A cloud service messaging monitoring system for high performance computing and methods of use

Abstract

The application discloses a cloud service message transmission monitoring system for high-performance computing and a use method thereof, and relates to the technical field of data transmission monitoring. The system comprises a security block, an arrangement block, a trusted block and a receiving block. The security block is an intranet where a cloud service message data source is located. The security block comprises a management module and a processing module connected with the management module. The trusted block is a monitoring system server deployment area and is used for monitoring transmitted cloud service messages. The trusted block comprises a monitoring module. The use method comprises the following steps: a first monitoring unit performs integrity monitoring on each piece of split sub-information, and if the sub-information data is complete, the sub-information is continuously processed; if the sub-information data is incomplete, the first monitoring unit encodes the sub-information and sends the encoded sub-information to an operation module through the monitoring module. The monitoring module greatly improves the efficiency of data information monitoring, and realizes high-speed and safe cloud service message transmission.

Description

TECHNICAL FIELD

[0001] The application relates to a cloud service message transmission monitoring system for high-performance computing and a use method thereof, and belongs to the technical field of data transmission. BACKGROUND

[0002] High-performance computing refers to using gathered computing power to process large-scale scientific and engineering computing tasks, including simulation, modeling and rendering, which cannot be completed by a standard workstation. A device for realizing high-performance computing is called a high-performance computer, which has the characteristics of fast operation speed, large storage capacity and high reliability. In a general sense, high-performance computing, parallel computing and supercomputing are synonymous. As an important means of scientific and technological innovation, high-performance computing is a strategic commanding point of contemporary scientific and technological competition, and it reflects the comprehensive strength of a country. It plays an important strategic role in national defense security, high-tech development and national economic construction. Cloud service messages need to use high-performance computing to support their operation. Cloud message services should first realize the high availability and scalability of message middleware to ensure uninterrupted service capability for a large number of users and dynamically adjust resources according to load changes. Secondly, it provides sequence guarantee for message transmission under the premise of guaranteeing the former.

[0003] At present, in order to ensure the safety and stability of cloud service message transmission, the cloud service message needs to be monitored in the whole transmission process to avoid the situation that the data is intercepted or lost during data transmission. However, the general cloud service message transmission monitoring system usually monitors each kind of data, and it is not convenient to monitor the integrity and safety of a part of large-capacity data information, which leads to slow monitoring efficiency.

[0004] The existing cloud service message transmission monitoring system has low monitoring efficiency for relatively complex and large data information, and cannot realize high-speed and high-security cloud service message transmission. Therefore, a cloud service message transmission monitoring system for high-performance computing and a monitoring method thereof are proposed. SUMMARY

[0005] The technical problem to be solved by the application is to provide a cloud service message transmission monitoring system for high-performance computing and a use method thereof, which solves the problem of low monitoring efficiency of the existing cloud service message transmission monitoring system for relatively complex and large data information and the problem of inability to realize high-speed and high-security cloud service message transmission.

[0006] The technical scheme adopted by the application is as follows: a cloud service message transmission monitoring system for high-performance computing, comprising a security block, an arrangement block, a trusted block and a receiving block, the security block is an internal network where a cloud service message data source is located, the security block comprises a management module and a processing module connected with the management module; the arrangement block is a data arrangement and distribution area of the monitoring system, the arrangement block comprises a distribution module; the trusted block is a server deployment area of the monitoring system and is used for monitoring transmitted cloud service messages, the trusted block comprises a monitoring module; and the receiving block is a cloud service message receiving area.

[0007] The processing module comprises a plurality of high-performance computing nodes connected with each other and used for executing cloud service message transmission monitoring tasks, is used for processing cloud service messages to be transmitted, each high-performance computing node comprises a data splitting unit, a marking unit and an encryption unit, the data splitting unit is used for splitting each cloud service message to be transmitted into sub-information, the split sub-information is marked by the marking unit one by one to form corresponding information codes, and the encryption unit is used for generating a combined key according to the information code corresponding to each sub-information and a preset cloud service message classification code according to a preset key coding rule, and each sub-information is encrypted by the combined key to generate a corresponding number of ciphertexts.

[0008] The management module is used for dynamically distributing and managing the plurality of high-performance computing nodes, integrating the ciphertexts generated by each high-performance node into a data packet, and sending a plurality of data packets to the distribution module.

[0009] The distribution module is used for recombining and distributing the plurality of data packets, the number of recombined data ciphertexts in each group is consistent, and the plurality of groups of recombined data are re-packed and transmitted after distribution.

[0010] The monitoring module comprises a first monitoring unit, a second monitoring unit and a third monitoring unit, a plurality of first monitoring units are arranged in the security block, each high-performance computing node corresponds to a group of first monitoring units, the first monitoring unit is used for monitoring the integrity of each cloud service message sub-information after splitting, and the integrity of the sub-information data after splitting is ensured; the second monitoring unit is arranged in the arrangement block and is used for monitoring the number of ciphertexts of each re-packed data packet.

[0011] Preferably, the receiving block comprises a receiving module.

[0012] The receiving module is used for receiving the plurality of data packets transmitted by the distribution module, obtaining the sub-information ciphertexts corresponding to the plurality of data packets according to the cloud service message classification code of each cloud service message, decoding the ciphertexts according to the combined key, integrating the decoded sub-information according to the information code sequence, and forming a complete cloud service message.

[0013] Preferably, the above-mentioned security blocks, arrangement blocks and receiving blocks are isolated by firewalls, and information transmission between the blocks needs to pass through the trusted block monitoring to ensure that there is no problem before information transmission, and the trusted block has the right to directly monitor through the firewall of each block.

[0014] Preferably, the above-mentioned trusted block further comprises an operation module, and the operation module and the monitoring module are in communication connection;

[0015] The operation module is used for receiving and storing the monitoring results of each monitoring unit output by the monitoring module, and the monitoring personnel accesses the monitoring system through the terminal, judges a sub-information existing problem in the cloud service message transmission process through the monitoring results of each monitoring unit stored in the operation module, and replaces the corresponding sub-information with the problem according to the information coding of the sub-information using the operation module.

[0016] Preferably, the third monitoring unit is arranged in the receiving block, and is used for monitoring the information coding sequence of the re-integrated cloud service information, so as to ensure that the sub-information of each cloud service information is integrated in coding sequence during the integration process.

[0017] A use method of a cloud service message transmission monitoring system for high-performance computing, comprising the following steps:

[0018] S1, the management module obtains the cloud service message to be transmitted, and numbers the cloud service messages according to the preset cloud service message classification coding table as A1, A2……A n ;

[0019] S2, according to the number of cloud service message numbers, a corresponding number of high-performance computing nodes B1, B2……B n are selected, each high-performance computing node B randomly selects a cloud service message A to be processed, divides the cloud service message into equal parts to obtain sub-information, and a marking unit encodes all sub-information of the cloud service message in sequence and forms corresponding information coding, at this time, a first monitoring unit monitors the integrity of each sub-information after the splitting, if the sub-information data is complete, it continues to process, if the sub-information data is not complete, the first monitoring unit sends the sub-information coding to the operation module through the monitoring module, the monitoring personnel replaces the complete sub-information data according to the sub-information coding, and an encryption unit generates a combination key according to the information coding corresponding to each sub-information and the preset cloud service message classification coding according to the preset key coding rule, encrypts each sub-information through the combination key to generate a corresponding number of ciphertexts and sends them to the management module again;

[0020] S3, the management module receives each batch of ciphertexts, and integrates them into multiple groups of data packets and sends them to the distribution module, the distribution module decomposes and scrambles the multiple groups of data packets together, recombines a certain number of arbitrary ciphertexts to form new multiple groups of data packets, at this time, the second monitoring unit monitors the number of ciphertexts in each group of data packets, if the number of ciphertexts in each group of data packets is the same, each group of data packets is sent to the receiving module in turn according to the preset time interval, if the number is not the same, the second monitoring unit sends the ciphertext loss information to the operation module through the monitoring module, the monitoring personnel manually monitor the sub-information encoding of each cloud service message, and supplement the lost sub-information according to the sub-information encoding;

[0021] S4, the receiving module receives the multiple groups of data packets in turn and uniformly decomposes the data packets, obtains the corresponding sub-information ciphertexts in the multiple groups of data packets according to the classification encoding of each cloud service message, and decodes the ciphertexts according to the combination key, and integrates the decoded sub-information according to the information encoding order, at this time, the third monitoring unit monitors the sub-information encoding of each integrated cloud service message, and forms a complete cloud service message.

[0022] Preferably, in the step S4, the receiving module applies for the combination key of each cloud service message to the management module while receiving the multiple groups of data packets, the management module calls the combination key generated by each high-performance computing node and the classification encoding of the cloud service message, and uniformly sends them to the receiving module, and the receiving module decrypts the corresponding cloud service message sub-information ciphertext according to the classification encoding of the cloud service message and the corresponding combination key.

[0023] Preferably, in the step S4, the specific steps that the third monitoring unit monitors the sub-information encoding of each integrated cloud service message are as follows: the third monitoring unit monitors the information encoding order of each integrated cloud service message, if the order is accurate, a complete cloud service message is formed, if the order is not accurate, the cloud service message is packaged and sent to the operation module through the monitoring module, the information encoding is manually re-identified by the monitoring personnel, the cloud service message is integrated according to the information encoding order, and a complete cloud service message is formed.

[0024] Advantages of the present application: compared with the prior art, the present application has the following effects:

[0025] 1) The present application sets up each block, sets all modules in the system in each block, and sets up a firewall between each block. In this way, the independent architecture improves the reliability of the monitoring system and the safety of the whole system. When the monitoring system is operating, each block is a micro server. When a micro server fails or is attacked, it will not affect other servers, improving the reliability of the system. The trusted block with the monitoring module is connected to each block. In the trusted block, the monitoring module can directly access each block without condition, so that the monitoring module can monitor the cloud service message at a higher speed.

[0026] 2) The present application sets up a processing module and processes each cloud service message through a separate high-performance computing node. First, the data splitting unit splits the cloud service message into equal parts. The sub-information after splitting is marked by the marking unit in sequence to form information coding. Each sub-information is uniformly encrypted to form ciphertext, which ensures the security of each cloud service message. Each cloud service message is processed by a separate high-performance computing node, which can quickly and effectively process data. After the high-performance nodes process all cloud service messages, they are sent in time sequence after being rearranged as data packets. This ensures the absolute safety of cloud service messages during transportation and avoids interception of cloud service messages by the outside world.

[0027] 3) The present application sets up a monitoring module that is in communication with each block. The monitoring module is divided into a first monitoring unit, a second monitoring unit, and a third monitoring unit. The monitoring module subdivides the large data flow monitoring task in the cloud service message transmission process. The first monitoring unit monitors the integrity of each cloud service message sub-information after splitting. The second monitoring unit monitors the number of ciphertexts for each group of repackaged data packets. The third monitoring unit monitors the information coding sequence of the integrated cloud service information. This ensures that each cloud service information is integrated in coding sequence. The monitoring range is expanded to the whole process of cloud service message transmission. Therefore, the monitoring module can monitor the whole process of cloud service message transmission. Each monitoring unit has a different monitoring task for cloud service messages. According to the corresponding monitoring task, the problems that are likely to occur during cloud service message transmission are determined, and the problems are solved in real time by monitoring personnel. This avoids the inconvenience of solving subsequent problems. Therefore, the present application greatly improves the efficiency of data information monitoring and realizes high-speed and safe cloud service message transmission. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1This is a system block diagram of the high-performance computing cloud service message passing monitoring system of the present invention.

[0029] Figure 2 This is a flowchart illustrating the usage method of the high-performance computing cloud service message passing monitoring system of the present invention. Detailed Implementation

[0030] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0031] Example 1: As Figures 1-2 As shown, a high-performance computing cloud service message transmission monitoring system includes a security block, a processing block, a trusted block, and a receiving block. The security block is the intranet where the cloud service message data source is located, and includes a management module and a processing module that establishes a control connection with it. The processing block is the data processing and allocation area of ​​the monitoring system, and includes an allocation module. The trusted block is the server deployment area of ​​the monitoring system, used to monitor the transmitted cloud service messages, and includes a monitoring module.

[0032] Processing Module: Includes multiple interconnected high-performance computing nodes for performing cloud service message transmission monitoring jobs. It is used to process cloud service messages to be transmitted. Each high-performance computing node includes a data splitting unit, a tagging unit, and an encryption unit. The data splitting unit is used to split each cloud service message to be sent into sub-information. The split sub-information is tagged one by one by the tagging unit to form a corresponding information code. The encryption unit is used to generate a combined key according to the information code corresponding to each sub-information and the preset cloud service message classification code according to the preset key encoding rules. The combined key is used to encrypt each sub-information to generate a corresponding number of ciphertexts.

[0033] The data splitting unit in this invention effectively splits the messages of each cloud service using a data splitting algorithm. Simultaneously, the algorithm learns and records each split to ensure the stability of subsequent data splits. The data splitting algorithm is as follows:

[0034] Let θ be the percentage of tasks completed by the Mapper out of the total number of tasks, i.e.

[0035]

[0036] In the formula, Task finished `Task∑` and `Task∑` represent the number of completed tasks and the total number of tasks, respectively. The condition for transmission is that the values ​​of the two data fragments are greater than the average value of the fragments. The average value of the fragments is calculated as follows:

[0037]

[0038] In the formula, Saverage is the average of the shards, Num is the total number of Reducers, S i is the shard value.

[0039] After the completion of the Map phase, the master node calculates the data skew of all intermediate results, and recalculates the load average of the current phase according to formula (2), and the data exceeding the load average is split again, and the split formula is:

[0040]

[0041] In the formula: Hash(·) is a Hash function, mod(·) is a remainder function, is a rounding up function.

[0042] The management module is used for dynamically allocating and managing a plurality of high-performance computing nodes, integrating the same batch of ciphertext generated by each high-performance node into a data packet, and sending multiple groups of data packets to the distribution module.

[0043] The distribution module is used for recombining and distributing multiple groups of data packets, and the number of recombined data ciphertexts in each group is consistent, and the distributed several groups of recombined data are repackaged and transmitted.

[0044] The monitoring module includes a first monitoring unit, a second monitoring unit and a third monitoring unit, a plurality of first monitoring units are arranged in the security block, each high-performance computing node corresponds to a group of first monitoring units, the first monitoring unit is used for integrity monitoring of each split cloud service message sub-information, and the integrity of the split sub-information data is guaranteed; the second monitoring unit is arranged in the arrangement block and is used for monitoring the number of ciphertexts of each repackaged data packet.

[0045] Among them, the receiving block is a cloud service message receiving area, and the receiving block includes a receiving module.

[0046] The receiving module is used for receiving multiple groups of data packets transmitted by the distribution module, obtaining the corresponding sub-information ciphertext in the multiple groups of data packets according to the classification coding of each cloud service message, decoding the ciphertext according to the combined key, integrating the decoded sub-information according to the information coding order to form a complete cloud service message.

[0047] In the present application, the Leaders algorithm is used as the method of cloud service message integration. The Leaders algorithm is an incremental clustering algorithm, and its clustering result depends on the selection of the representative point leader. Specifically, the algorithm first takes a point from the data set as the starting leader, and then calculates the Euclidean distance between the remaining data points and the existing leader in turn: if it is less than or equal to a given threshold, the data point is classified into the same class as the current leader; if it is greater than the given threshold, the data point is taken as a new leader. Repeat the above steps until the clustering of the entire data set is completed, and the algorithm is as follows:

[0048] Input: data set X; distance threshold T

[0049] 1. Initialize the leader set L as an empty set, and the number of leaders l as 0;

[0050] 2. For each data point x in the data set X do

[0051] 3. Find a leader that satisfies: leader element-of L and leader-x <= T;

[0052] 4. IF (there is no leader that satisfies the condition or ) then

[0053] 5. L = L union {x}; l = l + 1;

[0054] 6. Else

[0055] 7. Classify the data point x into the category represented by the leader;

[0056] 8. End if

[0057] 9. End for

[0058] Output: leader set L; number of leaders l: clustering result

[0059] Among them, the security block, the arrangement block and the receiving block are isolated by setting firewalls between each block. During the information transmission process between each block, the information transmission needs to pass through the trusted block monitoring to ensure that there is no problem, and then the information transmission is carried out. The trusted block has the right to directly cross the firewall of each block to directly monitor.

[0060] Among them, the trusted block further includes an operation module, and a communication connection is established between the operation module and the monitoring module.

[0061] Operation Module: The operation module is used to receive and store the monitoring results of each monitoring unit output by the monitoring module. Monitoring personnel access the monitoring system through the terminal, use the monitoring results of each monitoring unit stored in the operation module to determine a sub-information with a problem in the cloud service message transmission process, and use the operation module to replace the corresponding sub-information with a problem according to the information encoding of the sub-information.

[0062] The third monitoring unit is located within the receiving block and is used to monitor the information encoding order of the re-integrated cloud service information to ensure that the sub-information of each cloud service information is integrated in the encoding order during the integration process.

[0063] Example 2: Figure 2 As shown, a method for using a high-performance computing cloud service message passing monitoring system includes the following steps:

[0064] S1. The management module obtains the cloud service messages to be transmitted and assigns them numbers A1, A2...A1 according to a preset cloud service message classification code table. n ;

[0065] S2. Select the corresponding number of high-performance computing nodes B1, B2...B2 based on the number of cloud service message numbers. n Each high-performance computing node B randomly selects a cloud service message A to be processed, and splits the cloud service message into equal parts to obtain sub-information. The marking unit encodes all the sub-information of the cloud service message in sequence to form corresponding information codes. At this time, the first monitoring unit performs integrity monitoring on each sub-information after splitting. If the sub-information data is complete, it continues to process it. If the sub-information data is incomplete, the first monitoring unit sends the sub-information code to the operation module through the monitoring module. The monitoring personnel replace the complete sub-information data according to the sub-information code. The encryption unit generates a combined key according to the information code corresponding to each sub-information and the preset cloud service message classification code according to the preset key encoding rules. The combined key is used to encrypt each sub-information to generate a corresponding number of ciphertexts and resends them to the management module.

[0066] S3. The management module receives each batch of ciphertext and integrates it into multiple data packets, which are then sent to the allocation module. The allocation module decomposes and shuffles the multiple data packets and recombines a certain number of arbitrary ciphertexts to form new data packets. At this time, the second monitoring unit monitors the number of ciphertexts in each data packet. If the number of ciphertexts in each data packet is the same, the data packets are sent to the receiving module in sequence at a preset time interval. If the number is different, the second monitoring unit sends the ciphertext loss information to the operation module through the monitoring module. The monitoring personnel manually monitor the sub-information codes of each cloud service message and supplement the lost sub-information according to the sub-information codes.

[0067] In the present application, each group of data packets mentioned in the above steps is set to send one data packet every 5s according to a preset time interval until all data packets are sent.

[0068] S4, the receiving module receives the multiple groups of data packets in sequence and uniformly decomposes the data packets, obtains the corresponding sub-information ciphertext inside the multiple groups of data packets according to the classification coding of each cloud service message, and decodes the ciphertext according to the combined key, integrates the decoded sub-information according to the information coding order, at this time the third monitoring unit monitors the sub-information coding of each integrated cloud service message, and forms a complete cloud service message.

[0069] In step S4, the receiving module applies the combined key of each cloud service message to the management module while receiving the multiple groups of data packets, the management module calls the combined key generated by each high-performance computing node and the classification coding of the cloud service message, and uniformly sends it to the receiving module, and the receiving module decrypts the corresponding cloud service message sub-information ciphertext according to the classification coding of the cloud service message and the corresponding combined key.

[0070] In step S4, the specific steps of the third monitoring unit monitoring each integrated cloud service message sub-information coding are as follows: the third monitoring unit monitors the information coding order of each cloud service information that is re-integrated and complete, if the order is accurate, a complete cloud service message is formed, if the order is not accurate, the cloud service message is packaged and sent to the operation module through the monitoring module, the information coding is manually re-identified by the monitoring personnel, the cloud service message is integrated according to the information coding order, and a complete cloud service message is formed.

[0071] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any skilled person in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application, therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A cloud service messaging monitoring system for high performance computing, characterized by: The security block is an internal network where the cloud service message data source is located, and the security block includes a management module and a processing module connected with the management module; The arrangement block is an arrangement area for monitoring system data, and the arrangement block includes an arrangement module; the trusted block is a monitoring system server deployment area, used for monitoring the transmitted cloud service messages, and the trusted block includes a monitoring module; and the receiving block is a cloud service message receiving area; The receiving block includes a receiving module; The receiving module is used for receiving multiple groups of data packets transmitted by the arrangement module, obtaining corresponding sub-information ciphertext in the multiple groups of data packets according to the classified coding of each cloud service message, decoding the ciphertext according to the combined key, and integrating the decoded sub-information according to the information coding sequence to form a complete cloud service message; The processing module includes multiple high-performance computing nodes connected with each other and used for performing cloud service message transmission monitoring, and each high-performance computing node includes a data splitting unit, a marking unit and an encryption unit, the data splitting unit is used for splitting each cloud service message to be sent into sub-information, the split sub-information is marked by the marking unit one by one to form corresponding information coding, and the encryption unit is used for generating a combined key according to the information coding corresponding to each sub-information and the preset cloud service message classified coding according to the preset key coding rule, and encrypting each sub-information by the combined key to generate corresponding number of ciphertexts; The management module is used for dynamically allocating and managing the multiple high-performance computing nodes, integrating the ciphertexts generated by each high-performance node into a data packet, and sending multiple groups of data packets to the arrangement module; The arrangement module is used for recombining and distributing the multiple groups of data packets to obtain recombined data ciphertexts, the number of the recombined data ciphertexts in each group is consistent, and the distributed several groups of recombined data are repackaged and transmitted; The monitoring module includes a first monitoring unit, a second monitoring unit and a third monitoring unit, multiple first monitoring units are arranged in the security block, each high-performance computing node corresponds to a group of first monitoring units, and the first monitoring unit is used for monitoring the integrity of each cloud service message sub-information after splitting; The second monitoring unit is arranged in the arrangement block and is used for monitoring the number of ciphertexts of each repackaged data packet, and the third monitoring unit is arranged in the receiving block and is used for monitoring the information coding sequence of the completely integrated cloud service information.

2. The cloud service messaging monitoring system for high performance computing of claim 1, wherein: Firewalls are arranged between the security block, the arrangement block and the receiving block to isolate them, and information transmission between the blocks needs to pass through the trusted block for monitoring before being transmitted, and the trusted block has the right to directly cross the firewall of each block to directly monitor.

3. The cloud service messaging monitoring system for high performance computing of claim 1, wherein: The trusted block further includes an operation module, and the operation module is connected with the monitoring module in communication. The operation module is used for receiving and storing the monitoring results of each monitoring unit output by the monitoring module, and a monitoring personnel accesses the monitoring system through a terminal, judges a certain sub-information existing problem in a cloud service message transmission process through the monitoring results of each monitoring unit stored in the operation module, and replaces the corresponding sub-information existing problem using the operation module according to the information coding of the sub-information.

4. A method of using a cloud service messaging monitoring system for high performance computing, the method comprising: The method comprises the following steps: S1, the management module obtains the cloud service message to be transmitted, and numbers the cloud service messages according to a preset cloud service message classification coding table 、 …… ; S2, selecting a corresponding number of high-performance computing nodes according to the number of cloud service message numbers , … each high-performance computing node randomly selecting a cloud service message to be processed , the cloud service message is equally divided to obtain sub-information, and the marking unit encodes and processes all sub-information of the cloud service message in order and forms corresponding information codes. At this time, the first monitoring unit monitors the integrity of each sub-information after the splitting, if the sub-information data is complete, it continues to process it, if the sub-information data is not complete, the first monitoring unit sends the sub-information code to the operation module through the monitoring module, the monitoring personnel replaces the complete sub-information data according to the sub-information code, the encryption unit generates a combined key according to the information code corresponding to each sub-information and the preset cloud service message classification code according to the preset key coding rule, encrypts each sub-information through the combined key to generate a corresponding number of ciphertexts and sends them to the management module again; S3, the management module receives each batch of ciphertexts, and integrates the ciphertexts into multiple groups of data packets and sends the data packets to the distribution module, the distribution module decomposes and scrambles the multiple groups of data packets together, recombines a set number of arbitrary ciphertexts to form new multiple groups of data packets, at this time, the second monitoring unit monitors the number of ciphertexts in each group of data packets, if the number of ciphertexts in each group of data packets is the same, the groups of data packets are sent to the receiving module in turn according to the preset time interval, if the number is not the same, the second monitoring unit sends the ciphertext loss information to the operation module through the monitoring module, the monitoring personnel manually monitors the sub-information coding of each cloud service message, and supplements the lost sub-information according to the sub-information coding; S4, the receiving module receives the multiple groups of data packets in turn and uniformly decomposes the data packets, obtains the sub-information ciphertexts corresponding to the multiple groups of data packets according to the classification coding of each cloud service message, and decodes the ciphertexts according to the combination key, integrates the decoded sub-information according to the information coding sequence, at this time, the third monitoring unit monitors the sub-information coding of each integrated cloud service message, and forms a complete cloud service message; The specific steps that the third monitoring unit monitors the sub-information coding of each integrated cloud service message in step S4 are as follows: the third monitoring unit monitors the information coding sequence of each integrated cloud service message, if the sequence is accurate, a complete cloud service message is formed, if the sequence is not accurate, the cloud service message is packaged and sent to the operation module through the monitoring module, the information coding is manually re-identified by the monitoring personnel, the cloud service message is integrated according to the information coding sequence, and a complete cloud service message is formed.

5. The method of using a cloud service messaging monitoring system for high performance computing of claim 4, wherein, In step S4, the receiving module receives the multiple groups of data packets in turn and applies for the combination key of each cloud service message to the management module, the management module calls the combination key generated by each high-performance computing node and the classification coding of the cloud service message, and uniformly sends the combination key and the classification coding to the receiving module, and the receiving module decrypts the corresponding cloud service message sub-information ciphertext according to the classification coding of the cloud service message and the corresponding combination key.

Images