Intelligent test support software platform and resource scheduling method

Abstract

The invention discloses an intelligent test support software platform and a resource scheduling method. The resource scheduling method comprises the following steps: detecting abnormal events of an infrastructure as a service layer, a joint test platform as a service layer and / or a joint test service as a service layer; the event level of the abnormal event is obtained, and the event level comprises a first-level abnormal event, a second-level abnormal event and a third-level abnormal event; the first-level abnormal event is solved through resource scheduling, the second-level abnormal event is solved through resource scheduling and joint test parameter adjustment, and the third-level abnormal event is solved through joint test parameter adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of simulation experiments, and in particular to an intelligent experiment support software platform and a resource scheduling method. Background Art

[0002] Simulation experiments do not have the necessary equipment for experiments in the ordinary sense, but use simulation software on computers to simulate the effects of reality. Resource scheduling technology is one of the basic technologies of simulation experiments. Whether it is physical resources or virtual resources, the rationality and correctness of resource scheduling are the performance indicators that simulation experiments have always pursued. This article refers to the resources involved in simulation experiments as experimental resources. At present, a single algorithm is usually used for experimental resource scheduling in China, which makes it difficult to cope with the complex situation of joint experiments. Among them, joint experiments refer to the use of two or more experimental resources to conduct simulation experiments on the same research object.

[0003] like Figure 1 As shown, the test resource scheduling in the prior art takes the joint test middleware as the core, and connects the test resource application dedicated to the test range and the general joint test utility program of the joint test software support platform through the joint test middleware as the common intermediate layer. Although it can effectively improve the software reuse rate and development efficiency of the joint test program (shorten the test deployment cycle), problems such as cumbersome application, insufficient system capability, and poor anti-interference ability are still prominent, and it lacks dynamic scheduling and control technology, which makes it difficult to cope with the complex situation of joint tests.

[0004] Therefore, in a complex and changeable software operating environment, completing the joint test with high quality is the technical problem to be solved by the present invention. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent test support software platform and resource scheduling method, which uses two methods: abnormal self-reporting and system active detection to automatically capture abnormal events. On this basis, the abnormal events are graded and adaptively pre-processed according to the degree of impact of the abnormal events. Among them, the processing of important abnormal events uses a greedy algorithm to achieve dynamic adaptive scheduling of resources based on health management.

[0006] According to one aspect of the present invention, at least one embodiment provides a resource scheduling method applicable to an intelligent test support software platform, comprising: detecting abnormal events at the infrastructure as a service layer, the joint test platform as a service layer and / or the joint test service as a service layer; obtaining event levels of the abnormal events, wherein the event levels include first-level abnormal events, second-level abnormal events and third-level abnormal events; resolving the first-level abnormal events through resource scheduling, resolving the second-level abnormal events through resource scheduling and adjusting joint test parameters, and resolving the third-level abnormal events by adjusting joint test parameters.

[0007] According to another aspect of the present invention, at least one embodiment also provides an intelligent test support software platform, including: a processor, suitable for implementing various instructions; and a memory, suitable for storing multiple instructions, wherein the instructions are suitable for being loaded and executed by the processor: the resource scheduling method developed by the present invention.

[0008] According to another aspect of the present invention, at least one embodiment further provides a distributed management and control system, comprising: an intelligent test support software platform developed by the present invention and multiple resources, wherein the resources include virtual machines, virtual containers and / or network resources.

[0009] According to another aspect of the present invention, at least one embodiment further provides a computer-readable non-volatile storage medium storing computer program instructions, and when the computer executes the program instructions, it executes: the resource scheduling method developed by the present invention.

[0010] The present invention adopts the above-mentioned method, through health management methods such as exception handling, uses the test resource network topology to obtain the network distance, and uses the greedy algorithm to realize the dynamic allocation and adaptive scheduling of test resources. Compared with the single resource scheduling method of the prior art, it can ensure that the expected joint test content can still be completed smoothly in the event of an abnormal event, taking into account health monitoring and adaptive decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 It is a schematic diagram of a joint test software support platform in the prior art;

[0013] Figure 2 is a schematic diagram of a distributed management and control system according to an embodiment of the present invention;

[0014] Figure 3 is a schematic diagram of an intelligent test support software platform according to an embodiment of the present invention;

[0015] Figure 4 is a flow chart of a joint test method according to an embodiment of the present invention;

[0016] Figure 5 is a flow chart of a resource scheduling method according to an embodiment of the present invention;

[0017] Figure 6 4 is a schematic diagram of a server cluster network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Joint tests often involve test resources in different regions and with different compositions, and the test operation environment is complex. The current test resource scheduling technology using a single algorithm is difficult to cope with the complex situation of joint tests and does not have the ability to deal with abnormal situations. As a result, the test resource scheduling technology using a single algorithm is limited in actual joint test applications.

[0020] In view of the fact that the research on adaptive test resource scheduling has just started, the present invention starts from the actual needs and aims at the complex and changeable software running environment of the joint test. At least one embodiment provides a distributed control system, such as Figure 2 As shown, it includes: an intelligent test support software platform 100 and multiple resources 200. The distributed control system uses the joint test software support platform (referred to as "intelligent test support software platform" in the present invention) as a common middle layer to connect the test resource application dedicated to the test range and the joint test utility program, and develops a set of adaptive test resource dynamic scheduling control technology (or "test resource adaptive scheduling technology based on health management", referred to as "resource scheduling method" in the present invention) for the intelligent test support software platform 100, so as to ensure that the environment of the joint test (mainly test resources) is adaptively adjusted in the event of an abnormal event as much as possible, so that the expected joint test content can still be completed smoothly and with high quality.

[0021] The intelligent test support software platform 100, such as Figure 3As shown, it includes: a processor 301, suitable for implementing various instructions; and a memory 303, suitable for storing multiple instructions, and the instructions are suitable for being loaded and executed by the processor 301: the resource scheduling method developed by the present invention (this article will be described in detail later). Specifically, Figure 4 As shown, when simulation experimenters conduct joint experiments, the intelligent experiment support software platform 100 sequentially performs experiment resource preparation, experiment scenario design, experiment environment construction, experiment task management and experiment result analysis. Among them, the intelligent experiment support software platform 100 can be loaded and executed by the processor 301 in the experiment scenario design, experiment environment construction and experiment task management: the resource scheduling method developed by the present invention.

[0022] The processor 301 may be any applicable processor, such as a central processing unit, a microprocessor, an embedded processor, etc., and may adopt an architecture such as X86 or ARM. The memory 303 may be any applicable storage device, such as a non-volatile storage device, including but not limited to a magnetic storage device, a semiconductor storage device, an optical storage device, etc., and may be arranged as a single storage device, a storage device array, or a distributed storage device, which is not limited in the embodiments of the present invention.

[0023] It can be understood by those skilled in the art that the structure of the intelligent test support software platform 100 is only for illustration and does not limit the structure of the device. Figure 3 The transmission device is used to receive or send data via a network. In one embodiment, the transmission device is a radio frequency (RF) module, which is used to communicate with multiple resources 200 wirelessly.

[0024] The multiple resources 200, as test software tools and / or test basic equipment, support the intelligent test support software platform 100, and the test software tools and / or test basic equipment constitute a server cluster network. The resources mainly refer to joint test resources, including IT infrastructure, virtual machines, virtual containers and / or network resources, etc.; in addition, in the joint test scenario of the shooting range, test resources such as aircraft and ships and virtual construction resources can also be regarded as joint test resources.

[0025] It should be noted that from the perspective of the intelligent test support software platform, joint test resources can be further divided into joint test resources at the infrastructure as a service layer, joint test platform as a service layer and joint test service as a service layer. Joint test resources at the infrastructure as a service layer refer to IaaS (Infrastructure as a Service) joint test resources, that is, IT infrastructure is provided to the outside world as a service through the network. The service provided to users is the use of all facilities, including processing, storage, network and other basic computing resources; joint test platform as a service layer joint test resources refer to PaaS (Platform as a Service) joint test resources, and the service provided to users is to deploy applications developed or acquired by users to the infrastructure; joint test service as a service layer joint test resources refer to SaaS (Software as a Service) joint test resources, and the service provided to users is applications running on the infrastructure, which users can access through the client interface on various devices.

[0026] The present invention adopts the above-mentioned method and dynamically schedules joint test resources such as virtual machines, virtual containers and / or network resources through an intelligent test support software platform, and has strong anti-interference ability. Even if an abnormality occurs in a joint test resource during the joint test, the expected joint test content can still be completed smoothly and with high quality.

[0027] In the above operating environment, at least one embodiment of the present invention provides a resource scheduling method suitable for being loaded and executed by the processor 301 of the intelligent test support software platform 100 in the test scenario design, test environment construction and test task management and control links, such as Figure 5 It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here. The resource scheduling method may include the following steps:

[0028] Step S501, detecting abnormal events of the infrastructure as a service layer, the joint test platform as a service layer and / or the joint test service as a service layer;

[0029] Step S503, obtaining the event level of the abnormal event, wherein the event level includes a first level abnormal event, a second level abnormal event and a third level abnormal event;

[0030] Step S505, solving the first level abnormal events through resource scheduling, solving the second level abnormal events through resource scheduling and adjusting the joint test parameters, and solving the third level abnormal events through adjusting the joint test parameters.

[0031] The present invention improves the fault tolerance of complex test environments in the above manner. Abnormal events in the infrastructure as a service layer, the joint test platform as a service layer, and the joint test software as a service layer are captured through an abnormal event detection algorithm. Joint test parameter adjustment and resource dynamic scheduling and other processing measures are performed according to the impact of the abnormal events. Adaptive adjustment of the joint test environment (mainly referring to test resources) is achieved, so that even if an abnormal event occurs in a certain test resource, the expected joint test content can be completed smoothly and with high quality.

[0032] In step S501, abnormal events of the infrastructure as a service layer, the joint test platform as a service layer and / or the joint test service as a service layer are detected. There are two main methods for detecting abnormal events in the present invention: passive detection and active detection. Passive detection: that is, receiving abnormal events reported by the IaaS infrastructure as a service layer, the PaaS joint test platform as a service layer and / or the SaaS joint test service as a service layer. Active detection, that is, establishing multiple checkpoints on the IaaS infrastructure as a service layer, the PaaS joint test platform as a service layer and / or the SaaS joint test service as a service layer; sending check messages or update messages to multiple checkpoints, receiving multiple resource states reported by multiple checkpoints through status messages and / or timeout messages, wherein the resource states include the input state of the joint test resources, the output state of the joint test resources and / or the occupancy of the joint test resources; if the intelligent test support software platform is not in a globally consistent state, obtaining abnormal events of the IaaS infrastructure as a service layer, the PaaS joint test platform as a service layer and / or the SaaS joint test service as a service layer according to the multiple resource states. The status message encapsulates the test resource status information detected by the checkpoint, and the timeout message encapsulates the information that the defect occurs in the checked test resource.

[0033] It can be seen that the passive detection of the present invention mainly relies on the resources in the joint test environment to report abnormal events on their own, and the intelligent test support software platform detects abnormal events by receiving the abnormal events.

[0034] It can be seen that the active detection of the present invention is to obtain abnormal events by actively checking the status of each test resource by the intelligent test support software platform. In order to adapt to the abnormal event detection of complex and large-scale joint test environments, the present invention sets checkpoints on the joint test resources to obtain information such as the input, output, and resource occupancy of the joint test resources; at specified time intervals, different checkpoints uniformly report the resource status; based on the resource status reported by different checkpoints, it is determined whether the current distributed control system is in a globally consistent state. If the distributed control system is not in a consistent state, the abnormal events are sorted out based on the resource status reported by the checkpoint.

[0035] In step S503, the event level of the abnormal event is obtained, wherein the event level includes first-level abnormal events, second-level abnormal events, and third-level abnormal events. For example, the event level of abnormal events of the Infrastructure as a Service layer (IaaS) is obtained, wherein abnormal events of the Infrastructure as a Service layer (IaaS) include incorrect network message format, incorrect network message content, network transmission packet loss, network delay jitter, network unreachability, insufficient virtual machine memory, insufficient virtual machine CPU, virtual machine crash and / or storage resource failure, virtual machine crash and storage resource failure are first-level abnormal events, insufficient virtual machine memory and insufficient virtual machine CPU are second-level abnormal events, incorrect network message format, incorrect network message content, network transmission packet loss and network delay jitter are third-level abnormal events; the event level of abnormal events of the Joint Test Platform as a Service layer (PaaS) is obtained, wherein abnormal events of the Joint Test Platform as a Service layer (PaaS) include inability to deploy joint test resources, unsuccessful deployment of joint test resources, inability to deploy joint test data, etc. The following abnormal events are classified as level 1 abnormal events: failure to transmit back part of the joint test data, data collection service crash, unsuccessful issuance of control commands, joint test execution service crash, inability to obtain joint test resource monitoring data and / or inability to find registered joint test resources. The data collection service crash and the joint test execution service crash are level 1 abnormal events; failure to deploy joint test resources, inability to transmit back part of the joint test data and unsuccessful issuance of control commands are level 2 abnormal events; unsuccessful deployment of joint test resources, loss of part of the joint test data, inability to obtain joint test resource monitoring data and inability to find registered joint test resources are level 3 abnormal events; obtain the event level of the joint test service as a service layer (SaaS) abnormal event, among which the joint test service as a service layer (SaaS) abnormal event includes timeout of the joint test service external interface response, and timeout of the joint test service external interface response is level 1 abnormal event.

[0036] It can be seen that the present invention divides abnormal events into three categories: infrastructure as a service layer (IaaS) abnormalities, joint test platform as a service layer (PaaS) abnormalities, and joint test service as a service layer (SaaS) abnormalities, and positions the abnormalities into three levels according to the impact of the abnormalities (the lower the level, the higher the urgency). It should also be noted that the present invention can further subdivide the first-level abnormal events, the second-level abnormal events, and the third-level abnormal events, such as subdividing the third-level abnormal events into three levels according to the urgency, as shown in Table 1.

[0037] Table 1 Classification of abnormal events

[0038]

[0039]

[0040] In step S505, the first level abnormal event is solved by resource scheduling, the second level abnormal event is solved by resource scheduling and adjusting the joint test parameters, and the third level abnormal event is solved by adjusting the joint test parameters. For example, solving the third level abnormal event by adjusting the joint test parameters may include: solving the third level abnormal event by resending the joint test parameters for multiple deployments. For example, solving the second level abnormal event by resource scheduling and adjusting the joint test parameters may include: solving the second level abnormal event by adjusting the joint test parameters; if the second level abnormal event is not solved by adjusting the joint test parameters, solving the second level abnormal event by resource scheduling. For example, solving the first level abnormal event by resource scheduling may include: obtaining the network distance of multiple first resources connected to the intelligent test support software platform in the network topology, wherein the first resource includes a virtual machine, a virtual container and / or a network resource; sorting multiple second resources with the same network distance in the multiple first resources according to the error probability, and obtaining multiple second resources with the lowest error rate; using a greedy algorithm to traverse the multiple second resources according to the principle of network distance from small to large, and finding the target resource in the multiple second resources to solve the first level abnormal event, wherein solving the first level abnormal event includes resource copying, task migration and / or resource dynamic application.

[0041] That is to say, the present invention formulates different processing methods according to the event level of the abnormal event. (1) For the third level abnormal event (such as the 3-5 level abnormal events in Table 1), it is processed by adjusting the joint test parameters. For example, when the joint test resource deployment is unsuccessful (only a part of the joint test script is successfully deployed to the joint test tool), it can be deployed multiple times by adjusting the deployment retransmission parameters. The adjustment strategy for the joint test parameters does not involve complex technology, and it is only necessary to refine the abnormal event and formulate a detailed adjustment strategy. (2) For the second level abnormal event (such as the level 2 abnormal event in Table 1), it can be solved by adjusting the joint test parameters first. If it cannot be solved by adjusting the joint test parameters, it can be solved by reallocating resources. (3) For the first level abnormal event (such as the level 1 abnormal event in Table 1), it is directly solved by dynamically allocating resources. For the dynamic allocation of resources, it involves the selection and replication of test resources. It is necessary to select the optimal resource allocation strategy under the condition of limited test resources: through resource copying, task migration, dynamic resource application and other means, dynamically allocate test resources such as virtual machines, virtual containers or networks that can carry the original business.

[0042] It should be noted that the present invention solves the problem of first-level abnormal events or second-level abnormal events through resource scheduling, and mentions the concept of network distance. The network distance is the distance between any physical host in the distributed joint test environment of the distributed management and control system in the network topology, which reflects whether the two nodes are "close to each other" and the size of the bandwidth between them. In theory, the network distance cannot be too large, because if the distance is too large, the time cost of transmitting data will increase, but the network distance cannot be too small. If the distance is too small, the probability of simultaneous failure of the host will increase. In order to make it clear to those skilled in the art how the present invention solves abnormal events through resource scheduling, the following uses the use of dynamic allocation of virtual machines to migrate indicator calculation tools as an example.

[0043] like Figure 6 As shown, a typical distributed management and control system connects a server cluster network, which is composed of two local area networks. The internal part of the local area network is composed of server racks. The transmission speed between servers in one rack is very fast, while the transmission speed between servers in different racks may be slower, and the speed of servers between different local area networks may be even slower (in other words, the present invention refers to servers as physical machines). For example, the bandwidth between physical machines and whether they are in the same physical location are used as the standard for measuring network distance, that is, for the following four scenarios, the four network distances are expressed as follows, where the network distance can be expressed by a method of decreasing bandwidth in sequence, and any physical machine node can be expressed as d1 / r1 / n1, where d1 is local area network 1, r1 is rack 1, and n1 is physical machine 1.

[0044] (1) Different virtual machines in the same physical machine, distance( / d1 / r1 / n1, / d1 / r1 / n1) = 0;

[0045] (2) For virtual machines on different physical machines on the same rack, distance( / d1 / r1 / n1, / d1 / r1 / n2) = 1;

[0046] (3) For virtual machines on different racks in the same LAN, distance( / d1 / r1 / n1, / d1 / r2 / n3) = 2;

[0047] (4) For virtual machines in different LANs, distance( / d1 / r1 / n1, / d2 / r2 / n3)=3.

[0048] In view of the fact that the purpose of dynamically allocating virtual machines is to meet the needs of simulation experiment operation and improve fault tolerance, but the distributed copy of virtual machines increases the cost of network transmission, in order to balance the fault tolerance and transmission cost issues, the present invention allocates virtual machines based on network distance. Through the network distance, among the nodes that also meet the joint test conditions, the node with the optimal network distance is preferentially selected through a greedy algorithm to achieve rapid resource adjustment during the execution of the joint test: 1) A physical machine with a distance of 2 is selected from the local area network, and then the selected physical machines are sorted according to the error probability, and the machine with the lowest error rate is selected. Then a virtual machine is created in the physical machine, and the virtual machine is remotely copied; 2) If the physical machine with a distance of 2 cannot create available resources, the physical machines with network distances of 2, 3, and 1 are selected in turn to create virtual machines.

[0049] Compared with the existing resource scheduling technology, the present invention: 1) needs to set time points to mark which joint test data are affected by the dynamic allocation of scheduling resources according to the dynamic allocation process, and these data may be invalid data in the subsequent data analysis process; 2) The dynamic allocation of test resources takes into account the special needs of joint test resources, such as the indicator calculation tool has high requirements on the migration overhead and reliability of resources, while some fault injection tools have strict requirements on the network topology.

[0050] Through the above-mentioned method, the present invention can automatically capture abnormal events through the abnormal event detection algorithm. On this basis, according to the impact degree of the abnormal events, various adaptive scheduling methods such as joint test parameter adjustment and dynamic resource scheduling are proposed. Therefore, even if an abnormal event occurs in a certain test resource, the expected joint test content can be completed smoothly and with high quality.

[0051] At least one embodiment of the present invention further provides a computer-readable non-volatile storage medium that stores computer program instructions. When the computer executes the program instructions, the resource scheduling method developed by the present invention is executed.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A resource scheduling method, applicable to an intelligent test support software platform, characterized in that: include: Detect abnormal events at the infrastructure as a service layer, the joint test platform as a service layer, and / or the joint test service as a service layer; Acquire the event level of the abnormal event, wherein the event level includes a first-level abnormal event, a second-level abnormal event, and a third-level abnormal event; The first level abnormal events are resolved through resource scheduling, the second level abnormal events are resolved through resource scheduling and adjustment of joint test parameters, and the third level abnormal events are resolved through adjustment of joint test parameters.

2. The resource scheduling method according to claim 1, characterized in that: Detection of abnormal events at the infrastructure as a service layer, the joint test platform as a service layer, and / or the joint test service as a service layer includes: Receive abnormal events reported by the infrastructure as a service layer, the joint test platform as a service layer and / or the joint test service as a service layer.

3. The resource scheduling method according to claim 1, characterized in that: Detection of abnormal events at the infrastructure as a service layer, the joint test platform as a service layer, and / or the joint test service as a service layer includes: Establishing multiple checkpoints at the infrastructure as a service layer, the joint test platform as a service layer, and / or the joint test service as a service layer; Receiving multiple resource states reported by the multiple checkpoints through status messages and / or timeout messages, wherein the resource states include input states of joint test resources, output states of joint test resources and / or occupancy of joint test resources; If the distributed management and control system is not in a globally consistent state, abnormal events of the infrastructure as a service layer, the joint test platform as a service layer and / or the joint test service as a service layer are obtained according to the multiple resource states.

4. The resource scheduling method according to claim 1, characterized in that: Obtaining the event level of the abnormal event includes: Obtaining the event level of abnormal events at the infrastructure as a service layer, wherein the abnormal events at the infrastructure as a service layer include incorrect network message format, incorrect network message content, network transmission packet loss, network delay jitter, network unreachability, insufficient virtual machine memory, insufficient virtual machine CPU, virtual machine crash and / or storage resource failure, wherein the virtual machine crash and the storage resource failure are first-level abnormal events, the insufficient virtual machine memory and the insufficient virtual machine CPU are second-level abnormal events, and the incorrect network message format, incorrect network message content, network transmission packet loss and the network delay jitter are third-level abnormal events; Obtaining the event level of the abnormal event of the joint test platform as a service layer, wherein the abnormal event of the joint test platform as a service layer includes the inability to deploy joint test resources, unsuccessful deployment of joint test resources, inability to transmit back joint test data, loss of part of the joint test data, crash of the data acquisition service, unsuccessful issuance of the control command, crash of the joint test execution service, inability to obtain joint test resource monitoring data and / or inability to find the registered joint test resources, the crash of the data acquisition service and the crash of the joint test execution service are first-level abnormal events, the inability to deploy joint test resources, the inability to transmit back joint test data and the unsuccessful issuance of the control command are second-level abnormal events, the unsuccessful deployment of joint test resources, the loss of part of the joint test data, the inability to obtain joint test resource monitoring data and the inability to find the registered joint test resources are third-level abnormal events; The event level of the joint test service as a service layer abnormal event is obtained, wherein the joint test service as a service layer abnormal event includes a joint test service external interface response timeout, and the joint test service external interface response timeout is a first-level abnormal event.

5. The resource scheduling method according to claim 1, characterized in that: Resolving the first level of abnormal events through resource scheduling includes: Acquire network distances of a plurality of first resources connected to the intelligent test support software platform in a network topology, wherein the resources include virtual machines, virtual containers and / or network resources; sorting the plurality of second resources at the same network distance according to the error probability respectively, and obtaining a plurality of second resources with the lowest error rate; A greedy algorithm is used to traverse the multiple second resources according to the principle of network distance from small to large, and a target resource among the multiple second resources is found to solve the first-level abnormal event, wherein solving the first-level abnormal event includes resource copying, task migration and / or dynamic resource application.

6. The resource scheduling method according to claim 1, characterized in that: The second level of abnormal events can be solved by resource scheduling and adjusting joint test parameters, including: Resolve the second level abnormal events by adjusting the joint test parameters; If the second-level abnormal event is not resolved by adjusting the joint test parameters, the second-level abnormal event is resolved by resource scheduling.

7. The resource scheduling method according to claim 1, characterized in that: Solving the third level abnormal events by adjusting the joint test parameters includes: The third level abnormal event is resolved by resending the joint test parameters and performing multiple deployments.

8. An intelligent test support software platform, comprising: a processor adapted to implement the instructions; And a memory, suitable for storing multiple instructions, wherein the instructions are suitable for being loaded and executed by a processor: a resource scheduling method as described in any one of claims 1-7.

9. A distributed management and control system, comprising: The intelligent test support software platform and multiple resources as described in claim 8, wherein the resources include virtual machines, virtual containers and / or network resources.

10. A computer-readable non-volatile storage medium storing computer program instructions, wherein when a computer executes the program instructions, the resource scheduling method according to any one of claims 1 to 7 is executed.

Images