate (>50%). %) is a high usage rate. Set the temperature threshold (>25°C), resource utilization threshold (>50%) and load value upper limit (>3kW) of the server 100. When the temperature, resource utilization and load values of a group of servers 100 are too high, The remote monitoring device 300 provides virtual machine (VM) load transfer suggestions, recommending users to transfer the workload of the server 100 with high resource utilization, load, and temperature to low resource utilization (<25%) and low load. (<3kW), lower temperature (<25°C) server 100, or direct workload transfer. The user interface of the user equipment 400 can present the aforementioned load transfer suggestion, so that the maintenance personnel can further analyze and control the transfer of the workload. It should also be noted that the aforementioned threshold and upper and lower limit values can be adjusted according to actual needs.

Monitoring Operation - Air Conditioning Control Operation:

The remote monitoring device 300 sets the monitoring target and the upper and lower temperature limits. The target can be selected from the inlet and outlet air temperature of a single server 100 , the average inlet and outlet air temperature of all servers 100 , or the temperature of the hot and cold aisles. When the selected object and its conditions (eg, greater than the upper limit temperature) are satisfied, the remote monitoring device 300 notifies the air conditioner temperature controller 150 to automatically cool down the hot spot of the object, and when the temperature reaches the standard (eg, less than the upper limit temperature), the remote monitoring The device 300 then informs the air conditioner temperature controller 150 to stop cooling, which can effectively improve the air conditioner energy saving rate and the energy efficiency of the equipment room. In actual design, the air outlet of the air-conditioning equipment can be directed to a single or multiple servers 100 or the cabinets provided therein, so as to achieve targeted temperature regulation of hot spots.

(1) ，其中T_output 為控制溫度、T為溫度、t為時間、K_P 、K_t 、K_d 為調整參數。再透過死區運算分析技術解決控制穩定度問題，當即時量測值(PV)進入控制目標值(SP)之遲滯區範圍，便停止演算，即可解決上述問題。It should be noted that, in order to provide a more accurate temperature controller, the present invention uses a proportional-integral-derivative (PID) algorithm with a dead zone (dead zone) operation analysis technology, thereby improving the control accuracy and stability, and solve the nonlinear problem of control input caused by physical constraints, so as to master the control accuracy. And the equation looks like this:

(1) , where T _output is the control temperature, T is the temperature, t is the time, and K _P , K _t , and K _d are the adjustment parameters. Then, the problem of control stability is solved through the dead zone calculation analysis technology. When the real-time measured value (PV) enters the hysteresis zone of the control target value (SP), the calculation is stopped, and the above problem can be solved.

The aforementioned monitoring operations may be run concurrently or may be selected partially or entirely performed via the remote monitoring device 300 or the user device 400 .

To sum up, the remote management system and the remote management method of the present invention can facilitate maintenance personnel to know the status of the server 100 and the computer room anytime and anywhere, and can obtain alarm information in real time to speed up the elimination of obstacles through different monitoring operations. work. On the other hand, combined with air conditioning control, the embodiment of the present invention can automatically cool down the target server 100, effectively solve the problem of local hot spots in the cabinet, thereby achieving energy saving of the air conditioning and improving the energy efficiency of the equipment room.

Although the present invention has been disclosed above by the embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, The protection scope of the present invention shall be determined by the scope of the appended patent application.

1‧‧‧Remote Management System 100‧‧‧Server 101‧‧‧Server Management Software 103‧‧‧System Basic Input Output System 130‧‧‧Sensing Components

150: Air conditioner temperature controller

170: cold and hot aisle sensor in computer room

200:Gateway device

300: Remote monitoring equipment

400: User Equipment

S21~S25: Steps

FIG. 1 is a schematic structural diagram of a remote management system according to an embodiment of the present invention. FIG. 2 is a flowchart of a remote management method of a server according to an embodiment of the present invention.

Steps S21~S25‧‧‧

Claims (7)

A remote management system for servers, comprising: at least one server, monitoring its own central processing unit, memory and input/output device usage to calculate an operation performance, and sensing a sensor through at least one sensing element measurement information; a gateway device, coupled to the at least one server, and transmitting the operation performance and the sensing data; an air conditioner temperature controller, coupled to the gateway device, wherein the air conditioner temperature controller uses proportional-integral - Differential (proportional-integral-derivative, PID) algorithm combined with dead zone (dead zone) operation analysis technology; and a remote monitoring device, connected to the gateway device, and integrating the operation performance and the sensing data to provide a monitoring operation corresponding to the operation performance and the sensing data, and to control the air conditioner temperature controller to perform the air conditioner control operation, wherein the monitoring operation includes a load transfer operation, and the at least one server includes two of the servos When the operating performance of the server exceeds an operating threshold or the sensing data exceeds a sensing threshold, the remote monitoring device uses the operating performance, the sensing data and a workload value of the server Execute the load shifting operation to shift the workload of the server whose operational performance exceeds the operational threshold, the sensing data exceeds the sensing threshold, and a workload value of the server is high to that the operational performance does not exceed The operational threshold, the sensing data does not exceed the sensing threshold, and another server with a low workload value of the server. The remote management system as described in item 1 of the scope of the application further comprises a user equipment connected to the remote monitoring equipment, and the monitoring operation includes an alarm notification operation. When the operation performance and the sensing data are consistent with When a notification condition occurs, the remote monitoring device performs the alarm notification operation on the user equipment. The remote management system as described in item 1 of the scope of the application further comprises a user equipment connected to the remote monitoring equipment, and the monitoring operation includes an interface presentation operation, the remote monitoring equipment passes through the user equipment The operation performance and the sensing data are presented in a graphical user interface to perform the interface presentation operation. The remote management system as described in claim 3, wherein the remote monitoring device classifies the operating performance and the sensing data into different groups according to their degrees, and uses different colors to represent the different groups. The remote management system as described in claim 1 of the claimed scope further comprises a computer room cold and hot aisle sensors both coupled to the gateway device, the at least one server is installed in the at least one computer room, and the monitoring operation includes a In the air conditioning control operation, the remote monitoring device obtains the aisle temperature sensed by the cold and hot aisle sensors in the equipment room, and integrates the operation performance and the sensing data to control the air conditioning temperature controller to execute the air conditioning control operation. The remote management system of claim 5, wherein the at least one server includes a plurality of the servers, and the air conditioner temperature controller adjusts the temperature for one of the servers. A remote management method of a server, suitable for a remote monitoring device to monitor at least one server, the remote management method comprising: monitoring the CPU, memory and input/output device usage rate of the at least one server to calculate an operation performance; sensing a sensing data through at least one sensing element; and integrating the operation performance and the sensing data to provide monitoring operations corresponding to the operating performance and the sensing data, wherein the monitoring operations include an air-conditioning control operation, and the steps of integrating the operating performance and the sensing data to provide the corresponding monitoring operations include: obtaining a computer room cooling and heating a channel temperature sensed by a channel sensor; and integrating the operation performance and the sensing data to control an air conditioner temperature controller to perform the air conditioner control operation, wherein the air conditioner temperature controller utilizes proportional-integral-derivative (proportional-integral) -derivative, PID) algorithm and with dead zone (dead zone) operation analysis technology; wherein the monitoring operation includes a load transfer operation, the at least one server includes two of the servers, when the operation performance of one of the servers exceeds When an operating threshold or the sensing data exceeds a sensing threshold, the remote monitoring device performs the load shifting operation according to the operating performance, the sensing data and a workload value of the server, so that the operating performance exceeds The operating threshold, the sensing data exceeds the sensing threshold, and a workload of the server with a high workload value of the server is transferred to the operating performance that does not exceed the operating threshold, the sensing data does not exceed the sensing threshold A threshold is measured and another of the server has a low workload value for the server.

Images