JP2009287883A - Temperature anomaly cause portion determining device and temperature anomaly cause portion determining method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a temperature anomaly cause portion determining device and method, efficiently and automatically determining a temperature anomaly cause portion in a passage of cooling air supplied to an equipment group. <P>SOLUTION: The temperature anomaly cause portion detecting device 1 determining a temperature anomaly cause portion in the passage of cooling air supplied from one or a plurality of cooling sources, is equipped with: a temperature anomaly detection result acquiring part 11 acquiring a detection result of temperature anomaly of the equipment; and a determining part 12 determining the temperature anomaly cause portion in the passage of cooling air on the basis of passage information of the cooling air supplied to the equipment group and the detection result of temperature anomaly of the equipment. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a temperature abnormality cause location determination apparatus and a temperature abnormality cause location determination method for determining a temperature abnormality cause location in a flow path of cooling air supplied to a device group.

Based on the measurement result of the high-pressure side pressure of the air conditioner, it is determined whether there is a malfunctioning air conditioner, and based on the determination result, the air conditioner that performs complementary control to increase the cooling capacity of other air conditioners A monitoring system has been proposed (see, for example, Patent Document 1).
JP 2006-64254 A

  Equipment such as a server device has a determined temperature range in which the device can be operated. For example, in a space such as a data center where a large number of server devices are mounted at high density, a cooling function is important. As shown in FIG. 22, in the data center, the air conditioner 2 as a cooling source sends cooling air to the server group (a plurality of server devices 31) mounted on the rack 3 to cool the server device 31. Assume a system. The cooling air is sent from the air conditioner 2 to the server device 31 in the rack 3 and exhausted from the server device 31 according to the flow path indicated by the arrow in the figure. F in the drawing is intake air to the server device 31, and G is exhaust air from the server device 31. Hereinafter, in order to discuss cooling of the server device 31, attention is paid to intake air. The server device is connected to the monitoring device 10 through the network 4. Each server device 31 includes a temperature sensor. An obstacle (for example, a bundle of cables) or the like is installed in the flow path from the air conditioner 2 to the server device 31 due to some mistake, and the server device 31 is not sufficiently cooled. When the temperature of the intake air rises to an abnormal value, a temperature sensor provided in the server device 31 detects that the temperature of the intake air has risen to an abnormal value (temperature abnormality). Then, the server device 31 notifies the monitoring device 10 of the detection result of the temperature abnormality. The monitoring device 10 determines, for example, which server has detected the temperature abnormality based on the detection result of the temperature abnormality notified from the server device 31.

  However, depending on the system shown in FIG. 22, it is possible to determine which server device has detected a temperature abnormality. For example, a location that causes a temperature abnormality of the server device, such as a place where an obstacle is installed. (Temperature abnormality cause location) cannot be determined automatically. For example, when the air conditioner 2 is under the floor or the like, it is difficult for the monitor to visually identify the cause of the temperature abnormality. Furthermore, when a plurality of server devices 31 detect temperature abnormality, it is difficult to determine the cause of temperature abnormality. Although it is conceivable to install a monitoring camera or a sensor for detecting an obstacle to identify the cause of the temperature abnormality, there is a problem in that the installation of the monitoring camera or the like is costly.

  An object of the present invention is to provide a temperature abnormality cause point determination device that efficiently and automatically determines a temperature abnormality cause point in a flow path of cooling air supplied to a device group.

  It is another object of the present invention to provide a temperature abnormality cause location determination method for efficiently and automatically determining a temperature abnormality cause location in a flow path of cooling air supplied to a device group.

  This temperature abnormality cause location determination device is a temperature abnormality cause location determination device that determines a temperature abnormality cause location in a flow path of cooling air supplied from one or more cooling sources to a device group composed of a plurality of devices. Based on the temperature abnormality detection result acquisition means for acquiring the temperature abnormality detection result of the device, the flow path information of the cooling air supplied to the device group, and the temperature abnormality detection result of the device, Judgment means for judging the location of the temperature abnormality in the cooling air flow path is provided.

  The temperature abnormality cause location determination method is a temperature abnormality cause location determination method for determining a temperature abnormality cause location in a flow path of cooling air supplied from one or a plurality of cooling sources to a device group including a plurality of devices. A temperature abnormality detection result of the device is obtained, and the flow of the cooling air is determined based on flow path information of the cooling air supplied to the device group and the temperature abnormality detection result of the device. Determine the cause of temperature abnormality on the road.

  According to this temperature abnormality cause location determination apparatus and this temperature abnormality cause location determination method, a temperature abnormality cause location in a flow path of cooling air supplied from one or a plurality of cooling sources to a device group consisting of a plurality of devices is determined. It becomes possible to automatically specify the temperature abnormality detection result of the device and the flow path information of the cooling air.

  FIG. 1 is a diagram illustrating a configuration example of a temperature abnormality cause location determination apparatus according to the present embodiment. The temperature abnormality cause location determination apparatus 1 determines a location (temperature abnormality cause location) that causes a temperature abnormality in a flow path of cooling air supplied from one or a plurality of cooling sources to a device group including a plurality of devices. Is a processing device.

  In FIG. 1, a server apparatus is described as an example of the device. However, the temperature abnormality cause location determination device 1 according to the present embodiment notifies the temperature abnormality cause location determination device 1 of the temperature abnormality cause determination device 1 through the network 4. The present invention can be applied to any device (for example, a router, a storage device, etc.) including Further, the air conditioner 2, the rack 3, the plurality of server devices 31 stored in the rack 3, and the network 4 shown in FIG. 1 are respectively the air conditioner 2, the rack 3, and the server device 31 described above with reference to FIG. This is the same as the network 4. The air conditioner 2 and the server device 31 constitute a data center, and the temperature abnormality cause location determination apparatus 1 determines the temperature abnormality cause location using the data center as a monitoring target. Further, F in the figure is intake air to the server device 31, and G is exhaust air from the server device 31. In addition, although the some air conditioner 2 is shown in FIG. 1, the one air conditioner 2 may be sufficient. When the air conditioner 2 sends out the cooling air, the cooling air is sent to the server device 31 in the rack 3 according to the flow path indicated by the arrow in the figure, and is exhausted from the server device 31. In addition, the server device 31 includes a temperature sensor that detects that the temperature of the intake air has risen to an abnormal value (temperature abnormality). The determination device 1 is notified.

  The temperature abnormality cause location determination apparatus 1 includes a temperature abnormality detection result acquisition unit 11, a determination unit 12, a notification unit 13, and a flow path information database (DB) 14. The temperature abnormality detection result acquisition unit 11 acquires a temperature abnormality detection result (temperature abnormality detection result) from the server device 31 through the network 4. Based on the flow path information stored in the flow path information DB 14 and the temperature abnormality detection result of the server device 31 acquired by the temperature abnormality detection result acquisition unit 11, the determination unit 12 causes the temperature abnormality in the cooling air flow path. Determine the location. The flow path information is information on the flow path of the cooling air supplied from the air conditioner 2 to the server group (a plurality of server apparatuses 31). In the present embodiment, a path (route) through which cooling air flows from the air conditioner 2 to the server device 31 is defined as a cooling air flow path. That is, the determination unit 12 determines which part of the cooling air flow path causes a temperature abnormality in the server device 31. The notification unit 13 displays, for example, the temperature abnormality cause location determined by the determination unit 12 and notifies the user of the temperature abnormality cause location determination apparatus 1. The flow path information DB 14 stores flow path information.

  In addition, the function of the temperature abnormality cause location determination apparatus 1 of this embodiment is implement | achieved by CPU and the program run on it. The program can be stored in a computer-readable recording medium, such as a semiconductor memory, a hard disk, a CD-ROM, a DVD, or the like, provided by being recorded in these recording media, or a network via a communication interface. Provided by sending and receiving using.

  Below, the outline | summary of the determination process of the temperature abnormality cause location by the temperature abnormality cause location determination apparatus 1 of this embodiment is demonstrated. FIG. 2 is a model diagram of a data center that is a target (monitoring target) of temperature abnormality cause location determination processing by the temperature abnormality cause location determination apparatus 1 of the present embodiment. In order to efficiently cool the server apparatus in the data center, the air conditioner does not simply cool the air in the vicinity but needs to circulate the air. In addition, each server device has an operating temperature condition, and the environmental temperature should not be outside the range of the operating temperature condition. Further, if the cooling capacity of the air conditioner is insufficient, the server device is not sufficiently cooled. Such factors should be considered when adding or changing layouts in the data center. Therefore, in this embodiment, it is assumed that the flow of the cooling air is designed in advance. Since the server device in the data center operates constantly, the flow of cooling air is steady, and the flow of cooling air can be expressed as a route with a direction as shown in the model diagram shown in FIG.

  The arrows in FIG. 2 indicate the direction of the cooling air flow. The model diagram shown in FIG. 2 is a two-dimensional diagram, but the model diagram of the data center that is the target of the temperature abnormality cause location determination process may be three-dimensional. The arrows in FIG. 2 indicate the direction in which the cooling air flows. That is, the cooling air from the air conditioner C1 is sent to each of the server devices A1, A2, and A3, and is discharged from the server devices A1, A2, and A3. In addition, the cooling air sent from the air conditioner C2 is sent to each of the server devices A4, A5, and A6 and discharged from the server devices A4, A5, and A6. The flow path information (flow path information) of the cooling air indicated by the model diagram is stored in advance in the flow path information DB 14.

  Here, for example, it is assumed that the server device A5 and the server device A6 have detected a temperature abnormality. As shown in FIG. 3, the server apparatus A5 and the server apparatus A6 notify the temperature abnormality cause location determination device 1 of the temperature abnormality detection result (see # 1 in FIG. 3), and the temperature of the temperature abnormality cause location determination device 1 The abnormality detection result acquisition unit 11 acquires the temperature abnormality detection result. The determination unit 12 of the temperature abnormality cause location determination device 1 extracts the cooling air flow path information shown in the model diagram shown in FIG. 2 from the flow path information DB 14, and the extracted flow path information and the acquired temperature abnormality Based on the extraction result, the cause of temperature abnormality in the cooling air flow path is determined as follows.

  First, in the route from the air conditioner 2 to a normal server device 31 (a server device 31 that has not detected a temperature abnormality), there should be no cause for temperature abnormality. Therefore, the determination unit 12 obtains a route (first route) from the air conditioner 2 to the normal server device 31 indicated by a thick line in FIG.

  Next, the temperature abnormality cause location should be in the route from the air conditioner 2 to the server device 31 that detected the temperature abnormality. Therefore, the determination unit 12 obtains a route (second route) from the air conditioner C2 to the server device 31 that detects the temperature abnormality, which is indicated by a thick line in FIG.

  The determination unit 12 determines a portion (a portion indicated by a thick line in FIG. 6) included in the obtained second route and not included in the first route as a temperature abnormality cause location. The determination unit 12 determines a portion that is included in the second route and is not included in the first route as a temperature abnormality cause location, thereby efficiently and automatically obtaining the temperature abnormality cause location in the data center to be monitored. be able to.

  When there are a plurality of server devices 31 in which the temperature abnormality is detected, the determination unit 12 calculates the logical product of the set of edges from the server device 31 in which each temperature abnormality is detected to the air conditioner 2. A route may be used. The determination unit 12 uses the logical product of the set of sides from the server device 31 where each temperature abnormality is detected to the air conditioner 2 as the second route, thereby causing the temperature abnormality in the data center to be monitored. The location can be obtained with high accuracy.

  Below, the example of the flow-path information memorize | stored in flow-path information DB14 is demonstrated. For example, the flow of cooling air from the air conditioners C1 and C2 to each server device shown in the model diagram of the data center as shown in FIG. 2 can be expressed as a directed graph as shown in FIG. . The arrows in the directed graph shown in FIG. 7 indicate the direction of the cooling air flow. The directed graph shown in FIG. 7 includes nodes (Cn, Nn, An) and a route (hereinafter referred to as an edge) connecting adjacent nodes. Specifically, C1 and C2 in FIG. 7 are a node indicating the air conditioner C1 and a node indicating C2 in FIG. 2, respectively. A1, A2, A3, A4, A5, and A6 are respectively a node indicating the server device A1, a node indicating the server device A2, a node indicating the server device A3, a node indicating the server device A4, and a server in FIG. A node indicating the device A5 and a node indicating the server device A6. N1 to N6 are intermediate nodes. The intermediate node is a node indicating a point at which the cooling air branches or joins. In the example of the directed graph shown in FIG. 7, N1 to N6 indicate intermediate nodes indicating points where the cooling air branches.

  That is, the information indicated by the directed graph in FIG. 7 includes a set of nodes and a set of edges as shown in FIG. The information shown in FIG. 8 is the flow path information of the cooling air shown in the model diagram shown in FIG. In FIG. 8, the side is expressed in the format of “start node → end node”. For example, C1 → N1 indicates a side between the node C1 and the node N1, and N1 → A1 indicates a side between the node N1 and the node A1. In the flow path information DB 14, for example, flow path information of cooling air as shown in FIG. 8 is stored in a program format as described below.

For example, a set (list) of nodes is stored in advance in a program format in the flow path information DB 14. As an example of a program representation of a node, a C language structure is shown below.
typedef struct node {
string * indentfier; // identifier C1, N1, etc.
string * location; // location information pillar A3 etc.} node _t;
In the example of the program representation of the node, “indentfier” is a node identifier that uniquely identifies the node, such as “C1” or “N1”. In addition to being unique, in the case of a server, the identifier is associated with actual machine information such as a host name and an IP address. “location” is node position information such as a column number, for example. Any expression can be used as the position information as long as the information can identify the position of the node.

The list of nodes can be expressed by an array as shown below, for example.
node_t nodes [];
That is, the list of nodes included in the flow path information stored in the flow path information DB 14 has, for example, a data structure as shown in FIG. In FIG. 9, id is a node identifier, and loc is node position information.

Since an edge can be expressed by a pair of a start point node and an end point node, for example, it can be expressed as follows in C language. src and dst mean the start point node and end point node of the side, respectively.
struct edge {
node_t src; // starting point
node_t dst; // end point};
In addition, the set of edges (list) is an array,
struct edge edges [];
Can be expressed as

  That is, the list of sides included in the channel information stored in the channel information DB 14 has a data structure as shown in FIG. 10, for example.

Each server device 31 has a detection result of a temperature abnormality by a temperature sensor provided in the own device. That is, each server device 31 has information indicating a status of temperature abnormality (NG) or normal (OK). Therefore, the information of the set (list) of server devices 31 can be expressed using the following program. Note that the temperature abnormality detection result acquisition unit 11 of the temperature abnormality cause location determination apparatus 1 is configured to obtain information on the status of the server device 31 (status) based on the temperature abnormality detection result acquired from the server device 31 periodically or at an arbitrary timing. Information) is set in the list information of the server device 31, and the list information of the server device 31 is stored in the flow path information DB.
struct server {
node_t node; // indicate which node
int status; // Status (OK or NG)
};
struct server servers [];
In the program expression example, status indicates a temperature abnormality detection result acquired from the server device 31. status is either OK or NG. OK indicates that a temperature abnormality is not detected in the server device 31, and NG indicates that a temperature abnormality is detected in the server device 31. Node is a node indicating the server device 31.

  That is, the list of server devices 31 included in the flow path information stored in the flow path information DB 14 has a data structure as shown in FIG. 11, for example. In FIG. 11, nd indicates a node indicating the server device 31, and sts indicates a status of the server device 31.

  For example, the determination unit 12 outputs information on the start point node and the end point node of the side where the temperature abnormality cause location exists as the temperature abnormality cause location. Accordingly, the output information on the location of the cause of temperature abnormality has a data structure as shown in FIG. 12, for example.

  FIG. 13 is a diagram illustrating an example of a determination process flow of a temperature abnormality cause location by the temperature abnormality cause location determination apparatus of the present embodiment. First, the temperature abnormality detection result acquisition part 11 of the temperature abnormality cause location determination apparatus 1 acquires the temperature abnormality detection result from each server device 31, for example, with temperature information (step S1). Specifically, the temperature abnormality detection result acquisition unit 11 acquires a temperature abnormality detection result transmitted from the server device 31 in the monitoring target data center periodically or when a temperature abnormality is detected.

  The determination unit 12 determines whether there is a temperature abnormality based on the acquired temperature abnormality detection result (step S2). Note that the determination unit 12 determines whether the temperature indicated by the temperature information exceeds a predetermined threshold based on the temperature information acquired from each server device 31, and if the temperature exceeds the threshold, the temperature abnormality You may make it judge that there exists.

  If the determination unit 12 determines that there is no temperature abnormality, the process ends. When the determination unit 12 determines that there is a temperature abnormality, the determination unit 12 waits until the temperature abnormality detection result at the same time is acquired from all the server devices 31 in the monitored data center (step S3). In step S <b> 3, the temperature abnormality detection result acquisition unit 11 further displays information regarding the status of each server device 31 in the information of the server device 31 based on the temperature abnormality detection result acquired from each server device 31. Is stored in the flow path information DB 14 as part of the flow path information. Next, the determination part 12 narrows down and determines the temperature abnormality cause location (suspected location) (step S4). For example, the determination unit 12 determines the side as the suspected place.

  The determination unit 12 determines whether a suspected place has been found (step S5). That is, the determination unit 12 determines whether the suspected part has been successfully narrowed down or failed. If the determination unit 12 determines that the suspected place has not been found, the process ends. When the determination unit 12 determines that a suspected place has been found, the notification unit 13 notifies the suspected place (step S6). The notification unit 13 displays, for example, information on a side that is a suspected place on a display, or notifies by e-mail or the like. The notification unit 13 may display the suspected place in a graphic format, or may display the position information (post number etc.) of the start point node and end point node of the side that is the suspected place.

  FIG. 14 is a diagram illustrating details of the suspected place narrowing-down process in step S4 of FIG. First, the determination unit 12 obtains a set S1 (first set) of sides from the server device 31 from which no temperature abnormality is detected to the air conditioner 2 (step S41). Next, the determination part 12 calculates | requires the set S2 (2nd set) of the edge | side from the server apparatus 31 from which the temperature abnormality was detected to the air conditioner 2 (step S42). Then, the side included in S2 and not included in S1 is determined as the suspected place (step S43).

  FIG. 15 is a diagram for explaining the details of the processing for obtaining the edge set S1 in step S41 of FIG. First, the determination unit 12 extracts the flow path information from the flow path information DB 14, and sequentially extracts the nodes indicating the server apparatuses 31 from the list of server apparatuses 31 included in the flow path information (see FIG. 11) ( Step S101). Next, the determination unit 12 determines whether the status of the extracted node is OK (step S102). If the determination unit 12 determines that the status of the extracted node is not OK, the process proceeds to step S105. If the determination unit 12 determines that the status of the extracted node is OK, the determination unit 12 determines all sides connected to the node (step S103), and sets the determined set of sides as A. The determination unit 12 performs an OR operation on the set A of edges obtained in step S103, and obtains an OR operation result as S1 (step S104). Specifically, the result of ORing S1 and A is S1. The determination unit 12 determines whether there is a next node in the list of the server device 31 (step S105). If the determination unit 12 determines that there is a next node in the list of the server device 31, the process returns to step S101. If the determination unit 12 determines that there is no next node in the list of the server device 31, the process ends.

  FIG. 16 is a diagram for explaining the details of the processing for obtaining the edge set S2 in step S42 of FIG. First, the determination unit 12 extracts the flow path information from the flow path information DB 14, and sequentially extracts the nodes indicating the server apparatuses 31 from the list of server apparatuses 31 included in the flow path information (see FIG. 11) ( Step S201). Next, the determination unit 12 determines whether the status of the extracted node is NG (step S202). If the determination unit 12 determines that the status of the extracted node is not NG, the process proceeds to step S205. If the determination unit 12 determines that the status of the extracted node is NG, the determination unit 12 determines all sides connected to the node, and sets the obtained set of sides as B (step S203). The determination unit 12 performs an AND operation on the edge set B obtained in step S203, and obtains an AND operation result as S2 (step S204). Specifically, the result of the logical product of S2 and B is S2. The sides included in the obtained S2 are sides corresponding to the common part of the route from each server device 31 whose status is NG to the air conditioner 2. The determination unit 12 determines whether there is a next node in the list of the server device 31 (step S205). If the determination unit 12 determines that there is a next node in the list of the server device 31, the process returns to step S201. If the determination unit 12 determines that there is no next node in the list of the server device 31, the process ends.

  FIG. 17 is a diagram for explaining the details of the processing for obtaining the edge set A in S103 of FIG. The process for obtaining the edge set B in S203 of FIG. 16 is performed according to the same procedure as the process for obtaining the edge set A shown in FIG. 17, and thus the description of the details of the process for obtaining the edge set B is omitted. .

  First, the determination unit 12 extracts the side e from the list of sides (see FIG. 10) included in the flow path information extracted in step S101 of FIG. 15 (step S301). The determination unit 12 determines whether the end point node of the side e is a server device (step S302). When the determination unit 12 determines that the end point node of the edge e is not a server device (the end point node is an intermediate node), the determination unit 12 searches and obtains an edge having the intermediate node as the start point node (step S303). ), The process returns to step S302. Specifically, the obtained edge is set as the edge e, and the process is performed in S302.

  When the determination unit 12 determines that the end node of the side e is a server device, the determination unit 12 determines whether the end node is a server device to be processed (step S304). The server device to be processed is the server device indicated by the node that is the target for obtaining the connected side in step S103 of FIG. If the determination unit 12 determines that the end node is not the server server to be processed, the process proceeds to step S306. When the determination unit 12 determines that the end node is a server device to be processed, the side e is added to the side set A (step S305). By repeating the processing in step S305, a set A of edges is obtained. Next, the determination unit 12 determines whether there is a next side in the side list (step S306). If the determination unit 12 determines that there is a next side in the side list, the process returns to step S301. If the determination unit 12 determines that there is no next side in the side list, the process ends.

  FIG. 18 is a diagram illustrating a first application example of the determination process of the temperature abnormality cause location by the temperature abnormality cause location determination apparatus of the present embodiment. In this example, as shown in FIG. 18, the data center to be monitored includes one air conditioner C1 and one server device A1. Further, the status of the server device A1 is NG. Since there is no side from the server device in which no temperature abnormality is detected to the air conditioner, the side set S1 obtained by the temperature abnormality cause location determination device 1 is an empty set (S1 = φ). Moreover, since the edge from the server apparatus in which the temperature abnormality is detected to the air conditioner is e1, the edge included in the edge set S2 obtained by the temperature abnormality cause location determination apparatus 1 is e1 (S2 = {E1}). The side included in S2 and not included in S1 is e1. Therefore, the temperature abnormality cause location determination apparatus 1 determines e1 as the temperature abnormality cause location.

  FIG. 19 is a diagram illustrating a second application example of the temperature abnormality cause location determination process performed by the temperature abnormality cause location determination apparatus according to the present embodiment. In this example, as shown in FIG. 19, the data center to be monitored includes one air conditioner C1 and two server apparatuses A1 and A2. Further, the status of the server apparatus A1 is NG, and the status of the server apparatus A2 is OK. Since the sides from the server device where no temperature abnormality is detected to the air conditioner are e1 and e3, S1 = {e1, e3}. Further, since the sides from the server device where the temperature abnormality is detected to the air conditioner are e1 and e2, S2 = {e1, e2}. The side included in S2 and not included in S1 is e2. Therefore, the temperature abnormality cause location determination apparatus 1 determines e2 as the temperature abnormality cause location.

  FIG. 20 is a diagram for explaining a third application example of the temperature abnormality cause location determination process performed by the temperature abnormality cause location determination apparatus according to the present embodiment. In this example, as shown in FIG. 20, the data center to be monitored includes two air conditioners C1 and C2 and two server apparatuses A1 and A2. Further, the status of the server apparatus A1 is NG, and the status of the server apparatus A2 is OK. Since the sides from the server device where no temperature abnormality is detected to the air conditioner are e1, e2, e3, e5, S1 = {e1, e2, e3, e5}. Further, since the sides from the server device where the temperature abnormality is detected to the air conditioner are e1, e2, e3, e4, S2 = {e1, e2, e3, e4}. The side included in S2 and not included in S1 is e4. Therefore, the temperature abnormality cause location determination apparatus 1 determines e4 as the temperature abnormality cause location.

  FIG. 21 is a diagram illustrating a fourth application example of the temperature abnormality cause location determination process performed by the temperature abnormality cause location determination apparatus according to the present embodiment. In this example, since there is no side from the server apparatus in which no temperature abnormality is detected to the air conditioner, S1 = φ. A set of sides from the server device A1 where the temperature abnormality is detected to the air conditioner is {e1, e2, e3, e4}. A set of sides from the server device A2 where the temperature abnormality is detected to the air conditioner is {e1, e2, e3, e5}. Therefore, the edge set S2 is obtained as a result of a logical product of the edge set {e1, e2, e3, e4} and the edge set {e1, e2, e3, e5}, and {e1, e2, e3} Become. The sides included in S2 and not included in S1 are e1, e2, and e3. Therefore, the temperature abnormality cause location determination apparatus 1 determines e1, e2, and e3 as temperature abnormality cause locations.

It is a figure which shows the structural example of the temperature abnormality cause location determination apparatus of this embodiment. It is a model figure of the data center made into the object of temperature abnormality cause location determination processing. It is a figure explaining the outline | summary of the determination process of a temperature abnormality cause location. It is a figure explaining the outline | summary of the determination process of a temperature abnormality cause location. It is a figure explaining the outline | summary of the determination process of a temperature abnormality cause location. It is a figure explaining the outline | summary of the determination process of a temperature abnormality cause location. It is a figure which shows the directed graph expressing the flow of the cooling air from an air conditioner to a server apparatus. It is an example of the information which a directed graph shows. It is a figure which shows the list | wrist of the node contained in flow path information. It is a figure which shows the list | wrist of the edge contained in flow path information. It is a figure which shows the list | wrist of the server apparatus contained in flow path information. It is a figure which shows the example of a data structure of the information of a temperature abnormality cause location. It is a figure which shows an example of the determination processing flow of a temperature abnormality cause location. It is a figure explaining the detail of the narrowing-down process of a suspected place. It is a figure explaining the detail of the process which calculates | requires edge set S1. It is a figure explaining the detail of the process which calculates | requires edge set S2. It is a figure explaining the detail of the process which calculates | requires edge set A. FIG. It is a figure explaining the 1st application example of the determination process of a temperature abnormality cause location. It is a figure explaining the 2nd application example of the determination process of a temperature abnormality cause location. It is a figure explaining the 3rd application example of the determination process of a temperature abnormality cause location. It is a figure explaining the 4th example of application of judgment processing of a temperature abnormal cause part. It is a figure which shows the example of the system which cools a server apparatus in a data center.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Temperature abnormality cause location determination apparatus 2 Air conditioner 3 Rack 4 Network 10 Monitoring apparatus 11 Temperature abnormality detection result acquisition part 12 Judgment part 13 Notification part 14 Flow path information DB
31 Server device

Claims (6)

A temperature abnormality cause location determination device that determines a temperature abnormality cause location in a flow path of cooling air supplied from one or a plurality of cooling sources to a device group composed of a plurality of devices,
A temperature abnormality detection result acquisition means for acquiring a temperature abnormality detection result of the device;
And a determination unit that determines a cause of temperature abnormality in the flow path of the cooling air based on flow path information of the cooling air supplied to the device group and a detection result of the temperature abnormality of the device. Temperature abnormality cause location determination device. The determination means reaches the cooling source from a device in which no temperature abnormality is detected, with the cooling source, the point where the cooling air branches or joins, and the route connecting the adjacent nodes as a side. Is obtained as a first set, and a set of sides from the device in which the temperature abnormality is detected to the cooling source is obtained as a second set, and is included in the second set, and The temperature abnormality cause location determination apparatus according to claim 1, wherein a side that is not included in the set of 1 is determined as a temperature abnormality cause location in the flow path of the cooling air. When there are a plurality of devices in which a temperature abnormality is detected, the determination means uses the logical product of a set of edges from the device in which each temperature abnormality is detected to the cooling source as the second set. The temperature abnormality cause location determination apparatus according to claim 2. A temperature abnormality cause location determination method for determining a temperature abnormality cause location in a flow path of cooling air supplied from one or more cooling sources to a device group consisting of a plurality of devices,
Obtain the temperature abnormality detection result of the device,
Based on flow path information of cooling air supplied to the device group and a detection result of temperature abnormality of the device, a temperature abnormality cause location in the flow path of the cooling air is determined. Cause location determination method. The cooling source, the point at which cooling air branches or joins, the set of sides from the device in which no temperature abnormality is detected to the cooling source, with the device as a node and the route connecting adjacent nodes as the side Is obtained as a first set, and a set of sides from the device in which the temperature abnormality is detected to the cooling source is obtained as a second set, and is included in the second set and included in the first set. The temperature abnormality cause location determination method according to claim 4, wherein a side that is not present is determined as a temperature abnormality cause location in the cooling air flow path. When there are a plurality of devices in which a temperature abnormality is detected, a logical product of a set of edges from the device in which each temperature abnormality is detected to the cooling source is defined as the second set. Item 6. The method for determining the cause of temperature abnormality according to Item 5.

Images