A QoS cloud host communication queue guarantee method, storage device and intelligent terminal for quantum computing equipment

Abstract

The present invention discloses a QoS cloud host communication queue guarantee method for quantum computing devices, which applies a cloud host and multiple quantum computing devices, including the following steps: N QoS queues of different levels are set in the cloud host, and the levels of the N queues are arranged in the order of Q1>Q2>Q3>...>Qn; the Q1 and Q2 queues in the N queues are set as SP scheduling groups, and the rest are set as WDRR scheduling groups, and the queues of the WDRR scheduling group are all set with different scheduling weights; the cloud host initiates service requests to multiple quantum computing devices, and the service request with the highest priority enters the SP scheduling group queue, and the rest enters the WDRR scheduling group queue. The present invention adjusts the message by giving the cloud host message capacity space and monitoring and judging whether the queue exceeds the message capacity space, thereby avoiding network congestion and the random discarding of messages due to insufficient queue space.

Description

Technical Field

[0001] The present invention relates to the field of quantum computing, and in particular to a QoS cloud host communication queue guarantee method, a storage device and an intelligent terminal for quantum computing equipment. Background Art

[0002] When quantum computing devices communicate, the QoS-based cloud host communication queue guarantee solution includes the message scheduling and sending stage and the message receiving stage. For the message sending stage, the cloud hosts with different Qos levels are pre-allocated with the corresponding physical network card sending queue space, and then the message is judged and sent; for the message receiving stage, according to the remaining space of the receiving queue, when the remaining space is insufficient, some of the messages that need to be received are discarded according to the Qos level of the message, and the remaining messages are sent to the receiving queue.

[0003] The defects in this solution are: if the receiving queue space is insufficient, the network resources consumed when the message is forwarded from the sending queue to the discarded location have been wasted, and the sending queue space and the receiving space are not allocated properly, resulting in many message sending failures, affecting the system business.

[0004] Therefore, it is necessary to improve the existing technology to solve the problem that the cloud host communication queue of QoS used for quantum computing device business is randomly discarded due to network congestion and insufficient queue space, which affects the system business and may cause avalanche in the cloud host system, and the low QoS level business cannot be processed for a long time. Summary of the invention

[0005] In order to solve the above technical problems, a cloud host communication queue guarantee method for QoS of a cloud host communication queue for quantum computing device business due to network congestion and insufficient queue space is proposed.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is as follows: A QoS cloud host communication queue guarantee method for quantum computing devices, which uses a cloud host and multiple quantum computing devices, and the cloud host communicates with the multiple quantum computing devices through a network. The method includes the following steps:

[0007] Step 1: N QoS queues of different levels are set in the cloud host, and the order of the levels of the N queues is Q1>Q2>Q3>...>Qn;

[0008] Step 2: Set the Q1 and Q2 queues among the N queues as the SP scheduling group, and set the rest as the WDRR scheduling group. The queues in the WDRR scheduling group are all set with different scheduling weights.

[0009] Step 3: The cloud host initiates service requests to multiple quantum computing devices. The service requests have different priority levels according to their types. The service request with the highest priority level enters the SP scheduling group queue, and the remaining service requests enter the WDRR scheduling group queue.

[0010] Preferably, the maximum message capacity of the SP scheduling group is 1024Bytes;

[0011] The maximum scheduling weight in the WDRR scheduling group is W max The lower limit of the message capacity of the queue with the largest scheduling weight is K min Bytes, upper limit is K max Bytes, the maximum discard rate is H;

[0012] The lower limit of the message capacity of other queues in the WDRR scheduling group is W i / W max *K min Bytes, upper limit is W i / W max *K min Bytes, the maximum discard rate is W max / W i *H, where W i is the weight value of the i-th queue.

[0013] Preferably, the types of the service request at least include obtaining device data, setting key generation rate parameters, querying key error rate parameters and enabling inspection strategies.

[0014] Preferably, the priorities of the service requests for acquiring device data, setting key generation rate parameters, querying key error rate parameters and enabling inspection strategies are P1, P2, P3 and P4 respectively, where P1>P2>P3>P4.

[0015] Preferably, the K min The value is 768, the K max取 The value is 1024.

[0016] Preferably, when the cloud host initiates a service request to obtain device data from the first quantum computing device and the second quantum computing device respectively, the request size is 2 messages of X Bytes, and the cloud host sends messages to the corresponding first quantum computing device and the second quantum computing device respectively through queues Q1 and Q2, 2X<1024.

[0017] Preferably, when the cloud host initiates a service request to set the key generation rate parameter, query the key error rate parameter, and start the inspection strategy to the i-th quantum computing device, the cloud host selects the queue Q of the WDRR scheduling group that meets the conditions according to the size Y Bytes of the sent service. i Send a message to the corresponding i-th quantum computing device.

[0018] Preferably, the queue Q i The conditions for satisfying the following are: Use the WDRR algorithm to queue Q i Set a counter Deficit;

[0019] The initial value of Deficit is W*MTU, where W is the weight of the queue and MTU is the maximum transmission unit;

[0020] Each time the queue Q is polled i When i A message is output and the counter Deficit is reduced by the message length;

[0021] When the counter reaches 0, stop scheduling the queue Q i , but continue to schedule other queues Q whose counters are not 0 j ;

[0022] When the counters of all queues are 0, the Deficit of all counters is added with W*MTU, and a new round of scheduling of the queues of the WDRR scheduling group begins, where i and j are positive integers.

[0023] Preferably, when the messages in queue Q1 and queue Q2 of the SP scheduling group have reached the maximum length of the queue, when network congestion occurs, queue Q1 and queue Q2 directly discard the newly arrived messages and return a message sending failure.

[0024] Preferably, for the queues in the WDRR scheduling group, when the queue Q i The number of queue messages reaches Q i When the packet capacity reaches the lower limit, if there are still packets to enter queue Q i , then the queue Q i Start discarding messages in proportion, the discarding proportion is A=(BBytes÷256Bytes)×100%, where the value of B is the size of the sent message;

[0025] When the queue Q i When the queue capacity reaches the upper limit, all packets that are about to enter queue Q are discarded. i Messages;

[0026] When the queue Q iThe packet capacity of each queue is between the lower limit and the upper limit, and the packet loss rate of each queue is not greater than the set maximum packet loss rate. Until the packet capacity upper limit of each queue is reached, all packets waiting to enter the queue are discarded.

[0027] Beneficial technical effects of the present invention: By providing cloud host message capacity space and monitoring and judging whether the queue exceeds the message capacity space, the present invention can adjust the messages entering the queue in advance, avoiding network congestion and random discarding of messages due to insufficient queue space, thereby ensuring that the information of the cloud host can be smoothly sent to each quantum computing device and facilitating the maintenance of the cloud host system; the SP scheduling algorithm and the WDRR scheduling algorithm are used in the cloud host queue. The combination of the two algorithms improves the communication service capability between the cloud host and the quantum computing device, and can process both low-latency and high-latency services at the same time, avoiding the situation where low-QoS-level services cannot be processed for a long time. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a principle block diagram of the system of the present invention;

[0029] Figure 2 It is a regular diagram of SP scheduling in the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments, but the scope of protection claimed in the present invention is not limited to the following specific embodiments.

[0031] like Figure 1 and Figure 2 As shown, a QoS cloud host communication queue guarantee method for a quantum computing device, using a cloud host and multiple quantum computing devices, the cloud host communicates with the multiple quantum computing devices through a network, wherein:

[0032] The quantum computing device in this embodiment is a quantum computer, which is a physical device that follows the laws of quantum mechanics to perform high-speed mathematical and logical operations, store and process quantum information. When a device processes and calculates quantum information and runs quantum algorithms, it is a quantum computer. The main characteristics of quantum computers are fast operation speed, strong information processing capabilities, and a wide range of applications. Compared with general computers, the more information is processed, the more advantageous it is for quantum computers to perform operations, and the more accurate the operations can be.

[0033] The method comprises the following steps:

[0034] Step 1: N QoS queues of different levels are set in the cloud host, and the order of the levels of the N queues is Q1>Q2>Q3>...>Qn;

[0035] Step 2: Set the Q1 and Q2 queues among the N queues as the SP scheduling group, and set the rest as the WDRR scheduling group. The queues in the WDRR scheduling group are all set with different scheduling weights.

[0036] Step 3: The cloud host initiates service requests to multiple quantum computing devices. The service requests have different priority levels according to their types. The service request with the highest priority level enters the SP scheduling group queue, and the remaining service requests enter the WDRR scheduling group queue.

[0037] Specifically, the SP scheduling group adopts the SP scheduling algorithm. The SP scheduling algorithm preferentially sends packets in higher priority queues in order of priority from high to low. When the higher priority queue is empty, the packets in the lower priority queue are sent. The WDRR scheduling algorithm (Weighted Round Robin) is based on the weighted scheduling WRR of the message. It can be configured how many messages are scheduled from each queue before transferring to the next queue.

[0038] The maximum message capacity of the SP scheduling group is 1024 Bytes;

[0039] The maximum scheduling weight in the WDRR scheduling group is W max The lower limit of the message capacity of the queue with the largest scheduling weight is K min Bytes, upper limit is K max Bytes, the maximum discard rate is H;

[0040] The lower limit of the message capacity of other queues in the WDRR scheduling group is W i / W max *K min Bytes, upper limit is W i / W max *K min Bytes, the maximum discard rate is W max / W i *H, where W i is the weight value of the i-th queue. In this embodiment, K min The value is 768, the K max The value is 1024.

[0041] The scheduling strategy is: the cloud host initiates service requests to multiple quantum computing devices, and the service requests have different priority levels according to their types; the service request with the highest priority level enters the SP scheduling group queue, and the remaining service requests enter the WDRR scheduling group queue.

[0042] The types of service requests include at least obtaining device data, setting key generation rate parameters, querying key error rate parameters, and enabling patrol strategies. The priorities of service requests for obtaining device data, setting key generation rate parameters, querying key error rate parameters, and enabling patrol strategies are P1, P2, P3, and P4, respectively, where P1>P2>P3>P4.

[0043] When the cloud host initiates a service request for obtaining device data from the first quantum computing device and the second quantum computing device respectively, the size is 2 messages of X Bytes. According to the scheduling strategy, the priority level of the service for obtaining device data is P1, which has the highest priority level. Therefore, the service request enters the SP scheduling group queue, and the sum of the sizes of the 2 messages of X Bytes should be less than the maximum message capacity of the SP scheduling group, which is 1024Bytes, that is, 2X<1024. At this time, the cloud host sends messages to the corresponding first quantum computing device and the second quantum computing device through queues Q1 and Q2 respectively.

[0044] When the cloud host initiates a service request to set the key generation rate parameter, query the key error rate parameter, and start the inspection strategy to the i-th quantum computing device, the service request enters the WDRR scheduling group queue according to the scheduling strategy. The cloud host selects the queue Q of the WDRR scheduling group that meets the conditions according to the size Y Bytes of the sent service. i Send a message to the corresponding i-th quantum computing device, that is, Y should be less than the queue Q i The maximum message capacity.

[0045] Preferably, the queue Q i The conditions for satisfying the following are: Use the WDRR algorithm to queue Q i Set a counter Deficit;

[0046] The initial value of Deficit is W*MTU, where W is the weight of the queue and MTU is the maximum transmission unit;

[0047] Each time the queue Q is polled i When i A message is output and the counter Deficit is reduced by the message length;

[0048] When the counter reaches 0, stop scheduling the queue Q i , but continue to schedule other queues Q whose counters are not 0j ;

[0049] When the counters of all queues are 0, the Deficit of all counters is added with W*MTU, and a new round of scheduling of the queues of the WDRR scheduling group begins, where i and j are positive integers.

[0050] Specifically, when the messages in queues Q1 and Q2 of the SP scheduling group have reached the maximum length of the queues, when network congestion occurs, queues Q1 and Q2 directly discard the newly arrived messages and return a message sending failure.

[0051] For the queues in the WDRR scheduling group, when the queue Q i The number of queue messages reaches Q i When the packet capacity reaches the lower limit, if there are still packets to enter queue Q i , then the queue Q i Start discarding messages in proportion, the discarding proportion is A=(B Bytes÷256Bytes)×100%, where the value of B is the size of the sent message;

[0052] When the queue Q i When the queue capacity reaches the upper limit, all packets that are about to enter queue Q are discarded. i Messages;

[0053] When the queue Q i The packet capacity of each queue is between the lower limit and the upper limit, and the packet loss rate of each queue is not greater than the set maximum packet loss rate. Until the packet capacity upper limit of each queue is reached, all packets waiting to enter the queue are discarded.

[0054] The specific working process of the present invention is described below by taking N=5 as an example:

[0055] When N=5, the system of the present invention includes 1 cloud host and 5 quantum computing devices quantum computing 1, quantum computing 2, quantum computing 3, quantum computing 4 and quantum computing 5.

[0056] There are five different levels of QoS queues in the cloud host: Q1, Q2, Q3, Q4 and Q5. The QoS levels of these five queues are arranged in the following order: Q1>Q2>Q3>Q4>Q5, among which queues Q2 and Q2 are SP scheduling groups; queues Q3, Q4 and Q5 are WDRR scheduling groups.

[0057] Before the cloud host communicates with the five quantum computing devices, the maximum message capacity of queues Q1 and Q2 in the SP scheduling group is pre-set to 1024Bytes;

[0058] The lower limit of the message capacity of queue Q3 in the WDRR scheduling group is 768 bytes, the upper limit is 1024 bytes, and the maximum discard rate is 50% (meaning that messages smaller than 768 bytes are not discarded; messages between 768 bytes and -1024 bytes have a discard rate of 50%; messages larger than 1024 bytes are directly discarded).

[0059] The lower limit of the message capacity of queue Q4 is 640 bytes, the upper limit is 853 bytes, and the maximum discard rate is 60% (indicates that if the message is less than 640 bytes, the message will not be discarded; if the message size is between 640 bytes and -853 bytes, the message discard rate is 60%, and if the message size is greater than 853 bytes, the message is directly discarded). The weight ratio of queue Q4 to queue Q3 is W4 / W3=640 / 768.

[0060] The lower limit of the message capacity of queue Q5 is 512Bytes, the upper limit is 683Bytes, and the maximum discard rate is 75% (meaning that messages less than 512Bytes will not be discarded; messages between 512Bytes and -683Bytes have a discard rate of 75%, and messages larger than 683Bytes are directly discarded). The weight ratio of queue Q5 to queue Q4 is W5 / W3=512 / 768. If the total amount of messages in the queue does not exceed the maximum message capacity or the lower limit of the capacity, they can be completely sent out of the queue.

[0061] The cloud host initiates real-time device data acquisition requests to quantum computing device 1 and quantum computing device 2 respectively. The request size is 2 messages of 200 bytes. According to the setting of the scheduling strategy, the cloud host can completely send messages to quantum computing device 1 and quantum computing device 2 through queues Q1 and Q2 respectively.

[0062] Cloud host A initiates a request to set the key generation rate parameters to quantum computing device 3. The request size is 2 150-Bytes messages. According to the setting of the scheduling strategy, the requested message can be completely sent to quantum computing device 3 through queue Q3.

[0063] Cloud host A initiates a key bit error rate parameter query request to quantum computing device 4. The size of this request is 4 100Bytes. According to the setting of the scheduling strategy, this request message can be completely sent to quantum computing device 4 through queue Q4.

[0064] Cloud host A initiates a request to enable the inspection strategy to quantum computing device 5. The packet size of this request is 6 50-Bytes messages. According to the setting of the scheduling strategy, this request message can be completely sent to quantum computing device 5 through queue Q5.

[0065] First, the QoS level of queue Q1 is greater than that of queue Q2. Both queues Q1 and Q2 are in the SP scheduling group, that is, queues Q1 and Q2 are scheduled according to the SP scheduling algorithm (the SP (Strict Priority) scheduling algorithm is strictly scheduled in the order of queue priority. Only after all the high-priority queue data has been scheduled, the low-priority queue has the opportunity to be scheduled). Therefore, the scheduling order of the SP scheduling group is as follows: Figure 1 That is, the two 200-Byte messages in queue Q1 are first dequeued one by one and sent to quantum computing device 1; after the messages in queue Q1 are dequeued, the two 200-Byte messages in queue Q2 can be dequeued one by one and sent to quantum computing device 2.

[0066] After the queues of the SP group are scheduled, that is, after the messages of queue Q2 are scheduled out of the queue, the queues of the WDRR scheduling group are scheduled according to the QoS level of the queues in the WDRR group and the weight ratio of the queues. The WDRR algorithm (Weighted Deficit Round Robin (DWRR) WDRR algorithm sets a counter Deficit for each queue, and Deficit is initialized to Weight*MTU, where Weight is the weight of the queue and MTU is the maximum transmission unit. Each time a queue is polled, the queue outputs a message and the counter Deficit is reduced by the message length. When the counter is 0, the scheduling of the queue is stopped, but other queues whose counters are not 0 are continued. When the counters of all queues are 0, the Deficit of all counters is added with Weight*MTU, and a new round of scheduling begins) is used to schedule the queues of the WDRR scheduling group. Assuming that the MTU of the WDRR scheduling group is 100Bytes, the scheduling order of the WDRR scheduling group is as follows: Figure 2 , the messages of each queue are dispatched and sent to the device. The specific steps are as follows:

[0067] First scheduling: Deficit[3] of queue Q3 = W3*MTU = 3*100 = 300, then a 150-byte message is taken out of queue Q3 and sent to quantum computing device 3. After sending, Deficit[3] = 150; Deficit[4] of queue Q4 = W4*MTU = 2*100 = 200, then a 100-byte message is taken out of queue Q4 and sent to quantum computing device 4. After sending, Deficit[4] = 100; Deficit[5] of queue Q5 = W5*MTU = 1*100 = 100, then a 50-byte message is taken out of queue Q5 and sent to quantum computing device 5. After sending, Deficit[5] = 50;

[0068] Second scheduling: Take a 150-Byte message from queue Q3 and send it to quantum computing device 3. After sending, Deficit[3]=0; take a 100-Byte message from queue Q4 and send it to quantum computing device 4. After sending, Deficit[4]=0; take a 50-Byte message from queue Q5 and send it to quantum computing device 5. After sending, Deficit[5]=0;

[0069] The third scheduling: Since queue Q2 is empty after two schedulings, while queues Q3 and Q4 are not empty, queue Q2 does not participate in the next scheduling, and queues Q3 and Q4 continue to participate in the scheduling. At this time, Deficit[3] = 200, so a 200-Byte message is taken from queue Q3 and sent to quantum computing device 4. After sending, Deficit[3] = 100; Deficit[4] of queue Q4 = 100, so a 50-Byte message is taken from queue Q4 and sent to quantum computing device 5. After sending, Deficit[4] = 50;

[0070] Fourth scheduling: a 100-Byte message is taken from queue Q4 and sent to quantum computing device 4. After sending, Deficit[4] = 0; a 50-Byte message is taken from queue Q5 and sent to quantum computing device 5. After sending, Deficit[5] = 0;

[0071] Fifth scheduling: Since queue Q4 is empty after four schedulings, and queue Q5 is not empty, queue Q5 will no longer participate in the next scheduling, and queue Q5 will continue to be scheduled. Deficit[5] of queue Q5 = 100, so a 50-Byte message is taken from queue Q5 and sent to quantum computing device 5. After sending, Deficit[5] = 50;

[0072] The sixth scheduling: a 50-Byte message is taken out from queue Q5 and sent to quantum computing device 5. After sending, Deficit[5]=0. At this time, queue Q5 is also empty, so the scheduling ends.

[0073] Monitor each queue and its usage. The processing of congestion by each queue is as follows:

[0074] When the messages in queues Q1 and Q2 have reached the maximum length of the queues, since both queues Q1 and Q2 belong to the SP scheduling group, when network congestion occurs, queues Q1 and Q2 directly discard the newly arrived messages and return a message sending failure;

[0075] For queues in the WDRR scheduling group, the queues start to drop packets before congestion. That is, when the queue packet volume of queues Q3, Q4, and Q5 reaches the lower limit of their respective packet capacity, if there are still packets to enter queues Q3, Q4, and Q5, queues Q3, Q4, and Q5 start to drop packets at a certain ratio; when the queue packet capacity of queues Q3, Q4, and Q5 reaches the upper limit, all packets that are about to enter queues Q3, Q4, and Q5 are dropped; when the packet capacity of queues Q3, Q4, and Q5 is between the lower limit and the upper limit, the packet loss rate of each queue is not greater than the maximum packet loss rate set for each queue, until the packet capacity upper limit of each queue is reached, and all packets that are about to enter the queue are dropped.

[0076] For example, for queue Q3, the lower limit of the message capacity is 768 bytes, the upper limit is 1024 bytes, the maximum discard rate is 50%, and the difference between the upper limit and the lower limit of queue Q3 is 256 bytes. If the message capacity of queue Q2 has reached 768 bytes, in case 1, there is still one message of 100 bytes to enter queue Q3; in case 2, there are still two messages of 300 bytes to enter queue Q2.

[0077] For case 1, the packet discard rate of 100 Bytes is (100 Bytes ÷ 256 Bytes) × 100% ≈ 39%, which is less than the maximum discard rate of the queue. Therefore, only 61 Bytes of the 100 Bytes packet will enter queue Q3. At this time, the total number of packets in queue Q3 is 829 Bytes, which has not reached the upper limit of the packet capacity and can still receive new packets.

[0078] For case 2, there are two 300-Byte packets. The discard rate of the first 300-Byte packet is (300-Bytes ÷ 256-Bytes) × 100% > 50%. This is greater than the maximum discard rate of the packet. Therefore, the first 300-Byte packet has only 300-Bytes-(300 × 50%) = 150-Bytes. After the first 300-Byte packet enters, the queue capacity is 918-Bytes. At this time, the queue capacity of Q3 has not reached the upper limit. Therefore, the second 300-Byte packet is discarded. s, 300Bytes-(300×50%)=150Bytes, but 918Bytes+150Bytes=1068Bytes, which is greater than the upper limit of the message capacity of queue Q3 (1024Bytes). Therefore, of the second 300Bytes message, only 1024Bytes-918Bytes=106Bytes can enter queue Q3. At this time, if there are new messages to enter queue Q3, because the capacity of Q3 is full, queue Q3 directly discards all new messages and returns the message failure.

[0079] The present invention provides cloud host message capacity space and monitors and determines whether the queue exceeds the message capacity space, so as to adjust the messages entering the queue in advance, avoid network congestion and the random discarding of messages due to insufficient queue space, thereby ensuring that the information of the cloud host can be smoothly sent to each quantum computing device and facilitating the maintenance of the cloud host system; the SP scheduling algorithm and the WDRR scheduling algorithm are used in the cloud host queue. The combination of the two algorithms improves the communication service capacity between the cloud host and the quantum computing device, and can process both low-latency services and high-latency services at the same time, avoiding the low QoS level services from not being processed for a long time.

[0080] According to the disclosure and teaching of the above description, those skilled in the art to which the present invention belongs may also make changes and modifications to the above embodiments. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and some modifications and changes to the invention should also fall within the scope of protection of the claims of the present invention. In addition, although some specific terms are used in this specification, these terms are only for the convenience of description and do not constitute any limitation to the invention.

Claims

1. A QoS cloud host communication queue guarantee method for quantum computing devices, characterized in that: A cloud host and a plurality of quantum computing devices are applied, and the cloud host communicates with the plurality of quantum computing devices through a network. The method comprises the following steps: Step 1: N QoS queues of different levels are set in the cloud host, and the order of the levels of the N queues is Q1>Q2>Q3>...>Qn; Step 2: Set the Q1 and Q2 queues among the N queues as the SP scheduling group, and set the rest as the WDRR scheduling group. The queues in the WDRR scheduling group are all set with different scheduling weights. Step 3: The cloud host initiates service requests to multiple quantum computing devices. The service requests have different priority levels according to their types. The service request with the highest priority level enters the SP scheduling group queue, and the remaining service requests enter the WDRR scheduling group queue. When the cloud host initiates a service request for obtaining device data from the first quantum computing device and the second quantum computing device respectively, it requests two messages of size X Bytes. According to the scheduling strategy, the priority level of the service for obtaining device data is P1, which has the highest priority level. Therefore, the service request enters the SP scheduling group queue, and the sum of the sizes of the two messages of X Bytes is less than the maximum message capacity of the SP scheduling group, 1024Bytes, that is, 2X<1024. At this time, the cloud host sends messages to the corresponding first quantum computing device and the second quantum computing device respectively through queues Q1 and Q2; When the cloud host initiates a service request to set the key generation rate parameters, query the key error rate parameters, and enable the inspection strategy to the i-th quantum computing device, the service request enters the WDRR scheduling group queue according to the scheduling strategy. The cloud host selects the queue Q of the WDRR scheduling group that meets the conditions according to the size Y Bytes of the request message. i Send a message to the corresponding i-th quantum computing device, that is, Y is less than the queue Q i Maximum message capacity; When the messages in queues Q1 and Q2 of the SP scheduling group have reached the maximum length of the queues, queues Q1 and Q2 directly discard the newly arrived messages when network congestion occurs, and return a message sending failure; For the queues in the WDRR scheduling group, when the queue Q i The number of queue messages reaches Q i When the packet capacity reaches the lower limit, if there are still packets to enter queue Q i , then the queue Q i Start discarding messages in proportion, the discarding proportion is A = (B Bytes ÷ 256 Bytes) × 100%, where the value of B is the size of the message being sent; When the queue Q i When the queue capacity reaches the upper limit, all packets that are about to enter queue Q are discarded. i Messages; When the queue Q i The packet capacity of each queue is between the lower limit and the upper limit, and the packet loss rate of each queue is not greater than the set maximum packet loss rate. Until the packet capacity upper limit of each queue is reached, all packets waiting to enter the queue are discarded.

2. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 1, characterized in that: The maximum message capacity of the SP scheduling group is 1024Bytes.

3. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 2, characterized in that: The maximum scheduling weight in the WDRR scheduling group is W max The lower limit of the message capacity of the queue with the largest scheduling weight is K min Bytes, upper limit is K max Bytes, the maximum discard rate is H.

4. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 3, characterized in that: The types of the service request include at least obtaining device data, setting key generation rate parameters, querying key bit error rate parameters and enabling inspection strategies.

5. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 4, characterized in that: The priorities of the service requests for obtaining device data, setting key generation rate parameters, querying key error rate parameters and enabling inspection strategies are P1, P2, P3 and P4 respectively, where P1>P2>P3>P4.

6. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 3, characterized in that: The K min The value is 768, the K max The value is 1024.

7. A QoS cloud host communication queue guarantee method for quantum computing equipment as claimed in claim 6, characterized in that: Queue Q i The conditions are as follows: Use the WDRR algorithm to queue Q i Set a counter Deficit; The initial value of Deficit is W*MTU, where W is the weight of the queue and MTU is the maximum transmission unit; Each time the queue Q is polled i When i A message is output and the counter Deficit is reduced by the message length; When the counter reaches 0, stop scheduling the queue Q i , but continue to schedule other queues Q whose counters are not 0 j ; When the counters of all queues are 0, the Deficit of all counters is added with W*MTU, and a new round of scheduling of the queues of the WDRR scheduling group begins, where i and j are positive integers.

8. A storage device, wherein a plurality of instructions are stored in the storage device, characterized in that: The instructions are suitable for being loaded by a processor and executing the steps of a QoS cloud host communication queue guarantee method for a quantum computing device as described in any one of claims 1-7.

9. An intelligent terminal, comprising a processor for executing instructions and a storage device for storing a plurality of instructions, characterized in that: The instructions are suitable for being loaded by the processor and executing the steps of a QoS cloud host communication queue guarantee method for a quantum computing device as described in any one of claims 1-7.

Images