High-reliability data service system and method

Abstract

A high-reliability data service system comprises more than two token consumption node groups, a high-reliability service quality management cluster and more than two storage nodes, and the token consumption node groups comprise a token consumption node group A and a token consumption node group B; the token consumption node group comprises a token generation node and at least one token consumption node; the token generation node comprises a token generation module and a total token bucket; the token consumption node comprises a local token bucket; a token consumption node periodically receives tokens from the total token bucket and stores the tokens in a local token bucket; the token consuming node receives the token at the local token bucket.

Description

Technical Field

[0001] The present application belongs to the field of big data technology, and in particular relates to a high-reliability data service system and method. Background Art

[0002] Usually QOS (Quality of Service) provides the following three service models:

[0003] l. Best-Effort service--Best-Effort service model is a single service model and the simplest service model. This model does not provide any guarantee for performance such as latency and reliability.

[0004] 2. Integrated service--Integrated service model, which can meet multiple QoS requirements. This system can clearly distinguish and guarantee the service quality of each business flow, providing the most fine-grained service quality distinction for the network.

[0005] 3. Differentiated service--Differentiated service model, which can meet different QOS requirements. Unlike Int-Serv, it does not need to notify the network to reserve resources for each service. Differentiated service is simple to implement and has good scalability.

[0006] The position of QOS technology in the network, flow classification: using certain rules to identify messages that meet certain characteristics, it is the premise and basis for differentiating network services.

[0007] Traffic policing: Regulate specific traffic entering or leaving the device. When the traffic exceeds the set value, restriction or penalty measures can be taken to protect network resources from damage. It can be applied to both the inbound and outbound directions of the interface.

[0008] Traffic shaping: A traffic control measure that actively adjusts the output rate of a flow to adapt the flow to the network resources available on the downstream device and avoid unnecessary packet discarding. It usually acts in the direction of the interface.

[0009] Congestion management: It is how to formulate a resource scheduling strategy when congestion occurs to determine the processing order of message forwarding, which usually acts in the direction of the interface.

[0010] Congestion avoidance: monitors the usage of network resources and takes the initiative to discard packets when congestion is found to be increasing. It adjusts the queue length to relieve network overload, usually at the interface.

[0011] However, traditional QOS does not have a metric to evaluate node QOS performance, let alone a technical means to evaluate the overall QOS performance of the network. Traffic monitoring of a single node cannot evaluate node QOS performance. When traffic congestion occurs, it can only discard messages without any technical means to adjust and improve.

[0012] The principle of the token bucket management unit algorithm is that the system will put tokens into the bucket at a constant rate, and if a request needs to be processed, it needs to get a token from the bucket first. When there is no token to be taken from the bucket, the service is denied. When the bucket is full, the newly added token is discarded or rejected.

[0013] The token bucket management unit algorithm is a bucket that stores fixed-capacity tokens, and adds tokens to the bucket at a fixed rate. The token bucket management unit algorithm can basically be described by the following concepts: Tokens will be put into the token bucket management unit at a fixed rate. For example, 100 tokens are put in per second. The bucket can store at most b tokens. When the bucket is full, the newly added tokens are discarded or rejected.

[0014] The token bucket has several important parameters:

[0015] Token generation rate: indicates the number of tokens generated per second, that is, the number of tokens generated per unit time. This parameter affects the maximum throughput of the system and is also the core parameter of current limiting.

[0016] Token bucket management unit capacity: indicates the maximum number of tokens that can be stored in the token bucket management unit. When a request comes in, if the number of tokens in the bucket is insufficient, the request is rejected.

[0017] Warm-up time: indicates that after the system is started, the token bucket management unit will gradually increase the token generation rate until it reaches the set maximum rate. This parameter can avoid a large number of tokens when the system is started.

[0018] In a distributed network, there are many nodes with equal network positions. For example, in a distributed storage network, there are many gateway nodes. These gateway nodes can run token buckets for QOS management, but the QOS of a single gateway node cannot improve the overall network. The token bucket technology can prevent the subsequent network from crashing due to traffic overload. It is essentially a flow-limiting technology that cannot improve the QOS capabilities of the nodes, nor can it improve the traffic balance and overall network QOS.

[0019] Definitions of Abbreviations and Key Terms

[0020] QOS Quality of Service

[0021] ID Identity document

[0022] TPS Transactions Per Second The number of requests transmitted per second Summary of the invention

[0023] A high-reliability data service system comprises more than two token consumption node groups, wherein the token consumption node groups comprise a token consumption node group A and a token consumption node group B, a high-reliability service quality management cluster, and more than two storage nodes; the token consumption node groups are signal-connected to the high-reliability service quality management cluster; the token consumption node groups comprise a token generation node, and at least one token consumption node; the token generation node comprises a token generation module and a total token bucket; the token consumption node comprises a local token bucket; the token consumption node periodically obtains tokens from the total token bucket and stores them in a local token bucket; the token consumption node obtains tokens from the local token bucket; the high-reliability service quality management cluster is used to configure the speed at which the token generation module generates tokens; the token generation node and the token consumption node report the status of themselves or other nodes to the high-reliability service quality management cluster, and the high-reliability service quality management cluster establishes a node status table according to the status report; the high-reliability service quality management cluster republishes the updated node status table after the node status changes.

[0024] According to the above-mentioned high reliability data service system, the status report includes the remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

[0025] According to the above-mentioned high-reliability data service system, the high-reliability service quality management cluster collects the operating parameters or operating status of the token consumption nodes in the token consumption node group. The high-reliability service quality management cluster sets the traffic weight in the node status table according to the operating status of the token consumption node. The token consumption nodes in each token consumption node group can obtain the token generation node of this token consumption group through HASH calculation based on the node status table and each token consumption node group and the traffic weight; the service access end selects the corresponding token consumption node to process the request based on the node status table, each token consumption node group and the traffic weight.

[0026] According to the above-mentioned high-reliability data service system, the operating parameters include network bandwidth and CPU type; the operating status includes transmission delay, CPU utilization, and network bandwidth ratio.

[0027] According to the above-mentioned high-reliability data service system, token consumption node group A is used to process business A, and token consumption node group B is used to process business B.

[0028] According to the above-mentioned high-reliability data service system, the high-reliability service quality management cluster outputs a visual operation status diagram, which includes the operation status of the token generation node, and the status report of the token generation node includes the remaining token capacity ratio.

[0029] According to the above-mentioned high-reliability data service system, the operating status of the token generation node includes the token generation rate and the remaining token capacity ratio.

[0030] According to the above-mentioned high-reliability data service system, the token consumption node or the token generation node in the token consumption node group includes a heartbeat management module; the high-reliability service quality management cluster includes a node status module; the heartbeat management module monitors the heartbeat status of other nodes, identifies the nodes without heartbeat messages as faulty nodes, and reports the faulty nodes to the high-reliability service quality management cluster.

[0031] According to the above-mentioned high-reliability data service system, if the difference between the remaining token capacity ratios of token consumption node group A and token consumption node group B is greater than the set value, and the remaining token capacity ratio of token consumption node group A is higher than the remaining token capacity ratio of token consumption node group B, the high-reliability service quality management cluster deploys at least one token consumption node of token consumption node group A to token consumption node group B; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0032] According to the above-mentioned high-reliability data service system, combined with the load status of the token consumption node and the remaining token capacity ratio of the token consumption node group, the high-reliability service quality management cluster can dynamically adjust the traffic weight of each token consumption node, generate and publish a new version of the node status table.

[0033] According to the above-mentioned high-reliability data service system, when the proportion of the remaining token capacity of the total token bucket is greater than the set value, the high-reliability service quality management cluster reduces or adjusts to 0 the weight of at least one token consumption traffic associated with the token consumption node group; the corresponding token consumption node reduces the processing of related requests to the token consumption node group.

[0034] According to the above high reliability data service system, if the remaining token capacity of the total token bucket is less than the set value, the high reliability service quality management cluster will increase the traffic weight of the token consumption node group in each token consumption node.

[0035] According to the above-mentioned high reliability data service system, when the flow weight of the token consumption node in the group of all token consumption nodes is 0, the token consumption node enters a low power consumption waiting state.

[0036] According to the above-mentioned high-reliability data service system, when the proportion of the remaining token capacity of the total token bucket is less than the set value, the high-reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0037] According to the above-mentioned high-reliability data service system, when the proportion of the remaining token capacity of the total token bucket is greater than the set value, the high-reliability service quality management cluster deploys at least one token consumption node to the backup token consumption node group C; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0038] According to the above-mentioned high-reliability data service system, the high-reliability service quality management cluster builds a token consumption node business flow distribution ratio table; the high-reliability service quality management cluster periodically publishes the business flow distribution ratio table to the business access terminal.

[0039] A high-reliability data service method is used to manage a high-reliability service quality management cluster, a token generation node, and at least two token consumption nodes; a token generation node and at least one token consumption node form a token consumption node group; the high-reliability service quality management cluster is used to record token bucket configuration information, the token bucket configuration information includes a token generation speed for setting the token generation speed of the token generation node; the token generation node includes a total token bucket, which generates a specified number of tokens within a specified time according to the token generation speed and places them in the total token bucket; the token consumption node includes a local token bucket, which receives tokens from the total token bucket of the token generation node in the same group and places them in the local token bucket; the token consumption node receives a service request and receives a token from the local token bucket; the token generation node and the token consumption node report the status of themselves or other nodes to the high-reliability service quality management cluster, and the high-reliability service quality management cluster establishes a node status table according to the status report; the high-reliability service quality management cluster periodically publishes the node status table through a broadcast channel; the node status table includes the token generation node and the token consumption node members of the token consumption node group.

[0040] According to the above-mentioned high-reliability data service method, when a token generation node fails, the token consumption node selects a node as a new token generation node from among the token consumption nodes based on the updated node status table, and the high-reliability service quality management cluster publishes the updated node status table through the broadcast channel.

[0041] According to the above high reliability data service method, the status report includes the remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

[0042] According to the above-mentioned high-reliability data service method, the token consumption node groups are token consumption node group A and token consumption node group B, the remaining token capacity ratio of token consumption node group A is higher than the remaining token capacity ratio of token consumption node group B, and at least one token consumption node of token consumption node group A is deployed to token consumption node group B; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0043] According to the above-mentioned high-reliability data service method, if the proportion of the remaining token capacity of the total token bucket is greater than the set value, at least one token consumption node of the token consumption node group is deployed to the backup token consumption node group C; the token consumption node of the backup token consumption node group C enters a low-power waiting state; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0044] According to the above-mentioned high-reliability data service method, the proportion of the remaining token capacity of the total token bucket is less than the set value, and the high-reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0045] According to the above high-reliability data service method, the high-reliability service quality management cluster builds a token consumption node service flow distribution ratio table; the high-reliability service quality management cluster periodically publishes the service flow distribution ratio table to the service access terminal.

[0046] One of the technical effects of the above technical solution is that in a large-scale data service system, different services require different network resources. By forming a token consumption node group with different numbers of servers, one server is responsible for the generation of tokens. Compared with the traditional method, each server manages its own token cylinder, the token issuance of a service can be effectively controlled by one token consumption node. Each service corresponds to multiple token consumption nodes, and there will be no situation where one service corresponds to one WEB page server, and when the WEB page server fails, the service of one service will be interrupted.

[0047] One of the technical effects of the above technical solution is that by setting independent token nodes for different businesses, the token consumption requirements of different businesses can be managed dynamically.

[0048] One of the technical effects of the above technical solution is: using a high-reliability service quality management cluster to supervise the working status of each node in the token consumption node group. When a node fails, especially a token node fails, the network node status can be published in the first time through the node status table, which greatly improves the reliability of the system.

[0049] One of the technical effects of the above technical solution is: in a large data center, a large database, hundreds of businesses are running at the same time, each business corresponds to a different token consumption node group, and the operating status of the token nodes in each token consumption node group reflects the business volume and operating status of a business. Through the remaining token capacity ratio of the token nodes in the token consumption node group, the data center administrator can intuitively observe the operating health of each business. With one parameter, the operating health of the entire network can be observed.

[0050] One of the technical effects of the above technical solution is: through the token consumption node business processing traffic weight configuration, servers with different processing capabilities can be coordinated to work together. Servers with strong capabilities can process more business, and older servers can process fewer businesses.

[0051] One of the technical effects of the above technical solution is: although the remaining token capacity ratio of the token consumption node can be used to observe the overall operation status of the network, the collection of network bandwidth, CPU type; transmission delay, CPU utilization, network bandwidth ratio and other information can be used as a reference when analyzing the network status in detail.

[0052] One of the technical effects of the above technical solution is: displaying the proportion of remaining tokens in the token bucket. The remaining token capacity proportion can be the proportion of remaining tokens in the total token bucket. The total number of token buckets is small, which directly corresponds to the business. Network administrators can intuitively observe the health of different business operations.

[0053] One of the technical effects of the above technical solution is: the proportion of remaining token capacity can be the proportion of remaining tokens in the local token buckets of different services. If there is a problem with the proportion of remaining tokens in the total token bucket of a service, such as the proportion becomes larger, the network administrator can check whether the configured number of token consumption nodes is insufficient through the proportion of remaining tokens in the local token bucket.

[0054] One of the technical effects of the above technical solution is: in traditional technology, heartbeat monitoring runs in the nodes of the entire network. The scale of heartbeat monitoring is large, and it is easy to have phenomena such as heartbeat monitoring islands. By forming a small heartbeat monitoring ring in the token consumption node group, the accuracy of heartbeat monitoring can be improved.

[0055] One of the technical effects of the above technical solution is that the traffic demand of business A becomes smaller, while the traffic demand of business B becomes smaller. By deploying the token consumption node 13 in the token consumption node group A to the token consumption node group B, the processing capacity of business B can be improved. After the deployment, the token generation speed of the token generation node can be adjusted so that the proportion of remaining tokens in the adjusted token bucket runs between a reasonable 30% and 50%.

[0056] One of the technical effects of the above technical solution is: service node 31 runs the token node of service A, and service node 33 runs the token node of service B; service node 31, service node 32, and service node 33 all run the token consumption nodes of service A and service B. By adjusting the traffic weights of the token consumption nodes of service A or service B in different service nodes, the resource occupancy ratio of service A and service B to the entire network can be changed.

[0057] One of the technical effects of the above technical solution is: according to the above high-reliability data service system, when the traffic weight of the token consumption node in all token consumption node groups is 0, it enters a low-power waiting state, which can save the power consumption of the service cluster.

[0058] One of the technical effects of the above technical scheme is: by setting the token consumption node group C as the backup support group, the idle servers are centrally deployed to the group for management and control. In the backup support group, a certain number of services can be dynamically set to be in the power-on backup state, and other servers are in the motherboard power-off state. The motherboard power-off state can greatly reduce the power consumption of the server. When the business demand increases, the server in the power-on backup state is first dispatched to the token consumption node group with demand. When the number of services in the power-on backup state drops to the set value, the server is gradually mobilized to turn on the motherboard power supply through the network card power-on mode, and gradually enters the backup support state. This method can greatly improve the efficiency of server operation. During low business time periods, some servers can be powered off and exited from service, greatly saving the power consumption of the data center.

[0059] One of the technical effects of the above technical solution is that the high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node after deployment. After adjustment, by modifying the token generation speed of the token generation node, the adjusted remaining token capacity ratio can be brought into the normal monitoring range. When the business volume changes, the remaining token capacity ratio will fluctuate and enter an abnormal state, increasing the number of nodes in the token consumption node group to be adjusted again.

[0060] One of the technical effects of the above technical solution is that the service access end obtains a service traffic distribution ratio table, which can distribute the traffic to nodes with different processing capabilities and dynamically balance the traffic of each service node.

[0061] One of the technical effects of the above technical solution is: through the three-level token management mode, the traditional token bucket technology is decomposed into a three-level mode of token bucket performance parameter management level, token generation level and token consumption level, so that in large-scale service clusters, the token bucket function division of labor is clear, the function of a single node is single, the operation efficiency is high, and the fault tolerance is good. At the token consumption level, multiple nodes share the business and obtain tokens on a token generation node, making the token management and control intuitive and efficient.

[0062] One of the technical effects of the above technical solution is: periodically obtaining the status of each token bucket, being able to monitor the QOS status of the entire network, and laying a foundation for scheduling the QOS of the entire network;

[0063] One of the technical effects of the above technical solution is that the ratio of the current number of tokens to the capacity of the token bucket management unit indicates the level of token consumption. If the ratio is high, it means that the current node has a high backlog of tokens and the node is in poor working condition. The periodic ratio of the current number of tokens to the capacity of the token bucket management unit can effectively evaluate the QOS performance of the node.

[0064] One of the technical effects of the above technical solution is that: the node service quality monitoring module collects each service quality QOS, can adjust the flow distribution size or flow distribution ratio of each gateway node, and adjust the balance of the entire network service;

[0065] One of the technical effects of the above technical solution is that the dynamic capacity consumption ratio of the token bucket can directly evaluate the QOS status of the node, and 50% can be regarded as healthy. If it is greater than 50%, it can be considered that the speed of token generation is greater than the speed of consumption; if it is less than 50%, it can be considered that the speed of consumption is greater than the speed of generation;

[0066] One of the technical effects of the above technical solution is that the dynamic capacity consumption ratio of the token bucket varies greatly, and the token bucket parameter reporting module can reduce the fluctuation frequency of the dynamic capacity consumption ratio of the token bucket by averaging the dynamic capacity consumption ratios of the token bucket collected multiple times;

[0067] One of the technical effects of the above technical solution is that: the main gateway node runs a token bucket and manages multiple sub-gateway nodes, which can greatly reduce the amount of information reported to the gateway management server. At the same time, gateway grouping can balance the speed and quantity of token allocation to sub-gateway nodes;

[0068] One of the technical effects of the above technical solution is: health sorting of network management nodes, sorting the same or similar token bucket management unit capacities together, and intuitively obtaining the QOS capabilities of each node;

[0069] One of the technical effects of the above technical solution is: health sorting of network management nodes, sorting nodes with the same token generation rate together, and intuitively obtaining the QOS capabilities of each node;

[0070] One of the technical effects of the above technical solution is: for nodes with the same token generation rate, the same token bucket management unit capacity, and a large proportion of token bucket dynamic capacity consumption, the service quality QOS of these nodes is the most difficult to guarantee, and it is necessary to give priority to reducing traffic allocation, so that the QOS capacity can be effectively scheduled. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is one of the implementation methods of a high-reliability data service system;

[0072] Figure 2 A schematic diagram of the operating principle of a high-reliability data service system

[0073] Figure 3 This is a schematic diagram of the operating status of multiple token buckets.

[0074] Figure 4 It is one of the implementation methods of a high-reliability data service system;

[0075] Figure 5 It is a schematic diagram of node status monitoring of a high reliability data service system;

[0076] Figure 6 It is a schematic diagram of a token generation node selection method;

[0077] Figure 7 A high reliability data service system implementation method. DETAILED DESCRIPTION

[0078] The content of this application is further described in detail below in conjunction with the accompanying drawings. It should be noted that the following is a description of the preferred embodiments of the present invention and does not constitute any limitation to the present invention. The description of the preferred embodiments of the present invention is only an illustration of the general principles of the present invention. The numbers such as "first", "second" and "A" and "B" involved in the present invention are only for the convenience of explanation and do not represent the order relationship in time or space. The combination of letters and numbers "TA", "TB" and "H" involved in the present invention are only for the convenience of explanation, and the specific meaning is determined by the specific words referred to.

[0079] like Figure 1 A high-reliability data service system includes multiple service access terminals, such as service A access terminal and service B access terminal in the figure, more than two token consumption node groups, the token consumption node group includes token consumption node group A and token consumption node group B, a high-reliability service quality management cluster, more than two storage nodes, such as storage node 11, storage node 12, storage node 13, storage node 21, storage node 22, storage node 23 in the figure; the token consumption node group is signal-connected to the high-reliability service quality management cluster; the token consumption node group includes a token generation node as shown in the figure token node 10 and token node 20; multiple token consumption nodes, such as token consumption node 11, token consumption node 12, token consumption node 13 in token consumption node group A; token consumption node 21, token consumption node 22, token consumption node 23 in token consumption node group B; as shown in the figure Figure 2, the token generation node includes a token generation module and a total token bucket; the token consumption node includes a local token bucket; the token consumption node periodically collects tokens from the total token bucket and stores them in the local token bucket; the token consumption node collects tokens from the local token bucket; the high reliability service quality management cluster is used to configure the speed at which the token generation module generates tokens; such as Figure 5 ,like Figure 6 The token generation node and the token consumption node report their own or other nodes' status to the high-reliability service quality management cluster. The high-reliability service quality management cluster establishes a node status table based on the status report. The high-reliability service quality management cluster republishes the updated node status table after the node status changes.

[0080] The token consumption node may be a WEB page server, the token consumption node group A corresponds to processing business A, and the token consumption node group B corresponds to processing business B.

[0081] The token consumption node may be a gateway server, the token consumption node group A corresponds to the data storage service for processing service A, and the token consumption node group B corresponds to the data storage service for processing service B.

[0082] In large-scale data service systems, different businesses require different network resources. By combining different numbers of servers into a token consumption node group, one server is responsible for the generation of tokens. Compared with the traditional method, each server manages its own token cylinder, it is possible to effectively control the issuance of tokens for a business through one token consumption node. Each business corresponds to multiple token consumption nodes, so there will be no situation where one business corresponds to one WEB page server, and when the WEB page server fails, the service of a business will not be interrupted.

[0083] By setting up independent token nodes for different businesses, the token consumption requirements of different businesses can be managed dynamically.

[0084] A high-reliability service quality management cluster is used to monitor the working status of each node in the token consumption node group. When a node fails, especially a token node fails, the network node status can be published in the first time through the node status table, which greatly improves the reliability of the system.

[0085] According to the above-mentioned high reliability data service system, the status report includes the remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

[0086] In large data centers and large databases, hundreds of businesses are running at the same time. Each business corresponds to a different token consumption node group. The operating status of the token nodes in each token consumption node group reflects the business volume and operating status of a business. Through the remaining token capacity ratio of the token nodes in the token consumption node group, the data center administrator can intuitively observe the operating health of each business. With one parameter, the operating health of the entire network can be observed.

[0087] According to the above-mentioned high-reliability data service system, the high-reliability service quality management cluster collects the operating parameters or operating status of the token consumption nodes in the token consumption node group. The high-reliability service quality management cluster sets the traffic weight in the node status table according to the operating status of the token consumption node. The token consumption nodes in each token consumption node group can obtain the token generation node of this token consumption group through HASH calculation based on the node status table and each token consumption node group and the traffic weight; the service access end selects the corresponding token consumption node to process the request based on the node status table, each token consumption node group and the traffic weight.

[0088] By configuring the traffic weight of the token consumption node business processing, servers with different processing capabilities can be coordinated to work together. Servers with strong capabilities can handle more business, while older servers can handle fewer businesses.

[0089] According to the above-mentioned high-reliability data service system, the operating parameters include network bandwidth and CPU type; the operating status includes transmission delay, CPU utilization, and network bandwidth ratio.

[0090] Although the remaining token capacity ratio of token-consuming nodes can be used to observe the overall operation status of the network, collecting information such as network bandwidth, CPU type, transmission delay, CPU utilization, and network bandwidth ratio can be used as a reference when analyzing the network status in detail.

[0091] According to the above-mentioned high-reliability data service system, token consumption node group A is used to process business A, and token consumption node group B is used to process business B.

[0092] like Figure 3 ,According to the above-mentioned high-reliability data service system, the high-reliability service quality management cluster outputs a visual operation status diagram, the visual operation status diagram includes the operation status of the token generation node, and the status report of the token generation node includes the remaining token capacity ratio.

[0093] like Figure 3 ,The above figure shows the proportion of remaining tokens in the token bucket. The proportion of remaining token capacity ,can be the proportion of remaining tokens in the total token bucket. The number of total token buckets is small, ,which directly corresponds to the business. Network administrators ,can intuitively observe the health of different business

[0094] The remaining token capacity ratio can be the ratio of the remaining tokens in the local token buckets of different services. If there is a problem with the ratio of the remaining tokens in the total token bucket of a service, such as the ratio becomes larger, the network administrator can check whether the configured number of token consumption nodes is insufficient through the ratio of the remaining tokens in the local token bucket.

[0095] According to the above-mentioned high-reliability data service system, the operating status of the token generation node includes the token generation rate and the remaining token capacity ratio.

[0096] like Figure 5 , Figure 6 The token consumption node or token generation node in the token consumption node group includes a heartbeat management module; the high reliability service quality management cluster includes a node status module; the heartbeat management module monitors the heartbeat status of other nodes, identifies nodes without heartbeat messages as faulty nodes, and reports the faulty nodes to the high reliability service quality management cluster. In traditional technologies, heartbeat monitoring runs in nodes of the entire network. The scale of heartbeat monitoring is large, and it is easy to have heartbeat monitoring islands and other phenomena. By forming a small heartbeat monitoring ring in the token consumption node group, the accuracy of heartbeat monitoring can be improved.

[0097] According to the above-mentioned high-reliability data service system, if the difference between the remaining token capacity ratios of token consumption node group A and token consumption node group B is greater than the set value, and the remaining token capacity ratio of token consumption node group A is higher than the remaining token capacity ratio of token consumption node group B, the high-reliability service quality management cluster deploys at least one token consumption node of token consumption node group A to token consumption node group B; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0098] like Figure 4 , the traffic demand of business A becomes smaller, while the traffic demand of business B becomes smaller. By deploying the token consumption node 13 in the token consumption node group A to the token consumption node group B, the processing capacity of business B can be improved. After the deployment, the token generation speed of the token generation node can be adjusted so that the proportion of remaining tokens in the adjusted token bucket runs between a reasonable 30% and 50%.

[0099] like Figure 7 In an actual physical server, there are token consumption nodes that run multiple services at the same time. According to the above high-reliability data service system, combined with the load status of the token consumption nodes and the remaining token capacity ratio of the token consumption node group, the high-reliability service quality management cluster can dynamically adjust the traffic weight of each token consumption node, generate and publish a new version of the node status table. When the network service traffic changes, the traffic weight of the node can be adjusted to achieve traffic scheduling between different services.

[0100] like Figure 7 Service node 31 runs the token node of service A, and service node 33 runs the token node of service B; service node 31, service node 32, and service node 33 all run the token consumption nodes of services A and B. By adjusting the traffic weights of the token consumption nodes of services A or B in different service nodes, the resource occupancy ratio of services A and B to the entire network can be changed.

[0101] According to the above-mentioned high-reliability data service system, if the remaining token capacity of the total token bucket is greater than the set value, the high-reliability service quality management cluster reduces or adjusts the weight of at least one token consumption flow associated with the token consumption node group to 0; the corresponding token consumption node reduces the processing of requests related to the token consumption node group. For example, if the remaining token capacity of a business is greater than 90%, it means that the current business flow demand is small. By reducing or adjusting the weight of a token consumption flow to 0, the token consumption node is withdrawn from the service and the server is trimmed.

[0102] According to the above high reliability data service system, if the remaining token capacity of the total token bucket is less than the set value, the high reliability service quality management cluster will increase the traffic weight of the token consumption node group in each token consumption node.

[0103] According to the above-mentioned high-reliability data service system, when the traffic weight of the token consumption node in the group of all token consumption nodes is 0, the token consumption node enters a low-power waiting state, which can save the power consumption of the service cluster.

[0104] According to the above-mentioned high-reliability data service system, when the proportion of the remaining token capacity of the total token bucket is less than the set value, the high-reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0105] By setting the token consumption node group C as the backup support group, the idle servers are centrally deployed to the group for management and control. In the backup support group, a certain number of services can be dynamically set to be in the power-on backup state, and other servers can be in the motherboard power-off state. The motherboard power-off state can greatly reduce the power consumption of the server. When the business demand increases, the server in the power-on backup state is first dispatched to the token consumption node group with demand. When the number of services in the power-on backup state drops to the set value, the server is gradually mobilized to turn on the motherboard power supply through the network card power-on mode, and gradually enters the backup support state. This method can greatly improve the efficiency of server operation. During low business time periods, some servers can be powered off and exited from service, greatly saving the power consumption of the data center.

[0106] According to the above-mentioned high-reliability data service system, when the remaining token capacity ratio of the total token bucket is greater than the set value, the high-reliability service quality management cluster deploys at least one token consumption node to the backup token consumption node group C; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node. After adjustment, by modifying the token generation speed of the token generation node, the adjusted remaining token capacity ratio can be brought into the normal monitoring range. When the business volume changes, the remaining token capacity ratio will fluctuate and enter an abnormal state, increasing the number of nodes in the token consumption node group to be adjusted again.

[0107] According to the above high reliability data service system, the high reliability service quality management cluster builds a token consumption node service flow distribution ratio table; the high reliability service quality management cluster periodically publishes the service flow distribution ratio table to the service access terminal. The service access terminal obtains the service flow distribution ratio table and can distribute the flow to nodes with different processing capabilities to dynamically balance the flow of each service node.

[0108] like Figure 2 , is a three-level token management method for managing a high-reliability service quality management cluster, a token generation node, and at least two token consumption nodes; a token generation node and at least one token consumption node form a token consumption node group; the high-reliability service quality management cluster is used as the first level to record token bucket configuration information, and the token bucket configuration information includes a token generation speed for setting the token generation speed of the token generation node; the token generation node, as the second level, includes a total token bucket, which generates a specified number of tokens within a specified time according to the token generation speed and places them in the total token bucket; the token consumption node is used as the third level, including a local token bucket, which collects tokens from the total token bucket of the token generation node in the same group and places them in the local token bucket; the token consumption node receives a service request and collects tokens from the local token bucket; the token generation node and the token consumption node report the status report of themselves or other nodes to the high-reliability service quality management cluster, and the high-reliability service quality management cluster establishes a node status table based on the status report; the high-reliability service quality management cluster periodically publishes the node status table through a broadcast channel; the node status table includes the token generation node and the token consumption node members of the token consumption node group.

[0109] Through the three-level token management model, the traditional token bucket technology is decomposed into three levels: token bucket performance parameter management level, token generation level and token consumption level. In large-scale service clusters, the token bucket function is clearly divided, the function of a single node is single, the operation efficiency is high, and the fault tolerance is good. In the token consumption level, multiple nodes share the business and obtain tokens on a token generation node, making the token management and control intuitive and efficient.

[0110] According to the above-mentioned high-reliability data service method, when a token generation node fails, the token consumption node selects a node as a new token generation node from among the token consumption nodes based on the updated node status table, and the high-reliability service quality management cluster publishes the updated node status table through the broadcast channel.

[0111] According to the above high reliability data service method, the status report includes the remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

[0112] According to the above-mentioned high-reliability data service method, the token consumption node groups are token consumption node group A and token consumption node group B, the remaining token capacity ratio of token consumption node group A is higher than the remaining token capacity ratio of token consumption node group B, and at least one token consumption node of token consumption node group A is deployed to token consumption node group B; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0113] According to the above-mentioned high-reliability data service method, if the proportion of the remaining token capacity of the total token bucket is greater than the set value, at least one token consumption node of the token consumption node group is deployed to the backup token consumption node group C; the token consumption node of the backup token consumption node group C enters a low-power waiting state; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0114] According to the above-mentioned high-reliability data service method, the proportion of the remaining token capacity of the total token bucket is less than the set value, and the high-reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node.

[0115] According to the above high-reliability data service method, the high-reliability service quality management cluster builds a token consumption node service flow distribution ratio table; the high-reliability service quality management cluster periodically publishes the service flow distribution ratio table to the service access terminal.

[0116] Although the present invention is illustrated and described according to the preferred embodiment and several alternatives, the invention is not limited by the specific description in this specification. Other additional replacement or equivalent components can also be used to practice the present invention.

Claims

1. A high reliability data service system, characterized in that: It includes more than two token consumption node groups, the token consumption node groups include a token consumption node group A and a token consumption node group B, a high reliability service quality management cluster, and more than two storage nodes; The token consumption node group is connected to the high reliability service quality management cluster signal; The token consumption node group includes a token generation node and at least one token consumption node; the token generation node includes a token generation module and a total token bucket; The token consumption node includes a local token bucket; The token consumption node periodically obtains tokens from the total token bucket and stores them in a local token bucket; The token consumption node obtains tokens from the local token bucket; The high reliability service quality management cluster is used to configure the speed at which the token generation module generates tokens; The token generation node and the token consumption node report their own or other nodes' status to the high reliability service quality management cluster, and the high reliability service quality management cluster establishes a node status table based on the status report; the high reliability service quality management cluster republishes the updated node status table after the node status changes.

2. The high reliability data service system according to claim 1, characterized in that: The status report includes a remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

3. The high reliability data service system according to claim 2, characterized in that: The high-reliability service quality management cluster collects the operating parameters or operating status of the token consumption nodes in the token consumption node group. The high-reliability service quality management cluster sets the traffic weight in the node status table according to the operating status of the token consumption node. The token consumption nodes in each token consumption node group can obtain the token generation node of the token consumption group through HASH calculation based on the node status table and the traffic weight; The service access end selects the corresponding token consumption node to process the request based on the node status table, each token consumption node group and the traffic weight.

4. The high reliability data service system according to claim 3, characterized in that: The operating parameters include network bandwidth and CPU type; the operating status includes transmission delay, CPU utilization, and network bandwidth ratio; Also includes any one or more of the following features: Feature TA1: Token consumption node group A is used to process business A, and token consumption node group B is used to process business B; Feature TA2: The high reliability service quality management cluster outputs a visual operation status diagram, the visual operation status diagram includes the operation status of the token generation node, and the status report of the token generation node includes the remaining token capacity ratio; Feature TA3: The operating status of the token generation node includes the token generation rate and the remaining token capacity ratio; Feature TA4: A token consumption node or a token generation node in a token consumption node group includes a heartbeat management module; The high reliability service quality management cluster includes a node status module; The heartbeat management module monitors the heartbeat status of other nodes, identifies nodes without heartbeat messages as faulty nodes, and reports the faulty nodes to the high-reliability service quality management cluster.

5. The high reliability data service system according to claim 4, characterized in that: Also includes any one or more of the following features: Feature TB1: the difference between the remaining token capacity ratios of the token consumption node group A and the token consumption node group B is greater than a set value, the remaining token capacity ratio of the token consumption node group A is higher than the remaining token capacity ratio of the token consumption node group B, and the high reliability service quality management cluster deploys at least one token consumption node of the token consumption node group A to the token consumption node group B; after the deployment, the high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node; Feature TB2: Based on the load status of the token consumption nodes and the remaining token capacity ratio of the token consumption node group, the high-reliability service quality management cluster can dynamically adjust the traffic weight of each token consumption node, generate and publish a new version of the node status table.

6. The high reliability data service system according to claim 4, characterized in that: Also includes any one or more of the following features: Feature TC1: The remaining token capacity of the total token bucket accounts for a larger proportion than a set value, and the high reliability service quality management cluster reduces or adjusts the weight of at least one token consumption traffic associated with the token consumption node group to 0; the corresponding token consumption node reduces the processing of requests related to the token consumption node group; Feature TC2: the remaining token capacity of the total token bucket accounts for less than the set value, and the high-reliability service quality management cluster will increase the traffic weight of the token consumption node group in each token consumption node; Feature TC3: When the traffic weight of the token consumption node in all token consumption node groups is 0, it enters a low power consumption waiting state; Feature TC4: When the remaining token capacity of the total token bucket accounts for less than a set value, the high reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node; Feature TC5: the remaining token capacity of the total token bucket accounts for a larger proportion than a set value, and the high-reliability service quality management cluster deploys at least one token consumption node to the backup token consumption node group C; The high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node after deployment; Feature TC6: High reliability service quality management cluster builds token consumption node business traffic distribution ratio table; The high-reliability service quality management cluster periodically publishes the service traffic distribution ratio table to the service access terminal.

7. A high reliability data service method, characterized in that: Used to manage a high-reliability quality of service management cluster, a token generation node, and at least two token consumption nodes; A token generating node and at least one token consuming node form a token consuming node group; A high reliability service quality management cluster, used to record token bucket configuration information, wherein the token bucket configuration information includes a token generation speed for setting a token generation speed of a token generation node; The token generation node includes a total token bucket, which generates a specified number of tokens within a specified time according to the token generation speed and places them into the total token bucket; The token consumption node, including the local token bucket, takes tokens from the total token bucket of the token generation node in the same group and places them in the local token bucket; The token consumption node receives the service request and collects the token from the local token bucket; The token generation node and the token consumption node report the status of themselves or other nodes to the high reliability service quality management cluster, and the high reliability service quality management cluster establishes a node status table according to the status report; the high reliability service quality management cluster periodically publishes the node status table through a broadcast channel; The node status table includes token generating nodes and token consuming node members of the token consuming node group.

8. The high reliability data service method according to claim 7, characterized in that: When the token generation node fails, the token consumption node selects a node from the token consumption nodes as a new token generation node based on the updated node status table, and the high-reliability service quality management cluster publishes the updated node status table through a broadcast channel.

9. The high reliability data service method according to claim 8, characterized in that: The status report includes a remaining token capacity ratio; the remaining token capacity ratio is equal to the ratio of the current number of tokens in the token bucket to the capacity of the token bucket management unit.

10. The high reliability data service method according to claim 9, characterized in that: Also includes any one or more of the following features: Feature TD1: the token consumption node groups are token consumption node group A and token consumption node group B, the remaining token capacity ratio of token consumption node group A is higher than the remaining token capacity ratio of token consumption node group B, at least one token consumption node of token consumption node group A is deployed to token consumption node group B; after deployment, the high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node; Feature TD2: Based on the load status of the token consumption node and the remaining token capacity ratio of the token consumption node group, the high-reliability service quality management cluster can dynamically adjust the traffic weight of each token consumption node, generate and publish a new version of the node status table; Feature TD3: When the remaining token capacity of the total token bucket accounts for a larger proportion than a set value, the high reliability service quality management cluster reduces or adjusts to 0 the weight of at least one token consumption traffic associated with the token consumption node group; the corresponding token consumption node reduces the processing of requests related to the token consumption node group; Feature TD4: the remaining token capacity of the total token bucket accounts for less than the set value, and the high reliability service quality management cluster will increase the traffic weight of the token consumption node group in each token consumption node; Feature TD5: When the traffic weight of the token consumption node in all token consumption node groups is 0, the token consumption node enters a low power consumption waiting state; Feature TD6: the remaining token capacity of the total token bucket accounts for a larger proportion than a set value, and at least one token consumption node of the token consumption node group is deployed to the backup token consumption node group C; The token consumption nodes of the backup token consumption node group C enter a low power consumption waiting state; The high reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node after deployment; Feature TD7: When the remaining token capacity of the total token bucket accounts for less than a set value, the high-reliability service quality management cluster deploys at least one token consumption node of the backup token consumption node group C to the token consumption node group; after deployment, the high-reliability service quality management cluster modifies or does not modify the token generation speed of the token generation node; Feature TD8: High reliability service quality management cluster builds token consumption node business traffic distribution ratio table; The high-reliability service quality management cluster periodically publishes the service traffic distribution ratio table to the service access terminal.

Images