Computing power network monitoring method and computing power network monitoring apparatus

Abstract

The present disclosure relates to the technical field of wireless communications, and provides a computing power network monitoring method and a computing power network monitoring apparatus. The computing power network monitoring method comprises: a computing power network operation administration and maintenance (OAM) module sends a network monitoring requirement to any two computing power resource pools, and starts monitoring a bidirectional network performance parameter between the computing power resource pools, monitored network performance parameters comprising network latency, network packet loss rate, and network jitter. By means of the technical solution of the present disclosure, the network monitoring requirement is sent by means of the OAM module, so as to start monitoring a bidirectional network performance parameter between computing power resource pools. The detection operation does not require sending a large number of test messages between network devices to test and sense network states of devices at two ends, and lossless monitoring is facilitated.

Description

Computing power network monitoring method and computing power network monitoring device

[0001] This disclosure claims priority to Chinese patent application number 202311254492.2, filed on September 26, 2023, entitled “Computing Power Network Monitoring Method and Computing Power Network Monitoring Device,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present disclosure relates to the field of wireless communication technology, and in particular to a computing power network monitoring method and a computing power network monitoring device. Background Art

[0003] Computing network technology refers to the distribution of computing power, storage, algorithms, and other resource information from service nodes through a network control plane. This information, combined with network information and centered around user needs, provides optimal distribution, association, transaction, and allocation of computing, storage, and network resources, thereby achieving optimal allocation and utilization of resources across the entire network. The computing network manages and controls various computing resource nodes, sensing their resource status and scheduling computing applications between them. To manage and control these nodes, computing networks typically build and operate an Operations Administration and Maintenance (OAM) system. This OAM system monitors not only the computing resource usage of computing resource nodes but also the network status of these nodes. Therefore, computing applications must transmit these monitoring results to clients or other computing resource nodes via the network. Current active network monitoring solutions require sending large numbers of test messages between network devices to detect the network status of both devices. This large number of messages can disrupt the devices' normal service processing capabilities, resulting in poor monitoring.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute prior art known to ordinary technicians in the field.

[0005] Summary of the Invention

[0006] The purpose of the present disclosure is to provide a computing power network monitoring method, apparatus, network equipment and storage medium, which at least to some extent overcome the problems in the related art.

[0007] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by practice of the present disclosure.

[0008] According to one aspect of the present disclosure, a computing power network monitoring method is provided, including: a computing power network operation, maintenance and management (OAM) module sends a network monitoring request to any two computing power resource pools, and starts monitoring of two-way network performance parameters between the computing power resource pools. The measured network performance parameters include network delay, network packet loss rate and network jitter.

[0009] In one embodiment, it also includes: after the computing power resource pool accepts the network monitoring request issued by the OAM module, it starts network measurement between itself and the designated computing power resource pool; the computing resource pool inserts a sequence number and a sending timestamp into the packet header of a service data packet whose target address is the designated computing power resource pool, and upgrades the service data packet to a measurement data packet; and sends the measurement data packet to the designated computing power resource pool to avoid affecting normal network services due to the measurement of the network performance parameters.

[0010] In one embodiment, the method further includes: after the designated computing resource pool receives the measurement data packet, recording the serial number and receiving timestamp of the measurement data packet.

[0011] In one embodiment, it also includes: after completing the network parameter measurement, the OAM module collects the serial number and timestamp of the measurement data packet recorded by the two computing power resource pools, and calculates the network performance parameters between the two computing power resource pools, and the network performance parameters include the network delay, the packet loss rate and the network jitter.

[0012] According to another aspect of the present disclosure, a computing power network monitoring device is provided, which is applied to a computing power resource pool and includes: a network communication interface for receiving network monitoring requirements sent by an OAM module; a computing power network monitoring function processing unit, which is connected to the network communication interface and is used to initiate two-way network performance parameter monitoring between a specified computing power resource pool based on the network monitoring request, wherein the monitored network performance parameters include network delay, network packet loss rate and network jitter.

[0013] In one embodiment, the computing power network monitoring function processing unit is specifically used to: start network measurement between the specified computing power resource pool based on the network monitoring requirements, and generate a business data packet with the target address being the specified computing power resource pool; the computing power network monitoring device also includes: a serial number generator, connected to the computing power network monitoring function processing unit, for generating a serial number for the business data packet; a business data packet routing / forwarding unit, connected to the computing power network monitoring function processing unit, for transmitting a business data packet with the target address being the specified computing power resource pool to the computing power network monitoring function processing unit; a network monitoring business data packet processing engine, connected to the computing power network monitoring function processing unit, for inserting the serial number and sending timestamp into the packet header of the business data packet, upgrading the business data packet to a measurement data packet, and transmitting the measurement data packet to the computing power network monitoring function processing unit; the network communication interface is also used to: send the measurement data packet to the specified computing power resource pool.

[0014] In one embodiment, the computing power network monitoring function processing unit is further configured to: when the measurement data packet is received as the designated computing power resource pool, record the serial number and reception timestamp of the measurement data packet.

[0015] In one embodiment, it further includes: a network monitoring database connected to the computing power network monitoring function processing unit, and used to store network parameters obtained based on the measurement data packet.

[0016] In one embodiment, the system further comprises: a task token storage for storing the task token carried by the network monitoring request, wherein the task token is used to verify the validity of the measurement data packet.

[0017] According to another aspect of the present disclosure, a computing power network monitoring device is provided, which is applied to the OAM module and includes: a monitoring module for sending network monitoring requirements to any two computing power resource pools, and starting the monitoring of two-way network performance parameters between the computing power resource pools, wherein the monitored network performance parameters include network delay, network packet loss rate and network jitter.

[0018] In one embodiment, the monitoring module is also used to: collect the serial numbers and corresponding sending timestamps and receiving timestamps of the measurement data packets recorded by any two computing power resource pools; the computing power network monitoring device also includes: a calculation module, which is used to calculate the network performance parameters between any two computing power resource pools based on the serial numbers and the corresponding sending timestamps and receiving timestamps, and the network performance parameters include the network delay, the packet loss rate and the network jitter.

[0019] According to another aspect of the present disclosure, a network device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; the processor is configured to execute the computing power network monitoring method described in the first aspect above by executing the executable instructions.

[0020] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned computing power network monitoring method is implemented.

[0021] The computing power network monitoring solution provided by the embodiments of the present disclosure sends network monitoring requirements through the OAM module to initiate bidirectional network performance parameter monitoring between computing power resource pools. This detection operation does not require sending a large number of test messages between network devices to test the network status of the devices at both ends, which is conducive to achieving lossless monitoring.

[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0024] FIG1 shows a schematic diagram of a computing power network monitoring system according to an embodiment of the present disclosure;

[0025] FIG2 shows a flow chart of a computing power network monitoring method according to an embodiment of the present disclosure;

[0026] FIG3 shows a flow chart of another computing power network monitoring method according to an embodiment of the present disclosure;

[0027] FIG4 shows a flow chart of another method for monitoring computing power network according to an embodiment of the present disclosure;

[0028] FIG5 shows a flow chart of another method for monitoring computing power network according to an embodiment of the present disclosure;

[0029] FIG6 shows a flow chart of another method for monitoring computing power network according to an embodiment of the present disclosure;

[0030] FIG7 shows a schematic diagram of a computing power network monitoring device according to an embodiment of the present disclosure;

[0031] FIG8A is a schematic flow chart showing the first part of another computing power network monitoring method according to an embodiment of the present disclosure;

[0032] FIG8B is a schematic flow chart showing the second part of another computing power network monitoring method according to an embodiment of the present disclosure;

[0033] FIG9 shows a schematic diagram of a computing power network monitoring device according to an embodiment of the present disclosure;

[0034] FIG10 shows a structural block diagram of a network device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0036] In addition, the accompanying drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0037] FIG1 shows a schematic diagram of a computing power network monitoring architecture in an embodiment of the present disclosure.

[0038] As shown in Figure 1, the computing power network monitoring architecture includes an OAM module 102, a first computing power node 104 and a second computing power node 106. The first computing power node includes a network device, namely a router / gateway 1042, and the second computing power node includes a network device, namely a router / gateway 1062, and a wide area transmission network 108.

[0039] Among them, the computing power network OAM module 102 in the orchestration management layer of the computing power network monitors the operation status and network status of each computing power resource node, and performs computing power scheduling based on the monitoring situation.

[0040] The computing resource nodes in the computing network, including the first computing node 104 and the second computing node 106, are distributed in different areas of the wide area transmission network 108. With the deployment of a large number of industry applications in computing resource nodes, especially the deployment and application of large AI models, there is a high-throughput data flow between computing nodes. In addition to the status of computing node resources, the network operation status between nodes is also an important factor affecting the efficiency of application data forwarding.

[0041] When there is high-throughput data flow between computing nodes, the OAM module of the computing network needs to monitor key network parameters between computing resource nodes, including latency, jitter, and packet loss rate.

[0042] The traditional active monitoring method is to use active monitoring tools such as Ping and Traceroute to send a large number of test traffic packets from the node's network device, such as the router / gateway 1042 shown in Figure 1, to the target node router / gateway 1062, that is, the designated computing power resource pool, so as to count the network delay, packet loss rate, jitter and other operating parameters between nodes. However, sending a large number of test packets is lossy monitoring, which will affect the operation of existing normal business. At the same time, these test packets may be ignored and discarded by the node's network equipment, resulting in inaccurate testing. Therefore, a lossless computing power network monitoring solution is needed.

[0043] Below, the various steps of the computing power network monitoring method in this example implementation will be described in more detail with reference to the accompanying drawings and examples.

[0044] FIG2 shows a flow chart of an interactive computing network monitoring method according to an embodiment of the present disclosure.

[0045] As shown in FIG2 , an interactive computing network monitoring method according to an embodiment of the present disclosure includes:

[0046] In step S202, the computing power network operation, maintenance and management (OAM) module sends a network monitoring request to any two computing power resource pools to start monitoring of bidirectional network performance parameters between the computing power resource pools. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

[0047] Among them, any two computing power resource pools include the following computing power resource pool and the designated computing power resource pool.

[0048] In addition, the computing power resource pool is also called the computing power node.

[0049] In this embodiment, a network monitoring request is sent through the OAM module to start bidirectional network performance parameter monitoring between computing resource pools. This detection operation does not require sending a large number of test messages between network devices to test the network status of the devices at both ends, which is conducive to achieving lossless monitoring.

[0050] For example, as shown in FIG3 , the OAM module, as a monitoring end for computing power monitoring, performs the following steps of a computing power monitoring method:

[0051] In step S302, the computing power network operation, maintenance and management (OAM) module sends a network monitoring request to the two computing power resource nodes. The network monitoring request is used to start bidirectional network performance parameter monitoring of the computing power resource nodes. In the process of monitoring the bidirectional network performance parameters, the two computing power resource nodes include a computing power resource pool and a designated computing power resource pool for mutually sending and receiving business data packets. The business data packets are upgraded to measurement data packets. The computing power resource pool records the serial number and sending timestamp of the measurement data packet, and the designated computing power resource pool records the serial number and receiving timestamp of the measurement data packet.

[0052] In this embodiment, the computing power resource pool can be understood as the sending end of the measurement data packet, and the designated computing power resource pool can be understood as the receiving end of the measurement data packet.

[0053] In addition, those skilled in the art will appreciate that the measurement data packet may also be sent from a designated computing resource pool to a computing resource pool.

[0054] Furthermore, the OAM module, as a monitoring end for computing power monitoring, executes the computing power monitoring method, further comprising:

[0055] In step S304, the OAM module sends a monitoring stop instruction to the two computing resource nodes to receive the monitoring data list sent by the two computing resource nodes. The monitoring data list includes the serial numbers and corresponding sending timestamps of all measurement data packets sent by the computing resource pool, and the serial numbers and corresponding receiving timestamps of all measurement data packets received by the specified computing resource pool.

[0056] In step S306, the OAM module calculates the network performance parameters when the computing power resource pool sends a service data packet to the designated computing power resource pool based on the monitoring data list.

[0057] Among them, network performance parameters include network delay, network packet loss rate and network jitter.

[0058] In one embodiment, the interactive computing power network monitoring method also includes: after the computing power resource pool accepts the network monitoring request issued by the OAM module, it initiates network measurement between itself and the designated computing power resource pool; the computing power resource pool inserts a sequence number and a sending timestamp into the packet header of the service data packet whose target address is the designated computing power resource pool, and upgrades the service data packet to a measurement data packet; and sends the measurement data packet to the designated computing power resource pool to avoid affecting normal network services due to the measurement of network performance parameters.

[0059] The IP header of the service data packet is modified, and the modification content includes:

[0060] (1) The first three bits of the TOS field in the packet header are set to "110", which is defined as internetwork control.

[0061] (2) Add three 32-bit contents to the optional area at the end of the original packet header. The first 32 bits are used to load the task token. A 32-bit random number sequence number is generated and placed in the second 32-bit field. In the third 32 bits, the first 8 bits are the TOS value of the original packet, and the last 24 bits are filled in according to the IP protocol.

[0062] (3) According to the new IP packet header, recalculate the header length, total length, and header checksum fields.

[0063] In this embodiment, since sending a large number of test packets is lossy monitoring, it will affect the operation of existing normal services. At the same time, these test packets may be ignored and discarded by the node's network equipment, which will cause deviations in network performance measurements. By inserting the serial number and sending timestamp for computing power monitoring into the packet header of the service data packet, the service data packet is optimized for the monitoring function, so that the service data packet is upgraded to a measurement data packet. In the process of daily business interaction, network performance monitoring is achieved without sending a large number of additional monitoring packets to affect the normal operation of the node network. Fast and accurate network operation status monitoring and statistics of computing power network resource nodes can be achieved, effectively preventing interference and impact on the operation of existing computing power node services. At the same time, the OAM component of the computing power network performs a global analysis of the detection data generated by each computing power node, and realizes the accurate calculation of network operation parameters under lossless working conditions.

[0064] Exemplarily, as shown in FIG4 , the computing power resource pool is recorded as a first computing power node, and serves as a transmitter of a measurement data packet. According to another embodiment of the present disclosure, a computing power network monitoring method is applied to the first computing power node, and the steps performed include:

[0065] Step S402: The first computing power node initiates bidirectional network performance parameter monitoring in response to a network monitoring request sent by the OAM module.

[0066] In step S404, the first computing power node monitors the service data packet to be sent to the second computing power node, upgrades the service data packet to a first measurement data packet, and records the sequence number and sending timestamp of the first measurement data packet.

[0067] Among them, the second computing power node is the above-mentioned designated computing power resource pool.

[0068] Among them, the computing power network monitoring function processing unit of the PE / gateway device of the first computing power node monitors the data packets sent by this node. When it is detected that the target IP address belongs to the second computing power node, the business data packet is upgraded to the first measurement data packet.

[0069] Step S406: The first computing power node sends the first measurement data packet to the second computing power node.

[0070] In step S408, the first computing power node sends a first monitoring data list to the OAM module in response to the monitoring stop instruction sent by the OAM module. The first monitoring data list includes the serial number and sending timestamp of the first measurement data packet, so that the OAM module calculates the network performance parameters when the first computing power node sends the service data packet to the second computing power node based on the first monitoring data list.

[0071] In this embodiment, the sequence number and sending timestamp of the measurement data packet are recorded, combined with the receiving timestamp of the measurement data packet, and fed back to the OAM module so that the OAM module can calculate parameters such as network delay, network packet loss rate and network jitter, which is conducive to ensuring the reliability of network performance parameter detection.

[0072] Exemplarily, the computing power network monitoring method executed by the first computing power node, that is, the first computing power node initiates bidirectional network performance parameter monitoring between itself and the second computing power node in response to the network monitoring requirement sent by the OAM module, also includes: the first computing power node obtains the task token sent together with the network monitoring requirement.

[0073] Exemplarily, the computing power network monitoring method executed by the second computing power node, that is, the second computing power node starts monitoring of bidirectional network performance parameters in response to the network monitoring requirement sent by the OAM module, also includes: the second computing power node obtains the task token sent together with the network monitoring requirement.

[0074] In this embodiment, the OAM module sends a task token, i.e., a task token, to the nodes participating in the monitoring, i.e., the first computing power node and the second computing power node. During the monitoring process, the designated computing power resource pool, i.e., the second computing power node, receives the measurement data packet and extracts the token in the data packet for comparison with the task token in the memory to verify whether it is a legitimate measurement data packet. The measurement data packet sent by this node also needs to extract the token value from the memory, thereby ensuring the legitimacy and reliability of the computing power network monitoring.

[0075] Exemplarily, the computing power network monitoring method executed by the first computing power node, that is, the first computing power node monitors the business data packet to be sent to the second computing power node, and upgrades the business data packet to a measurement data packet, including: the first computing power node modifies the IP packet header of the business data packet based on the task token to generate a marked measurement data packet.

[0076] Exemplarily, the computing power network monitoring method performed by the first computing power node, that is, the first computing power node modifies the IP header of the business data packet based on the task token to generate a marked measurement data packet, includes: the first computing power node configures the service type field of the IP header as the tag of the measurement data packet; and generates optional data of the IP header, the optional business data including the task token, serial number and original service type field loaded at the first position of the tail of the IP header, and the serial number is placed at the second position of the tail of the IP header.

[0077] Exemplarily, the computing power network monitoring method executed by the first computing power node, that is, the first computing power node responds to the monitoring stop instruction sent by the OAM module, and also includes: the first computing power node deletes the task token sent together with the network monitoring requirement.

[0078] Exemplarily, the computing power network monitoring method executed by the first computing power node, that is, before the first computing power node sends a first monitoring data list to the OAM module in response to a monitoring stop instruction sent by the OAM module, also includes: the first computing power node receives a second measurement data packet sent by the second computing power node; and extracts the serial number of the second measurement data packet and records the receiving timestamp.

[0079] In one embodiment, the interactive computing power network monitoring method further includes: after the designated computing power resource pool receives the measurement data packet, recording the serial number and receiving timestamp of the measurement data packet.

[0080] As shown in FIG5 , illustratively, the second computing power node, i.e., the designated resource pool, serves as the receiving end of the measurement data packet. According to another embodiment of the present disclosure, a computing power network monitoring method is applied to the second computing power node, and the steps performed include:

[0081] Step S502: The second computing power node starts monitoring of bidirectional network performance parameters in response to the network monitoring request sent by the OAM module.

[0082] In step S504, the second computing power node monitors the first measurement data packet sent by the first computing power node, extracts the sequence number of the first measurement data packet, records the receiving timestamp, and restores it to the corresponding service data packet.

[0083] In step S506, the second computing power node sends a second monitoring data list to the OAM module in response to the monitoring stop instruction sent by the OAM module. The second monitoring data list includes the serial number and sending timestamp of the second measurement data packet, so that the OAM module calculates the network performance parameters when the first computing power node sends the service data packet to the second computing power node based on the second monitoring data list.

[0084] Exemplarily, the computing power network monitoring method performed by the second computing power node, that is, the second computing power node monitors a first measurement data packet sent by the first computing power node, extracts a sequence number of the first measurement data packet, and records a receiving timestamp, including:

[0085] When the second computing power resource detects that the service type field of the IP packet header of the received service data packet is a detection mark, it is determined that the first measurement data packet is detected.

[0086] The second computing resource reads data at the first position of the tail of the IP packet header.

[0087] If the data at the first position is consistent with the acquired task token, it is confirmed that the first measurement data packet is a legitimate monitoring service data packet.

[0088] Exemplarily, the computing power network monitoring method executed by the second computing power node further includes: the second computing power resource extracts the serial number of the first measurement data packet from the second position of the tail of the IP packet header, and stores the serial number of the first measurement data packet based on the received sequence number.

[0089] Exemplarily, the computing power network monitoring method executed by the second computing power node, that is, after the second computing power node monitors the first measurement data packet sent by the first computing power node, also includes: the second computing power node restores the first network monitoring packet to the original business data packet.

[0090] Exemplarily, the computing power network monitoring method executed by the second computing power node, that is, the second computing power node restores the first network monitoring packet to the original business data packet, including: the second computing power node replaces the detection tag with the original service type field; and deletes the optional data of the IP packet header.

[0091] Restoring the first network monitoring packet to the original service data packet specifically includes:

[0092] (1) Read the first 8 bits of the third 32-bit content in the optional part of the IP packet header and replace the "TOS" field content of the packet header;

[0093] (2) Delete the content of the optional part of the data packet;

[0094] (3) Calculate and update the "header length" field of the packet header;

[0095] (4) Calculate and update the "total length" field of the packet header;

[0096] (5) Based on the above update operation, the checksum of the new packet header is calculated and the "header checksum" field is updated. At this time, the received data packet is rolled back to the original state when it was sent by the host of the first computing power node, and is sent to the corresponding host in the computing power resource node through the general data packet routing / forwarding unit.

[0097] Exemplarily, the computing power network monitoring method executed by the second computing power node, that is, the second computing power node responds to the monitoring stop instruction sent by the OAM module, and also includes: the second computing power node deletes the task token sent together with the network monitoring requirement.

[0098] Exemplarily, the computing power network monitoring method executed by the second computing power node, that is, before the second computing power node responds to the monitoring stop instruction sent by the OAM module and sends the second monitoring data list to the OAM module, also includes: the second computing power node records the serial number and sending timestamp of the second measurement data packet, and sends the second measurement data packet to the first computing power node.

[0099] In one embodiment, the interactive computing power network monitoring method further includes: after completing the network parameter measurement between any two computing power resource pools, the OAM module collects the serial numbers and corresponding timestamps and receiving timestamps of the measurement data packets recorded by any two computing power resource pools, and calculates the network performance parameters between any two computing power resource pools. The network performance parameters include network delay, packet loss rate and network jitter.

[0100] The network parameters include the serial number, the sending timestamp, and the receiving timestamp.

[0101] The corresponding timestamps include the measurement data packet sending timestamp and receiving timestamp.

[0102] As shown in FIG6 , an interactive computing network monitoring method according to another embodiment of the present disclosure includes:

[0103] In step S602, the computing power network OAM module issues instructions to the access network devices of the two computing power resource nodes, such as gateways or access routers, to start network operation status monitoring.

[0104] In step S604, after receiving the instruction, the network device of the first computing power node detects the sent service data packet. When a data packet sent to the second computing power node is detected, the service data packet is marked as a measurement data packet and the serial number and sending timestamp are recorded.

[0105] In step S606, the second computing power node receives the data packet sent by the first computing power node and detects that the packet is marked as a first measurement data packet, records the serial number and receiving timestamp, changes the data packet mark to a normal data packet, and then forwards it to the target host in the second computing power node.

[0106] In step S608, the network device of the second computing power node detects the data packet sent to the first computing power node, marks the packet as a second measurement data packet, and records the sequence number of the data packet and the sending timestamp.

[0107] In step S610, the network device of the first computing power node receives a data packet from the second computing power node and marks it as a second measurement data packet, records the serial number and receiving timestamp, changes the mark of the data packet to a business data packet, and forwards it to the target host in the first computing power node.

[0108] Step S612: After the monitoring time ends, the computing network OAM module issues a stop monitoring instruction to the network devices of the first computing node and the second computing node.

[0109] In step S614, the network access devices of the first computing power node and the second computing power node stop monitoring, and the network access device of the first computing power node sends all data packet sequence numbers and corresponding sending timestamp lists sent to the second computing power node, as well as the data packet sequence numbers and corresponding receiving timestamp lists received from the second node, to the OAM module.

[0110] In step S616, the network access device of the second computing power node sends to the OAM module the sequence numbers and corresponding sending timestamp lists of all data packets sent to the first computing power node, as well as the sequence numbers and corresponding receiving timestamp lists of data packets received from the first node.

[0111] In step S618, the OAM module receives the monitoring data list sent by the computing node network device, and then calculates the packet loss rate, delay and jitter of the data packet sent from the first computing node to the second computing node based on the detection data list.

[0112] Exemplarily, the OAM module calculates network performance parameters when the computing power resource pool sends a service data packet to the designated computing power resource pool based on the monitoring data list, including:

[0113] Network performance parameters include the network packet loss rate and the number of service data packets sent based on the sequence numbers of all measurement data packets sent.

[0114] The number of received service data packets is counted based on the sequence numbers of all received measurement data packets.

[0115] The network packet loss rate is calculated based on the number of service data packets sent and the number of service data packets received.

[0116] Exemplarily, the OAM module calculates network performance parameters when the computing power resource pool sends a service data packet to the designated computing power resource pool based on the monitoring data list, including:

[0117] Network performance parameters include network delay. The sending timestamp and receiving timestamp of each measurement data packet are counted based on the sequence number. The individual delay of each network monitoring packet is determined based on the sending timestamp and receiving timestamp of each measurement data packet. The maximum delay and minimum delay are extracted from all individual delays. The average delay is determined based on all individual delays and the number of service data packets received. The maximum delay, minimum delay and average delay are used as the network delay.

[0118] Exemplarily, the OAM module calculates the network performance parameters when the computing power resource pool sends service data packets to the specified computing power resource pool based on the monitoring data list, including: the network performance parameters include the maximum network jitter value, and the maximum network jitter value is calculated based on the maximum delay and the maximum delay.

[0119] Exemplarily, the OAM module sends a network monitoring request to two computing nodes, and also includes: the OAM module generates a task token and sends the task token along with the start monitoring command, wherein, in the computing resource pool, the task token is used to upgrade the service data packet to a legal measurement data packet.

[0120] Exemplarily, the OAM module calculates the network performance parameters when the computing power resource pool sends service data packets to the specified computing power resource pool based on the monitoring data list, including: the network performance parameters include the average network jitter parameter, and the delay difference between the single delay and the average delay of each measured data packet is counted; the average network jitter parameter is determined based on all delay differences and the number of service data packets received.

[0121] After receiving the monitoring data list sent by the computing node network device, the computing power network OAM module counts and compares the list of sent data packets sent by the first computing power node and the list of received data packets sent by the second computing power node. It can calculate the packet loss rate, delay, jitter and other parameters of the data packets sent by the first computing power node to the second computing power node. It counts and compares the list of sent data packets of the second computing power node and the list of received data packets of the first computing power node. It can calculate the packet loss rate, delay, jitter and other parameters of the data packets sent by the second computing power node to the first computing power node. The calculation process is as follows.

[0122] Assume that the number of data packets sent by the first computing power node is N1, and the number of data packets received by the second computing power node is N2. When N1>N2, packet loss occurs, and the packet loss rate R ls The calculation is shown in formula (1): R ls =(N1-N2) / N1×100% (1)

[0123] By counting the receiving timestamps of all data packets in the data list received by the second computing power node, and the sending timestamps of the corresponding serial number data packets in the data packets sent by the first computing power node, the transmission delay of all data packets sent to the second computing power node can be calculated. i is the sequence number of the data packet, and the maximum delay t is calculated. max , and the minimum delay t min The average delay t avg , needs to be calculated, as shown in formula (2):

[0124] For the network jitter parameter, the maximum network jitter value J can be calculated max , as shown in formula (3): J max =t max -t min (3)

[0125] Average network jitter parameter J avg The calculation is shown in formula (4):

[0126] By calculating the network packet loss rate R from the first computing power node to the second computing power node ls , maximum delay t max and average delay t avg , Maximum network jitter J max and average network jitter J avg Based on the same calculation method, the network performance parameters in the direction from the second computing power node to the first computing power node can be calculated.

[0127] As shown in FIG7 , a computing power network monitoring device 700 according to an embodiment of the present disclosure is applied to a computing power resource pool and includes: a network communication interface 702 for receiving a network monitoring request sent by an OAM module.

[0128] The computing power network monitoring function processing unit 704 is used to initiate bidirectional network performance parameter monitoring between the specified computing power resource pool based on the network monitoring request. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

[0129] Exemplarily, the computing power network monitoring function processing unit is the core module of computing power network monitoring, which is responsible for coordinating and managing other functional modules and completing the business process of active network monitoring function. The data packets entering the device are first processed by this module, and then forwarded to the general data packet routing / forwarding unit of this device. The measurement data packets sent by the device are also sent to this unit through the general unit, and are sent through the network communication interface 702 after monitoring function processing.

[0130] In one embodiment, the computing power network monitoring function processing unit 704 is specifically used to: initiate network measurement between the specified computing power resource pool based on network monitoring requirements, and generate a service data packet with the target address being the specified computing power resource pool.

[0131] The computing power network monitoring device 700 also includes: a serial number generator 706, which is connected to the computing power network monitoring function processing unit 704 and is used to generate a serial number for a service data packet.

[0132] For example, the measurement data packet sent by the sending computing power resource pool must be marked with a serial number to facilitate identification and record by the network equipment of the other node. The function of the serial number generator is to generate a 32-bit serial number under the instruction of the network measurement data packet processing engine. The sending computing power resource pool will also be recorded in the sent measurement data packet table.

[0133] The service data packet routing / forwarding unit 708 is connected to the computing power network monitoring function processing unit 704 and is used to transmit service data packets with the target address being the specified computing power resource pool to the computing power network monitoring function processing unit 704 .

[0134] The network monitoring service data packet processing engine 710 is connected to the computing power network monitoring function processing unit 704, and is used to insert a serial number and a sending timestamp into the packet header of the service data packet, upgrade the service data packet to a measurement data packet, and pass the measurement data packet to the computing power network monitoring function processing unit 704.

[0135] Exemplarily, the network measurement packet processing engine is used to process normal data packets to achieve data packet upgrades. Therefore, the received / sent detection data packets must be processed according to the monitoring method definition rules. The engine completes the data packet upgrade processing under the coordination of the computing power network monitoring function processing unit 704.

[0136] The network communication interface 702 is further used to send a measurement data packet to a designated computing resource pool.

[0137] In one embodiment, the computing power network monitoring function processing unit 704 is further configured to: when receiving a measurement data packet as a designated computing power resource pool, record the serial number and receiving timestamp of the measurement data packet.

[0138] In one embodiment, it further includes: a network monitoring database 712, which is connected to the computing power network monitoring function processing unit 704 and is used to store network parameters obtained based on the measurement data packets.

[0139] The network parameters include but are not limited to the sequence number, sending timestamp, and receiving timestamp of the measurement data packet.

[0140] Exemplarily, the network monitoring database stores active network monitoring data, including two data tables, namely the sending measurement data packet table and the receiving measurement data packet table. The two tables respectively record the serial number and sending timestamp of the detection data packet sent by this node, and the receiving measurement data packet table records the serial number and receiving timestamp of the detection data packet sent to this node by other computing nodes received by this node. The structures of the two tables are the same, as shown in Table 1.

[0141] Table 1

[0142] The data packet sequence number field is 32 bits, and the timestamp field is 64 bits, which is the length specified by the NTP protocol.

[0143] In one embodiment, the system further includes: a task token storage 714 for storing the task token required by the network monitoring, where the task token is used to verify the validity of the measurement data packet.

[0144] For example, when the computing power network OAM starts a network monitoring operation on the computing power node, it will simultaneously issue a task token to the nodes participating in the monitoring. When the computing power network monitoring function processing unit receives the instruction, it will store the task token in the task token memory. During the monitoring process, after receiving the measurement data packet, the token in the data packet must be extracted and compared with the task token in the task token memory to verify whether it is a legal measurement data packet. The measurement data packet sent by this node must also extract the token value from the memory.

[0145] In this embodiment, the system architecture of the network monitoring device disclosed herein can be implemented in existing computing power node network equipment by adding relevant software functions and database modules. There is no need to modify the system and hardware of the equipment, and non-destructive testing of computing power network performance is achieved without increasing hardware costs.

[0146] According to another embodiment of the present disclosure, a computing power network monitoring device is applied to an OAM module and includes: a monitoring module for sending network monitoring requirements to any two computing power resource pools, and starting two-way network performance parameter monitoring between the two computing power resource pools. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

[0147] In one embodiment, the monitoring module is also used to: collect the serial numbers and corresponding timestamps and receiving timestamps of the measurement data packets recorded by any two computing power resource pools as network parameters; the computing power network monitoring device also includes: a calculation module, which is used to calculate the network performance parameters between any two computing power resource pools based on the serial numbers and the corresponding sending timestamps and receiving timestamps, and the network performance parameters include network delay, packet loss rate and network jitter.

[0148] To perform accurate statistical monitoring, all resource nodes in the entire computing network are synchronized with the same time server (NTP Server). To further illustrate the computing network monitoring method of the present disclosure, in combination with the computing network monitoring device shown in FIG7 , as shown in FIG8A , a computing network monitoring method according to another embodiment of the present disclosure includes:

[0149] In step S802, the computing network OAM module plans to count the network operation status between the two computing nodes and generate a 32-bit random number as a task token.

[0150] Step S804: Issue a start network monitoring instruction to the computing network monitoring function processing unit of the PE device / gateway of the first computing power node and the second computing power node respectively, and the task Token is issued together with the instruction.

[0151] In step S806, the computing power network monitoring function processing unit of the PE / gateway device of the first computing power node receives the instruction, stores the Token in the instruction into the task Token memory, and creates a measurement data packet sending table in the network monitoring database.

[0152] In step S808, the computing power network monitoring function processing unit of the PE / gateway device of the first computing power node monitors the business data packet to be sent. When it is detected that the target IP address belongs to the second computing power node, the network measurement data packet processing engine is called for processing. The engine unit first modifies the IP header of the data packet.

[0153] The specific modification results are shown in Figure 9.

[0154] As shown in FIG9 , the content in bold with background filling is the newly added or modified content, wherein the last 5 bits of the TOS field of the packet header remain unchanged. After the modification is completed, the current timestamp is obtained and recorded in sequence in the sent measurement data packet table.

[0155] In step S810, the computing power network monitoring function processing unit of the PE / gateway device of the first computing power node obtains a measurement data packet after completing the above-mentioned IP packet header modification operation, and sends the measurement data packet through the network communication interface.

[0156] In step S812, after receiving the measurement data packet, the computing power network monitoring function processing unit of the PE / gateway of the second computing power node detects the TOS field in the IP packet header and identifies the packet as a network measurement data packet.

[0157] Specifically, the first three bits of the TOS field are detected to be "110", and the packet is identified as a network measurement data packet.

[0158] Step S814: read the task Token content and compare it with the task Token stored in the task Token memory. If they are consistent, it is a legal network measurement data packet. Otherwise, it is an illegal measurement data packet and is discarded.

[0159] Specifically, the first 32 bits of the optional part of the packet header are read, which is the task token content, and compared with the task token stored in the task token memory.

[0160] After step S814, as shown in FIG8B , the computing power network monitoring method according to another embodiment of the present disclosure further includes:

[0161] Step S816: When receiving a legitimate measurement data packet sent from the first computing power node for the first time, the computing power network monitoring function processing unit of the second computing power node PE / gateway device immediately generates a received measurement data packet table.

[0162] Step S818, read the second 32-bit content in the optional field of the IP packet header, which is the serial number of the measurement data packet, store the serial number of the packet, obtain and store the timestamp of the current device, and store the data packets received subsequently in ascending order of serial number.

[0163] Among them, the sequential storage can start from the sequence number "1".

[0164] In step S820, after the computing power network monitoring function processing unit of the PE / gateway device of the second computing power node completes the recording, it calls the network measurement data packet processing engine to process the IP packet header.

[0165] In step S822, when the second computing power node has a data packet to send to the second business data packet of the first computing power node, the PE / gateway device of the second computing power node performs upgrade processing to obtain a second measurement data packet, and the PE / gateway device of the first computing power node detects the second measurement data packet from the second computing power node and processes it.

[0166] Step S824: When the monitoring time ends, the computing network OAM module issues a stop monitoring instruction to the computing network monitoring function processing unit of the PE / gateway device of the first computing node and the second computing node, and the current task Token becomes invalid.

[0167] In step S826, the computing power network monitoring function processing unit instructs the task token memory to delete the recorded corresponding task token, and then sends the stored sent measurement data packet table and received measurement data packet table to the computing power network OAM module.

[0168] In step S828, the computing power network OAM module receives the detection data packet table sent by the computing power nodes participating in the computing power network monitoring, and calculates the bidirectional network performance parameters of the first computing power node and the second computing power node respectively.

[0169] The calculation process refers to the above formulas (1) to (4), including packet loss rate, maximum transmission delay, average transmission delay, maximum network jitter, average network jitter, etc.

[0170] In this embodiment, active monitoring of the computing network performance is achieved by setting up an OAM module and a computing power node. In the monitoring process summary, there is no need to send a large number of test packets. Therefore, when a large number of business data packets are transmitted back and forth between the first computing node and the second computing power node, accurate network performance parameters can be obtained. If network operation parameter statistics are to be obtained when there are relatively few business data packets, it is necessary to instruct the network equipment of the computing node to generate and send sufficient measurement data packets to record the statistical network parameters.

[0171] Specifically, based on normal business traffic, measurement data packets are obtained by optimizing the business data packets for monitoring functions, which prevents the problem that general active monitoring tools need to send a large number of additional monitoring packets and affect the normal operation of the node network. It can realize fast and accurate network operation status monitoring and statistics of computing power network resource nodes.

[0172] The following describes a network device 1000 according to an embodiment of the present invention with reference to FIG10 . This may be an OAM module or a computing resource pool. The network device 1000 shown in FIG10 is merely an example and should not limit the functionality or scope of use of the embodiments of the present invention.

[0173] As shown in Figure 10, network device 1000 is implemented as a general-purpose computing device. Components of network device 1000 may include, but are not limited to, the aforementioned at least one processing unit 1010, the aforementioned at least one storage unit 1020, and a bus 1030 connecting various system components (including storage unit 1020 and processing unit 1010).

[0174] The storage unit stores program code that can be executed by the processing unit 1010, causing the processing unit 1010 to perform the steps according to various exemplary embodiments of the present invention described in the "Exemplary Methods" section above. For example, the processing unit 1010 may perform the solution described in step S202 of FIG. 2 .

[0175] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 10201 and / or a cache memory unit 10202 , and may further include a read-only memory unit (ROM) 10203 .

[0176] The storage unit 1020 may also include a program / utility 10204 having a set (at least one) of program modules 10205, such program modules 10205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0177] Bus 1030 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0178] Network device 1000 may also communicate with one or more external devices 1070 (e.g., a keyboard, pointing device, Bluetooth device, etc.), one or more devices that enable a user to interact with network device 1000, and / or any device that enables network device 1000 to communicate with one or more other computing devices (e.g., a router, modem, etc.). Such communication may occur via input / output (I / O) interface 1050. Furthermore, network device 1000 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via network adapter 1060. As shown, network adapter 1060 communicates with other modules of network device 1000 via bus 1030. It should be understood that, although not shown, other hardware and / or software modules may be used in conjunction with network device 1000, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0179] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0180] In exemplary embodiments of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above-described methods of this specification is stored. In some possible implementations, various aspects of the present invention may also be implemented in the form of a program product, which includes program code. When the program product is executed on an electronic device, the program code is used to cause the electronic device to perform the steps according to various exemplary embodiments of the present invention described in the "Exemplary Methods" section of this specification.

[0181] According to an embodiment of the present invention, a program product for implementing the above-mentioned method can be a portable compact disc read-only memory (CD-ROM) and include program code, and can be run on an electronic device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0182] The program product may be implemented in any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0183] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0184] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0185] The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0186] It should be noted that although several modules or units of the device for action execution are mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0187] Furthermore, although the steps of the method of the present disclosure are described in a particular order in the accompanying drawings, this does not require or imply that the steps must be performed in this particular order, or that all steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0188] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0189] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the appended claims.

Claims

1. A computing power network monitoring method, comprising: The computing power network OAM module sends a network monitoring request to any two computing power resource pools, and starts the monitoring of bidirectional network performance parameters between the computing power resource pools. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

2. The computing power network monitoring method according to claim 1, wherein: Also includes: After the computing power resource pool accepts the network monitoring request issued by the OAM module, it starts the network monitoring between the computing power resource pool and the designated computing power resource pool; The computing resource pool inserts a sequence number and a sending timestamp into a packet header of a service data packet whose target address is the designated computing resource pool, and upgrades the service data packet to a measurement data packet.

3. The computing power network monitoring method according to claim 2, wherein: Also includes: After receiving the measurement data packet, the designated computing power resource pool records the serial number and receiving timestamp of the measurement data packet.

4. The computing power network monitoring method according to claim 3, wherein: Also includes: After completing the measurement of the sequence number and timestamp of the measurement data packet, the OAM module collects the sequence number and timestamp of the measurement data packet recorded by the two computing power resource pools, and calculates the network performance parameters between the two computing power resource pools, wherein the network performance parameters include the network delay, the packet loss rate and the network jitter.

5. A computing power network monitoring device, applied to a computing power resource pool, comprising: Network communication interface, used to receive network monitoring requirements sent by the OAM module; The computing power network monitoring function processing unit is connected to the network communication interface and is used to initiate the monitoring of two-way network performance parameters between the specified computing power resource pool based on the network monitoring request. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

6. The computing power network monitoring device according to claim 5, wherein: The computing power network monitoring function processing unit is specifically used to: start network measurement with the specified computing power resource pool based on the network monitoring requirement, and generate a service data packet with the target address being the specified computing power resource pool; The computing power network monitoring device also includes: A sequence number generator, connected to the computing power network monitoring function processing unit, for generating a sequence number for the service data packet; A service data packet routing / forwarding unit, connected to the computing power network monitoring function processing unit, and configured to transmit the service data packet whose destination address is the designated computing power resource pool to the computing power network monitoring function processing unit; A network monitoring service data packet processing engine, connected to the computing power network monitoring function processing unit, for inserting the sequence number and the sending timestamp into the header of the service data packet, upgrading the service data packet to a measurement data packet, and transmitting the measurement data packet to the computing power network monitoring function processing unit; The network communication interface is also used to: send the measurement data packet to the designated computing power resource pool.

7. The computing power network monitoring device according to claim 6, wherein: The computing power network monitoring function processing unit is also used to: when receiving the measurement data packet as the designated computing power resource pool, record the serial number and receiving timestamp of the measurement data packet.

8. The computing power network monitoring device according to claim 7, wherein: Also includes: A network monitoring database is connected to the computing power network monitoring function processing unit and is used to store network parameters obtained based on the measurement data packet.

9. The computing power network monitoring device according to any one of claims 6 to 8, wherein: Also includes: The task token storage is connected to the computing power network monitoring function processing unit and is used to store the task token required to be carried by the network monitoring, and the task token is used to verify the legitimacy of the measurement data packet.

10. A computing power network monitoring device, applied to an OAM module, comprising: The monitoring module is used to send a network monitoring request to any two computing resource pools, and start the monitoring of bidirectional network performance parameters between the two computing resource pools. The monitored network performance parameters include network delay, network packet loss rate and network jitter.

11. The computing power network monitoring device according to claim 10, wherein: The monitoring module is also used to: collect the serial numbers of the measurement data packets recorded by the two computing resource pools and the corresponding sending timestamps and receiving timestamps; The computing power network monitoring device also includes: A calculation module is used to calculate the network performance parameters between any two computing resource pools based on the serial number and the corresponding sending timestamp and receiving timestamp, and the network performance parameters include the network delay, the packet loss rate and the network jitter.

Images