WO2015067144A1

WO2015067144A1 - Method and apparatus for software disposition

Info

Publication number: WO2015067144A1
Application number: PCT/CN2014/090021
Authority: WO
Inventors: 王骅
Original assignee: 华为技术有限公司
Priority date: 2013-11-08
Filing date: 2014-10-31
Publication date: 2015-05-14
Also published as: CN103580923B; CN103580923A

Abstract

A method and apparatus for software deployment are disclosed in the present invention. The method includes: obtaining communication topology information of a cloud computing system; according to the communication topology information, selecting a host computer from a first host computer set for deploying a virtual machine for a software system, wherein, the first host computer set is a set of host computers of the virtual machine for the software system to be disposed in the cloud computing system. The method and apparatus for software deployment of the present invention can save communication bandwidth and reduce cost and power consumption.

Description

Software deployment method and device

The present application claims priority to Chinese Patent Application No. 20131055631, the entire disclosure of which is hereby incorporated by reference.

Technical field

The present invention relates to the field of information technology and, more particularly, to a method and apparatus for software deployment.

Background technique

In current cloud computing systems, software automation deployment is one of the biggest highlights of cloud computing. The so-called software automation deployment means that the system can automatically select the hardware host, and automatically configure and start the virtual machine on the selected host according to the requirements of the application software, and then automatically install the application software on the virtual machine, the entire process does not need to be Manual intervention. The system can also automatically repeat the above process after the system load increases, so that more virtual machines and application software can process the service; or when the system load is reduced, the reverse process of the above process is automatically performed, and the virtual machine for processing the service is reduced. And the number of application software, so that some physical hosts can be released to enter a power-saving state, or release resources for use by other applications.

The application software running on the cloud is generally distributed application software, that is, there are multiple software modules providing services at the same time, and there must be communication traffic between the processing modules of the same application software. When deploying software, the current cloud computing management system mainly relies on configuring enough hardware to ensure network bandwidth. For example, when deploying application software A and application software B from host 1 to host 10, the cloud management system only distributes application software A and application software B according to established rules, such as all host load averaging or integrated deployment to a small number of hosts. The processing module is deployed on the host 1 to the host 10.

Existing software deployment methods may result in discrete deployment results, resulting in most of the communication traffic between the same application software distributed modules being forwarded by multiple Ethernet switches, resulting in over-provisioning of network devices, which is not conducive to cost savings and energy consumption. Will also bring more delays. When the system exceeds a certain rule At the time of modulo, the traffic of the core switch may exceed the maximum transmission bandwidth that the prior art can achieve.

Summary of the invention

The embodiment of the invention provides a software deployment method and device, which can save communication bandwidth and reduce power consumption.

The first aspect provides a software deployment method, including: acquiring communication topology information of a cloud computing system; and selecting a virtual machine of the host deployment software system from the first host set according to the communication topology information, where the first The host collection is a collection of hosts of the virtual machine in the cloud computing system where the software system is to be deployed.

In a first possible implementation, the virtual machine of the host deployment software system is selected from the first host set according to the communication topology information, including:

When the host of the virtual machine of the software system is deployed in the cloud computing system, the affinity of the host in the first host set and all hosts in the second host set is determined according to the communication topology information, where The second host set is a set of hosts of the virtual machine in the cloud computing system where the software system has been deployed;

Selecting the host with high affinity from the first set of hosts to deploy the virtual machine of the software system.

With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, the communication topology information includes a hop count of a communication path between hosts in the cloud computing system;

Determining, according to the communication topology information, the affinity of the host in the first host set and all hosts in the second host set, including:

The affinity is determined according to the hop count of the communication path between the host in the first host set and all the hosts in the second host set.

In conjunction with the second possible implementation of the first aspect, in a third possible implementation, the affinity is between the host in the first host set and all hosts in the second host set. The average number of hops of the communication path or the sum of the sums is inversely proportional.

In combination with the second or third possible implementation of the first aspect, in a fourth possible implementation Obtaining communication topology information of the cloud computing system, including:

Receiving, by each host in the cloud computing system, a hop count of a communication path between each host and other hosts in the cloud computing system, wherein each host and other hosts in the cloud computing system The number of hops of the communication path is obtained by each host through link tracking;

The number of hops of the communication path between the hosts in the cloud computing system is determined based on the number of hops of the communication path between each host and other hosts in the cloud computing system.

With reference to the second or the third possible implementation manner of the first aspect, in a fifth possible implementation manner, the communications topology information of the cloud computing system is obtained, including:

Receiving neighbor node information sent by each host or network device in the cloud computing system, where neighboring node information sent by each host or network device is passed by the host layer or network device through a link layer discovery protocol LLDP Or Cisco Discovery Protocol CDP mode is obtained;

Determining the number of hops of the communication path between the hosts in the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system.

With reference to the first possible implementation manner of the first aspect, in a sixth possible implementation manner, the communication topology information includes a number of connections between hosts in the cloud computing system;

The affinity is determined according to the number of connections between the host in the first host set and all hosts in the second host set.

In conjunction with the sixth possible implementation of the first aspect, in a seventh possible implementation, the affinity is the same as the host in the first host set and all hosts in the second host set. The average number of connections or the sum of the sums is inversely proportional.

With reference to the sixth or the seventh possible implementation manner of the first aspect, in an eighth possible implementation manner, the communications topology information of the cloud computing system is obtained, including:

Receiving information about neighboring nodes sent by each host or network device in the cloud computing system, where The neighbor node information sent by each host or network device is obtained by the LLDP or CDP mode by each host or network device;

Determining the number of connections between hosts in the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system.

With reference to the sixth or the seventh possible implementation of the first aspect, in a ninth possible implementation manner, the communications topology information of the cloud computing system is obtained, including:

The number of connections between hosts in the cloud computing system is determined according to a pre-configured connection relationship of all hosts and network devices in the cloud computing system.

With reference to the first possible implementation of the first aspect, in a tenth possible implementation, the communication topology information includes network domain information of a host in the cloud computing system;

The affinity is determined according to network domain information of the host in the first host set and network domain information of all hosts in the second host set.

With reference to the tenth possible implementation manner of the foregoing aspect, in an eleventh possible implementation manner, the level of the affinity is the same as the network domain where the host in the first host set is located and the second host set The size of the difference in the network domain in which all hosts are located is inversely proportional.

With reference to the tenth or eleven possible implementation manners of the first aspect, in a twelfth possible implementation, the communications topology information of the cloud computing system is obtained, including:

Obtain the pre-configured network domain information of the host in the cloud computing system.

With reference to the first aspect or any one of the first to the twelve possible implementation manners of the first aspect, in the thirteenth possible implementation manner, according to the communication topology information, from the first host Select the virtual machine in the collection to deploy the software system to the host, including:

When the host of the virtual machine of the software system is not deployed in the cloud computing system, determining, according to the communication topology information, the affinity of the host in the first host set and all hosts in the third host set. The third host set is a set of hosts of the virtual machine in the cloud computing system where other software systems have been deployed;

A virtual machine having a low affinity with all hosts in the third host set is selected from the first set of hosts to deploy a virtual machine of the software system.

The second aspect provides a software deployment apparatus, including: an obtaining module, configured to acquire communication topology information of a cloud computing system; and a deployment module, configured to select a host deployment software from the first host set according to the communication topology information. The virtual machine of the system, wherein the first host set is a set of hosts of the virtual machine in the cloud computing system to be deployed the software system.

In a first possible implementation manner, the deployment module includes:

a first determining unit, configured to: when the host of the virtual machine that has deployed the software system is deployed in the cloud computing system, determine, according to the communication topology information, a host in the first host set and all hosts in the second host set Affinity, wherein the second set of hosts is a set of hosts of the virtual machine in the cloud computing system that have deployed the software system;

And a first deployment unit, configured to select, from the first set of hosts, the virtual machine with the high affinity to deploy the virtual machine of the software system.

With reference to the first possible implementation manner of the second aspect, in a second possible implementation manner, the communication topology information includes a hop count of a communication path between hosts in the cloud computing system;

The first determining unit is specifically configured to determine the affinity according to a hop count of a communication path between the host in the first host set and all hosts in the second host set.

With the second possible implementation of the second aspect, in a third possible implementation, the affinity is between the host in the first host set and all hosts in the second host set. The average number of hops of the communication path or the sum of the sums is inversely proportional.

With reference to the second or the third possible implementation of the second aspect, in a fourth possible implementation, the acquiring module is specifically configured to receive each host sent by each host in the cloud computing system and the The number of hops of the communication path between other hosts in the cloud computing system, where each host and the The number of hops of the communication path between other hosts in the cloud computing system is obtained by the each host through link tracking, and is determined according to the hop count of the communication path between each host and other hosts in the cloud computing system. The number of hops of the communication path between hosts in the cloud computing system.

With reference to the second or the third possible implementation of the second aspect, in a fifth possible implementation, the acquiring module is specifically configured to receive, by each host or network device, a neighboring node sent by the network computing system Information, wherein the neighbor node information sent by each host or network device is obtained by each host or network device by a link layer discovery protocol LLDP or a Cisco Discovery Protocol CDP mode, according to all hosts in the cloud computing system and The neighbor node information transmitted by the network device determines the hop count of the communication path between the hosts in the cloud computing system.

With reference to the first possible implementation of the second aspect, in a sixth possible implementation, the communication topology information includes a number of connections between hosts in the cloud computing system;

The first determining unit is specifically configured to determine the affinity according to the number of connections between the host in the first host set and all hosts in the second host set.

In conjunction with the sixth possible implementation of the second aspect, in a seventh possible implementation, the affinity is between the host in the first host set and all hosts in the second host set. The average number of connections or the sum of the sums is inversely proportional.

With reference to the sixth or the seventh possible implementation of the second aspect, in an eighth possible implementation, the acquiring module is specifically configured to receive, by each host or network device, a neighboring node sent by the network computing system Information, wherein the neighbor node information sent by each host or network device is obtained by the LLDP or CDP mode by each host or network device, and the neighbor node information sent by all hosts and network devices in the cloud computing system is obtained. Determine the number of connections between hosts in the cloud computing system.

With reference to the sixth or the seventh possible implementation of the second aspect, in a ninth possible implementation manner, the acquiring module is specifically configured to: according to a pre-configured connection relationship between all hosts and network devices in the cloud computing system Determine the number of connections between hosts in the cloud computing system.

In conjunction with the first possible implementation of the second aspect, in a tenth possible implementation, The communication topology information includes network domain information of a host in the cloud computing system;

The first determining unit is specifically configured to determine the affinity according to network domain information of the host in the first host set and network domain information of all hosts in the second host set.

With reference to the tenth possible implementation manner of the second aspect, in an eleventh possible implementation manner, the affinity is different from the network domain in which the host in the first host set is located and the second host set The size of the difference in the network domain in which all hosts are located is inversely proportional.

With reference to the tenth or eleven possible implementation manners of the second aspect, in a twelfth possible implementation, the acquiring module is specifically configured to obtain network domain information of a host in the pre-configured cloud computing system.

With reference to the second aspect or any one of the first to the twelve possible implementation manners of the second aspect, in a thirteenth possible implementation manner, the deployment module includes:

a second determining unit, configured to: when the host of the virtual machine of the software system is deployed in the cloud computing system, determine, according to the communication topology information, all hosts in the first host set and the third host set Affinity, wherein the third host set is a set of hosts of the virtual machine in the cloud computing system where other software systems have been deployed;

And a second deployment unit, configured to select, from the first set of hosts, a virtual machine that has low affinity with all hosts in the third host set to deploy the virtual machine of the software system.

Based on the foregoing technical solution, the software deployment method and device of the embodiment of the present invention can select a virtual machine of the host software system according to the communication topology information of the cloud computing system, and the virtual machine of the same software system can be deployed in a centralized manner, thereby saving communication. Bandwidth, reducing cost and power consumption.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only some of the present invention. For the embodiments, those skilled in the art can obtain other drawings according to the drawings without any creative work.

1 is a schematic flow chart of a method of software deployment in accordance with one embodiment of the present invention.

2 is a schematic flow chart of a method for software deployment according to another embodiment of the present invention.

3 is a schematic diagram of a cloud computing system in accordance with one embodiment of the present invention.

4 is a schematic diagram of a cloud computing system in accordance with another embodiment of the present invention.

FIG. 5 is a schematic diagram of a cloud computing system in accordance with yet another embodiment of the present invention.

6 is a schematic diagram of software deployment results in accordance with one embodiment of the present invention.

7 is a schematic block diagram of an apparatus for software deployment in accordance with one embodiment of the present invention.

Figure 8 is a schematic block diagram of a deployment module in accordance with one embodiment of the present invention.

9 is a schematic block diagram of a deployment module in accordance with another embodiment of the present invention.

FIG. 10 is a schematic structural diagram of an apparatus for software deployment according to another embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

FIG. 1 shows a schematic flow diagram of a method 100 of software deployment in accordance with an embodiment of the present invention. As shown in FIG. 1, the method 100 includes:

S110. Obtain communication topology information of the cloud computing system.

S120. Select a virtual machine of the host deployment software system from the first host set according to the communication topology information, where the first host set is a set of hosts of the virtual machine in the cloud computing system where the software system is to be deployed.

In the embodiment of the present invention, the execution subject of the method 100 is a device deployed by software. The device may be a device in the cloud computing management system or a cloud computing management system itself, which is not limited by the embodiment of the present invention.

It should be understood that in the embodiments of the present invention, the software system may also be expressed as software. In the cloud computing system When the software system is deployed, the virtual machine in which the software system is deployed is selected in the cloud computing system, and then the software system is installed on the virtual machine.

When the software system is deployed, the software deployed device first obtains the communication topology information of the cloud computing system, and then selects, according to the communication topology information, the virtual machine that the host deploys the software system from the first host set. In the embodiment of the present invention, the first host set is a set of hosts of the virtual machine in the cloud computing system where the software system is to be deployed. It should be understood that the term "first host set" is used in the embodiment of the present invention for convenience of description, and does not limit the embodiment of the present invention. In the embodiment of the present invention, the software deployment device selects the virtual machine of the host deployment software system according to the communication topology information of the cloud computing system, so that the result of the deployment is related to the communication topology of the cloud computing system, and is convenient for the same software system. Virtual machine centralized deployment. That is to say, virtual machines of the same software system can be deployed on hosts connected to the same or fewer switches, so communication traffic of the same software system can be forwarded by fewer switches, thereby saving communication bandwidth, reducing cost and power consumption. .

Therefore, the software deployment method of the embodiment of the present invention can centrally deploy the virtual machine of the same software system by selecting the virtual machine of the host deployment software system according to the communication topology information of the cloud computing system, thereby saving communication bandwidth and reducing cost and Power consumption.

In the embodiment of the present invention, optionally, as shown in FIG. 2, selecting a virtual machine of the host deployment software system from the first host set according to the communication topology information includes:

S121, when there is a host of the virtual machine in which the software system is deployed in the cloud computing system, determining, according to the communication topology information, an affinity between the host in the first host set and all hosts in the second host set, The second host set is a set of hosts of the virtual machine in which the software system has been deployed in the cloud computing system;

S122. Select a host with high affinity from the first host set to deploy a virtual machine of the software system.

In the embodiment of the present invention, the second host set is a set of hosts of the virtual machine in which the software system has been deployed in the cloud computing system. It should be understood that the embodiment of the present invention adopts the term second host set only for the purpose of The description is convenient and the invention is not limited in any way.

When the virtual machine of the software system is deployed according to the communication topology information, if the host of the virtual machine in which the software system is deployed is in the cloud computing system, that is, the second host set is not empty, the software deployment device first according to the communication topology information. Determining the affinity of the host in the first host set and all the hosts in the second host set, and then selecting, according to the affinity, the virtual machine that the host deploys the software system from the first host set. Specifically, the software deployed device selects a host with high affinity from the first set of hosts to deploy the virtual machine of the software system. For example, selecting the host with the highest affinity from the first host set to deploy the virtual machine of the software system, or selecting the remaining host with the highest affinity when the host with the highest affinity cannot deploy the virtual machine of the software system. The host deploys the virtual machine of the software system.

In the embodiment of the present invention, the affinity indicates the proximity relationship between a host and all hosts in a host set. Affinity can be determined based on communication topology information. There are various ways to determine the affinity for the content of different communication topology information, which are described separately below.

In the embodiment of the present invention, optionally, the communication topology information includes a hop count of a communication path between hosts in the cloud computing system.

In this case, determining, according to the communication topology information, the affinity of the host in the first host set and all hosts in the second host set, including:

Specifically, in this embodiment, the communication topology information is the hop count of the communication path between the hosts in the cloud computing system. For example, in the cloud computing system shown in FIG. 3, the hop count of the communication path between the host 1-1 and the host 1-2 is 2; the hop of the communication path between the host 1-1 and the host 2-1 The number is 4. In this case, the affinity of the host in the first host set and all the hosts in the second host set may be based on a communication path between the host in the first host set and all hosts in the second host set. The number of hops is determined.

Optionally, the affinity is the same as the host in the first host set and the second host set. The average or the sum of the hop counts of the communication paths between all the hosts is inversely proportional.

Taking the cloud computing system shown in FIG. 3 as an example, if the second host set includes the host 1-1 and the host 1-2, the first host set includes the host 1-3, the host 2-1, the host 2-2, and the host 2 -3, that is, the host computer 1-1 and the host 1-2 have deployed the virtual machine of the software system, the host 1-3, the host 2-1, the host 2-2, and the host 2-3 to be deployed the virtual machine of the software system, When selecting a virtual machine that deploys the software system from the first host set, it is required to determine the hop count of the communication path between each host in the first host set and all hosts in the second host set, thereby determining the number The affinity of each host in a set of hosts to all hosts in the second set of hosts.

The hop count of the communication path between the host 1-3 and the host 1-1 is 2, the hop count of the communication path with the host 1-2 is 2, the average value is 2, and the sum is 4;

The hop count of the communication path between the host 2-1 and the host 1-1 is 4, the hop count of the communication path with the host 1-2 is 4, the average value is 4, and the sum is 8;

The hop count of the communication path between the host 2-2 and the host 1-1 is 4, the hop count of the communication path with the host 1-2 is 4, the average value is 4, and the sum is 8;

The number of hops of the communication path between the host 2-3 and the host 1-1 is 4, the hop count of the communication path with the host 1-2 is 4, the average value is 4, and the sum is 8.

The level of affinity is inversely proportional to the average or sum of the hop counts of the communication paths between the hosts in the first set of hosts and all the hosts in the second set of hosts. Therefore, hosts 1-3 in the first set of hosts have the highest affinity, and hosts 1-3 are selected to deploy virtual machines of the software system. In this way, the virtual machines of the same software system are deployed on the host 1-1, the host 1-2, and the hosts 1-3. The traffic in the software system only passes through the Ethernet switch L2-1, thereby saving communication bandwidth and reducing network equipment. Cost of use and power consumption.

When the communication topology information is the hop count of the communication path between the hosts in the cloud computing system, the communication topology information is acquired in various ways.

In the embodiment of the present invention, optionally, acquiring communication topology information of the cloud computing system includes:

Specifically, each host obtains the hop count of the communication path between each host and other hosts in the cloud computing system through a link tracking manner. For example, in the cloud computing system shown in FIG. 3, the host 1-1 sends a link tracking message to all other hosts, wherein the link tracking message to the host 1-2 passes only through the Ethernet switch L2-1, the host 1-1 will receive two link tracking response messages from the Ethernet switch L2-1 and the host 1-2 on this path, so the host 1-1 can know the communication path between it and the host 1-2. The number of hops is 2; the link trace message to host 2-1 will pass through Ethernet switch L2-1, Ethernet switch L1-1, and Ethernet switch L2-2, so it will receive from this path. The four links of the Ethernet switch L2-1, the Ethernet switch L1-1, the Ethernet switch L2-2, and the host 2-1 track the response message, so the host 1-1 can know that it is between the host 2-1 and the host 2-1. The number of hops of the communication path is 4. By analogy, each host knows the number of hops of its communication path with other hosts in the cloud computing system.

Each host sends a hop count of the communication path between each host and other hosts in the cloud computing system to the software deployed device, such that the software deployed device can be based on each host and other hosts in the cloud computing system The hop count of the communication path determines the number of hops of the communication path between the hosts in the cloud computing system. In this embodiment, each host may send the hop count of the communication path between each host and other hosts in the cloud computing system to the software deployed device through the existing management channel in the cloud computing system.

Each host can periodically send link tracking messages to other hosts of the cloud computing system to automatically detect network changes and send the detected results to the software deployed device through the management channel to refresh the software deployed devices. Recorded results.

Receiving neighbor node information sent by each host or network device in the cloud computing system, where neighboring node information sent by each host or network device is passed by the host layer or network device through a link layer discovery protocol ( Link Layer Discovery Protocol (LLDP) or Cisco Discovery Protocol (CDP) method;

Specifically, each host or network device obtains neighbor node information through LLDP or CDP, and sends neighbor node information to the software deployment device, and the software deployment device determines the host in the cloud computing system according to the neighbor node information. The number of hops between the communication paths. For example, in the cloud computing system shown in FIG. 4, the virtual switch and the Ethernet switch on each host obtain neighbor node information through LLDP or CDP, and then each host and the Ethernet switch pass their respective management channels. Adjacent node information is sent to the device deployed by the software. In this embodiment, the software deployed device needs to establish a management channel with the Ethernet switch, and the Ethernet switch sends the neighbor node information to the software deployed device through the established management channel.

In the embodiment of the present invention, optionally, the communication topology information includes a number of connections between hosts in the cloud computing system.

Specifically, in this embodiment, the communication topology information is the number of connections between hosts in the cloud computing system. For example, in the cloud computing system shown in FIG. 4, the number of connections between the host 1-1 and the host 1-2 is 2; the number of connections between the host 1-1 and the host 2-1 is 4. In this case, the affinity of the host in the first host set and all the hosts in the second host set may be based on the first host set. The number of connections between the host in the middle and all hosts in the second set of hosts is determined.

Optionally, the affinity is inversely proportional to the average or the sum of the number of connections between the host in the first host set and all hosts in the second host set.

Taking the cloud computing system shown in FIG. 4 as an example, if the second host set includes the host 1-1 and the host 1-2, the first host set includes the host 1-3, the host 2-1, the host 2-2, and the host 2 -3.

The number of connections between the host 1-3 and the host 1-1 is 2, the number of connections with the host 1-2 is 2, the average value is 2, and the sum is 4;

The number of connections between the host 2-1 and the host 1-1 is 4, the number of connections with the host 1-2 is 4, the average value is 4, and the sum is 8.

The number of connections between the host 2-2 and the host 1-1 is 4, the number of connections with the host 1-2 is 4, the average value is 4, and the sum is 8.

The number of connections between the host 2-3 and the host 1-1 is 4, the number of connections with the host 1-2 is 4, the average value is 4, and the sum is 8.

The level of affinity is inversely proportional to the average or sum of the number of connections between the host in the first set of hosts and all hosts in the second set of hosts. Therefore, hosts 1-3 in the first set of hosts have the highest affinity, and hosts 1-3 are selected to deploy virtual machines of the software system. In this way, the virtual machines of the same software system are deployed on the host 1-1, the host 1-2, and the hosts 1-3. The traffic in the software system only passes through the Ethernet switch L2-1, thereby saving communication bandwidth and reducing network equipment. Cost of use and power consumption.

When the communication topology information is the number of connections between hosts in the cloud computing system, the communication topology information is acquired in various ways.

Receiving, by the host or the network device, neighboring node information sent by the host or the network device, where the neighboring node information sent by each host or network device is obtained by using the LLDP or CDP mode;

In this embodiment, the software deployed device determines the number of connections between the hosts based on the neighbor node information.

In this embodiment, the connection relationship of each node in the cloud computing system is pre-configured. That is to say, the connection relationship between each host and the network device in the cloud computing system is pre-configured to the software deployment device, so that the software deployed device can obtain the number of connections between the hosts according to the pre-configured connection relationship.

In the embodiment of the present invention, optionally, the communication topology information includes network domain information of the host in the cloud computing system.

Specifically, in this embodiment, the communication topology information is network domain information of a host in the cloud computing system. As shown in FIG. 5, the cloud computing system is divided into multiple network domains. For example, the network domain L2-1 includes hosts 1-1, 1-2, and 1-3, the network domain L2-2 includes hosts 2-1, 2-2, and 2-3, and the network domain L2-3 includes hosts 3-1. , 3-2, 3-3, the network domain L2-4 includes hosts 4-1, 4-2, 4-3. The network domain can also be recursively nested. For example, the network domain L1-1 includes the network domains L2-1 and L2-2, and the network domain L1-2 includes the network domains L2-3 and L2-4. Affinity can be determined based on the host's network domain information. If the host is in the same network domain as the host in the second set, the affinity is high; if it is in a different network domain, the affinity is low.

Optionally, the level of the affinity is inversely proportional to the size of the difference between the network domain in which the host in the first host set is located and the network domain in which all hosts in the second host set are located.

For example, in FIG. 5, if the second set of hosts includes the host 1-1 and the host 1-2, the other hosts constitute the first set of hosts. When selecting a virtual machine of the host deployment software system from other hosts, since the host 1-3 is in the same network domain as the host 1-1 and the host 1-2, the difference of the network domain is the smallest, so the host 1-3 is selected to deploy the software. The virtual machine of the system. In this way, the virtual machines of the same software system are deployed on the host 1-1, the host 1-2, and the hosts 1-3. The traffic in the software system only passes through the Ethernet switch L2-1, thereby saving communication bandwidth and reducing network equipment. Cost of use and power consumption.

Optionally, obtaining communication topology information of the cloud computing system, including:

That is to say, the network domain information of each host in the cloud computing system can be pre-configured to the software deployment device, so that the software deployed device can obtain the pre-configured network domain information.

In the software deployment method of the embodiment of the present invention, the virtual machine of the same software system can be centrally deployed by selecting the virtual machine of the host deployment software system according to the affinity, thereby saving communication bandwidth and reducing cost and power consumption.

The method for selecting a virtual machine for deploying the software system by the host when the host of the virtual machine of the software system is deployed in the cloud computing system is described above. The following describes the virtual machine in the cloud computing system where the software system has not been deployed. When the host is hosted, select the method by which the host deploys the virtual machine of the software system.

In the embodiment of the present invention, optionally, selecting a virtual machine of the host deployment software system from the first host set according to the communication topology information, including:

When the host of the virtual machine of the software system is deployed in the cloud computing system, the affinity of the host in the first host set and all hosts in the third host set is determined according to the communication topology information, where The third host set is a set of hosts of the virtual machine in the cloud computing system where other software systems have been deployed;

In the embodiment of the present invention, the third host is a host of a virtual machine that has deployed other software systems. A collection of hosts, that is, a collection of hosts that have deployed virtual machines that are different from the software systems of the software systems to be deployed. It should be understood that the term "third host set" is used in the embodiment of the present invention for convenience of description, and does not limit the embodiment of the present invention.

When there is no host of the virtual machine in which the software system is deployed in the cloud computing system, that is, when the host of the virtual machine that deploys the software system is selected for the first time, the device deployed by the software is based on the host in the first host set. Selecting the host to deploy the virtual machine of the software system with the affinity of all hosts in the third host set. Specifically, the software deployed device selects a virtual machine of the software system from a first host set that selects a host with low affinity to all hosts in the third host set.

It should be understood that, in the embodiment of the present invention, the manner in which the affinity between the host in the first host set and all the hosts in the third host set is determined, and all the hosts in the first host set and the second host set are determined. The affinity of the host is determined in a similar manner. For the sake of brevity, it will not be described again.

The software deployment method of the embodiment of the present invention can centrally deploy virtual machines of the same software system. As shown in FIG. 6, the virtual machine of the software system A is deployed on the hosts 1 to 5, and the virtual machines of the software system B are deployed on the hosts 5 to 10. The traffic in software system A is mainly exchanged in Ethernet switches 2 and 3, and a small amount of traffic passes through Ethernet switch 1; the traffic in software system B is mainly exchanged in Ethernet switches 3 and 4, and a small amount of traffic passes through the Ethernet switch. 1. Such a deployment result can save network equipment usage cost and power consumption, and can avoid network bottlenecks. As shown in FIG. 6, since the traffic passing through the Ethernet switch 1 is reduced, it is possible to prevent the transmission path bandwidth between the Ethernet switch 1 and other Ethernet switches from becoming a bottleneck.

It should be understood that, in the embodiment of the present invention, the software deployed device may also directly select the virtual machine of the host deployment software system according to the communication topology information. For example, selecting the average or the sum of the hops of the communication paths with all the hosts in the second set of hosts, or selecting the average or sum of the number of connections with all the hosts in the second set of hosts The smallest host, or the host with the smallest difference between the selected network domain and the network domain in which all hosts in the second host set are located, deploy the virtual machine of the software system. These embodiments are also intended to fall within the scope of the invention.

The software deployment method of the embodiment of the present invention can select a virtual machine of the host software system according to the communication topology information of the cloud computing system, and the virtual machine of the same software system can be deployed in a centralized manner, thereby saving communication bandwidth and reducing cost and power consumption. .

It should be noted that the examples in the embodiments of the present invention are only intended to help those skilled in the art to better understand the embodiments of the present invention, and do not limit the scope of the embodiments of the present invention.

It should be understood that, in various embodiments of the present invention, the size of the sequence numbers of the above processes does not mean the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not be taken to the embodiments of the present invention. The implementation process constitutes any limitation.

The method of software deployment according to an embodiment of the present invention is described in detail above with reference to FIG. 1 to FIG. 6. Hereinafter, an apparatus for software deployment according to an embodiment of the present invention will be described with reference to FIG. 7 to FIG.

FIG. 7 shows a schematic block diagram of an apparatus 700 for software deployment in accordance with an embodiment of the present invention. As shown in FIG. 7, the apparatus 700 includes:

The obtaining module 710 is configured to obtain communication topology information of the cloud computing system.

The deployment module 720 is configured to select, according to the communication topology information, a virtual machine of the host deployment software system from the first host set, where the first host set is a host of the virtual machine in the cloud computing system where the software system is to be deployed. Collection.

In the embodiment of the present invention, the software deployment device selects the virtual machine of the host deployment software system according to the communication topology information of the cloud computing system, so that the result of the deployment is related to the communication topology of the cloud computing system, and is convenient for the same software system. Virtual machine centralized deployment. That is to say, virtual machines of the same software system can be deployed on hosts connected to the same or fewer switches, so communication traffic of the same software system can be forwarded by fewer switches, thereby saving communication bandwidth, reducing cost and power consumption. .

Therefore, the software deployment device of the embodiment of the present invention can centrally deploy the virtual machine of the same software system by selecting the virtual machine of the host deployment software system according to the communication topology information of the cloud computing system, thereby saving communication bandwidth and reducing cost and Power consumption.

In the embodiment of the present invention, optionally, as shown in FIG. 8, the deployment module 720 includes:

The first determining unit 721 is configured to: when the host of the virtual machine of the software system is deployed in the cloud computing system, determine, according to the communication topology information, all the hosts in the first host set and the second host set The affinity of the host, wherein the second host set is a set of hosts of the virtual machine in the cloud computing system where the software system has been deployed;

The first deployment unit 722 is configured to select, from the first set of hosts, the virtual machine with the high affinity to deploy the virtual machine of the software system.

The first determining unit 721 is specifically configured to determine the affinity according to the hop count of the communication path between the host in the first host set and all hosts in the second host set.

In the embodiment of the present invention, optionally, the average or the sum of the hops of the communication path between the host in the first host set and all the hosts in the second host set is The size is inversely proportional.

In the embodiment of the present invention, the obtaining module 710 is specifically configured to receive, by each host in the cloud computing system, a communication path between each host and other hosts in the cloud computing system. The hop count, wherein the hop count of the communication path between each host and other hosts in the cloud computing system is obtained by each host through a link tracking manner, according to each host and the cloud computing system The number of hops of the communication path between other hosts determines the number of hops of the communication path between the hosts in the cloud computing system.

In the embodiment of the present invention, the acquiring module 710 is specifically configured to receive, by using each host or network device, the neighbor node information sent by the host or the network device. The neighbor node information is obtained by each host or network device by using a link layer discovery protocol LLDP or a Cisco discovery protocol CDP mode, and determining the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system. The number of hops of the communication path between the hosts in the host.

In the embodiment of the present invention, optionally, the communication topology information includes a host in the cloud computing system. The number of connections between.

The first determining unit 721 is specifically configured to determine the affinity according to the number of connections between the host in the first host set and all hosts in the second host set.

In the embodiment of the present invention, optionally, the affinity is inversely proportional to the average value or the sum of the connections between the host in the first host set and all the hosts in the second host set. .

In the embodiment of the present invention, the acquiring module 710 is specifically configured to receive, by using each host or network device, the neighbor node information sent by the host or the network device. The neighboring node information is obtained by the LLDP or CDP mode by the host or the network device, and the number of connections between the hosts in the cloud computing system is determined according to the information of the neighboring nodes sent by all the hosts and the network devices in the cloud computing system. .

In the embodiment of the present invention, the obtaining module 710 is specifically configured to determine, according to a pre-configured connection relationship between all hosts and network devices in the cloud computing system, the number of connections between the hosts in the cloud computing system. .

In the embodiment of the present invention, optionally, the communication topology information includes network domain information of a host in the cloud computing system.

The first determining unit 721 is specifically configured to determine the affinity according to network domain information of the host in the first host set and network domain information of all hosts in the second host set.

In the embodiment of the present invention, optionally, the level of the affinity is different from the size of the network domain where the host in the first host set is located and the network domain where all hosts in the second host set are located. In inverse proportion.

In the embodiment of the present invention, the acquiring module 710 is specifically configured to obtain the pre-configured network domain information of the host in the cloud computing system.

In the embodiment of the present invention, as shown in FIG. 9, the deployment module 720 optionally includes:

a second determining unit 723, configured to: in the cloud computing system, the virtual system that has deployed the software system The host of the machine determines the affinity of the host in the first host set and all the hosts in the third host set according to the communication topology information, where the third host set is deployed in the cloud computing system. a collection of hosts of a virtual machine of a software system;

The second deployment unit 724 is configured to select, from the first set of hosts, a virtual machine that has low affinity with all hosts in the third host set to deploy the virtual machine of the software system.

The software-deployed apparatus 700 according to an embodiment of the present invention may correspond to an execution subject in a software-deployed method according to an embodiment of the present invention, and the above-described and other operations and/or functions of the respective modules in the apparatus 700 are respectively implemented in order to implement FIG. The corresponding processes of the respective methods in FIG. 6 are not described herein again for the sake of brevity.

FIG. 10 shows a structure of an apparatus for software deployment according to still another embodiment of the present invention, including at least one processor 1002 (eg, a CPU), at least one network interface 1005 or other communication interface, a memory 1006, and at least one communication bus 1003. Used to implement connection communication between these devices. The processor 1002 is configured to execute executable modules, such as computer programs, stored in the memory 1006. The memory 1006 may include a high speed random access memory (RAM), and may also include a non-volatile memory such as at least one disk memory. A communication connection with at least one other network element is achieved by at least one network interface 1005, which may be wired or wireless.

In some embodiments, the memory 1006 stores a program 10061, and the processor 1002 executes the program 10061 for performing the following operations:

Obtaining communication topology information of the cloud computing system;

Determining, according to the communication topology information, a virtual machine of the host deployment software system from the first host set, where the first host set is a host of the virtual machine in the cloud computing system to be deployed by the software system Machine collection.

Optionally, the processor 1002 is specifically configured to: when the host of the virtual machine of the software system is deployed in the cloud computing system, determine, according to the communication topology information, the host and the second host set in the first host set. Affinity of all the hosts in the set, wherein the second set of hosts is a set of hosts of the virtual machine in the cloud computing system where the software system has been deployed; selecting the host with high affinity from the first set of hosts Deploy the virtual machine of the software system.

Optionally, the communication topology information includes a hop count of a communication path between hosts in the cloud computing system; the processor 1002 is specifically configured to: according to the host in the first host set and all hosts in the second host set The number of hops between the communication paths determines the affinity.

Optionally, the level of the affinity is inversely proportional to the average value or the sum of the hop counts of the communication paths between the host in the first host set and all the hosts in the second host set.

Optionally, the processor 1002 is specifically configured to receive, by using the network interface 1005, a hop count of a communication path between each host and each host in the cloud computing system sent by each host in the cloud computing system, where The number of hops of the communication path between each host and other hosts in the cloud computing system is obtained by each host through link tracking; according to each host and other hosts in the cloud computing system The number of hops of the communication path determines the number of hops of the communication path between the hosts in the cloud computing system.

Optionally, the processor 1002 is specifically configured to receive, by using the network interface 1005, neighbor node information sent by each host or network device in the cloud computing system, where the neighbor node information sent by each host or network device is configured by Each host or network device is obtained through a link layer discovery protocol LLDP or a Cisco Discovery Protocol CDP mode; determining, according to neighboring node information sent by all hosts and network devices in the cloud computing system, between the hosts in the cloud computing system The number of hops for the communication path.

Optionally, the communication topology information includes a number of connections between hosts in the cloud computing system; the processor 1002 is specifically configured to use, according to the hosts in the first host set and all hosts in the second host set. The number of connections determines the affinity.

Optionally, the processor 1002 is specifically configured to receive, by using the network interface 1005, neighbor node information sent by each host or network device in the cloud computing system, where the neighbor node information sent by each host or network device is configured by Each host or network device is obtained by means of LLDP or CDP; determining the number of connections between hosts in the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system.

Optionally, the processor 1002 is specifically configured to determine, according to a pre-configured connection relationship of all hosts and network devices in the cloud computing system, a number of connections between hosts in the cloud computing system.

Optionally, the communication topology information includes network domain information of the host in the cloud computing system; the processor 1002 is specifically configured to: according to network domain information of the host in the first host set and all hosts in the second host set The network domain information determines the affinity.

Optionally, the processor 1002 is specifically configured to obtain network domain information of the host in the pre-configured cloud computing system.

Optionally, the processor 1002 is specifically configured to: when the host of the virtual machine that has deployed the software system is not in the cloud computing system, determine, according to the communication topology information, the host in the first host set and the third host set. Affinity of all hosts, wherein the third set of hosts is a set of hosts of the virtual machine in which the other software systems have been deployed in the cloud computing system; selecting from the first set of hosts and the third set of hosts A host with low affinity for all hosts deploys virtual machines of the software system.

It can be seen from the foregoing technical solutions provided by the embodiments of the present invention that the virtual machine of the host software deployment system is selected according to the communication topology information of the cloud computing system, and the virtual machines of the same software system can be deployed in a centralized manner, thereby saving Communication bandwidth, reducing cost and power consumption.

It should be understood that in the embodiment of the present invention, the term "and/or" is merely a description of the associated object. The association indicates that there can be three relationships. For example, A and/or B may indicate that A exists separately, and A and B exist simultaneously, and B cases exist alone. In addition, the character "/" in this article generally indicates that the contextual object is an "or" relationship.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both, for clarity of hardware and software. Interchangeability, the composition and steps of the various examples have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods for implementing the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present invention.

A person skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system, the device and the unit described above can refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, or an electrical, mechanical or other form of connection.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the embodiments of the present invention.

In addition, each functional unit in various embodiments of the present invention may be integrated in one processing unit. It is also possible that each unit physically exists alone, or two or more units may be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention contributes in essence or to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium. A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any equivalent person can be easily conceived within the technical scope of the present invention by any person skilled in the art. Modifications or substitutions are intended to be included within the scope of the invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

A method of software deployment, comprising:

Obtaining communication topology information of the cloud computing system;

And selecting, according to the communication topology information, a virtual machine of the host deployment software system, where the first host set is a set of hosts of the virtual machine in the cloud computing system where the software system is to be deployed. .
The method according to claim 1, wherein the selecting a virtual machine of the host deployment software system from the first set of hosts according to the communication topology information comprises:

When the host of the virtual machine of the software system is deployed in the cloud computing system, determining, according to the communication topology information, the affinity of the host in the first host set and all hosts in the second host set. The second set of hosts is a set of hosts of the virtual machine in which the software system has been deployed in the cloud computing system;

Selecting the host with high affinity from the first set of hosts to deploy a virtual machine of the software system.
The method according to claim 2, wherein the communication topology information comprises a hop count of a communication path between hosts in the cloud computing system;

Determining, according to the communication topology information, the affinity of the host in the first host set and all the hosts in the second host set, including:

The affinity is determined according to a hop count of a communication path between a host in the first host set and all hosts in the second host set.
The method according to claim 3, wherein the level of the affinity is the same as the number of hops of the communication path between the host in the first host set and all the hosts in the second host set. The average or the size of the sum is inversely proportional.
The method according to claim 3 or 4, wherein the obtaining the communication topology information of the cloud computing system comprises:

Receiving, by each host in the cloud computing system, a hop count of a communication path between each host and other hosts in the cloud computing system, wherein each of the hosts and the cloud computing system The number of hops of the communication path between other hosts in the medium is obtained by each of the hosts through link tracking;

Determining a hop count of a communication path between hosts in the cloud computing system according to a hop count of a communication path between each host and other hosts in the cloud computing system.
The method according to claim 3 or 4, wherein the obtaining the communication topology information of the cloud computing system comprises:

Receiving neighbor node information sent by each host or network device in the cloud computing system, where neighboring node information sent by each host or network device is passed by the host layer or network device through a link layer Discovery protocol LLDP or Cisco Discovery Protocol CDP mode is obtained;

Determining a hop count of a communication path between hosts in the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system.
The method of claim 2, wherein the communication topology information comprises a number of connections between hosts in the cloud computing system;

Determining, according to the communication topology information, the affinity of the host in the first host set and all the hosts in the second host set, including:

The affinity is determined according to the number of connections between the host in the first host set and all hosts in the second host set.
The method according to claim 7, wherein the level of the affinity is the same as the average number of connections between the host in the first host set and all hosts in the second host set or The sum of the sum is inversely proportional.
The method according to claim 7 or 8, wherein the obtaining the communication topology information of the cloud computing system comprises:

Receiving neighbor node information sent by each host or network device in the cloud computing system, The neighbor node information sent by each host or network device is obtained by each host or network device by using LLDP or CDP;

Determining the number of connections between hosts in the cloud computing system according to neighboring node information sent by all hosts and network devices in the cloud computing system.
The method according to claim 7 or 8, wherein the obtaining the communication topology information of the cloud computing system comprises:

Determining the number of connections between hosts in the cloud computing system according to a pre-configured connection relationship of all hosts and network devices in the cloud computing system.
The method according to claim 2, wherein the communication topology information comprises network domain information of a host in the cloud computing system;

Determining, according to the communication topology information, the affinity of the host in the first host set and all the hosts in the second host set, including:

The affinity is determined according to network domain information of the host in the first host set and network domain information of all hosts in the second host set.
The method according to claim 11, wherein the level of the affinity is the same as the network in which the host in the first host set is located and the network in which all hosts in the second host set are located. The size of the difference in the domain is inversely proportional.
The method according to claim 11 or 12, wherein the obtaining the communication topology information of the cloud computing system comprises:

Obtaining network domain information of the host in the pre-configured cloud computing system.
The method according to any one of claims 1 to 13, wherein the selecting a virtual machine of the host deployment software system from the first set of hosts according to the communication topology information comprises:

When the host of the virtual machine of the software system is deployed in the cloud computing system, determining, according to the communication topology information, the affinity of the host in the first host set and all hosts in the third host set. Sex, wherein the third host set is another software system deployed in the cloud computing system a collection of hosts for the virtual machine;

Selecting, from the first set of hosts, a virtual machine having low affinity with all hosts in the third host set to deploy a virtual machine of the software system.
A device for software deployment, comprising:

An obtaining module, configured to obtain communication topology information of the cloud computing system;

a deployment module, configured to select a virtual machine of the host deployment software system from the first host set according to the communication topology information, where the first host set is a virtual to be deployed in the cloud computing system A collection of hosts for the machine.
The device according to claim 15, wherein the deployment module comprises:

a first determining unit, configured to determine, according to the communication topology information, a host and a second host set in the first host set, when a host of the virtual machine of the software system is deployed in the cloud computing system Affinity of all hosts in the cloud, wherein the second set of hosts is a set of hosts of the virtual machine in which the software system has been deployed in the cloud computing system;

a first deployment unit, configured to select, from the first set of hosts, the virtual machine with the high affinity to deploy a virtual machine of the software system.
The apparatus according to claim 16, wherein said communication topology information comprises a hop count of a communication path between hosts in said cloud computing system;

The first determining unit is specifically configured to determine the affinity according to a hop count of a communication path between a host in the first host set and all hosts in the second host set.
The apparatus according to claim 17, wherein the level of the affinity is the same as the number of hops of the communication path between the host in the first host set and all the hosts in the second host set. The average or the size of the sum is inversely proportional.
The device according to claim 17 or 18, wherein the obtaining module is specifically configured to receive each host sent by each host in the cloud computing system and other ones in the cloud computing system The number of hops of the communication path between the hosts, wherein each of the hosts and the cloud computing system The number of hops of the communication path between the other hosts in the system is obtained by the each host through the link tracking manner, according to the hop count of the communication path between each host and other hosts in the cloud computing system. Determining the number of hops of the communication path between the hosts in the cloud computing system.
The device according to claim 17 or 18, wherein the acquiring module is specifically configured to receive neighboring node information sent by each host or network device in the cloud computing system, where each of the The neighbor node information sent by the host or the network device is obtained by each host or network device through a link layer discovery protocol LLDP or a Cisco Discovery Protocol CDP mode, according to the phase sent by all hosts and network devices in the cloud computing system. The neighbor information determines the number of hops of the communication path between the hosts in the cloud computing system.
The apparatus according to claim 16, wherein said communication topology information comprises a number of connections between hosts in said cloud computing system;

The first determining unit is specifically configured to determine the affinity according to the number of connections between the host in the first host set and all hosts in the second host set.
The apparatus according to claim 21, wherein the level of the affinity is an average of the number of connections between the host in the first host set and all hosts in the second host set or The sum of the sum is inversely proportional.
The device according to claim 21 or 22, wherein the acquiring module is specifically configured to receive neighboring node information sent by each host or network device in the cloud computing system, where each of the The neighboring node information sent by the host or the network device is obtained by the LLDP or CDP mode by the host or the network device, and the cloud computing is determined according to the neighboring node information sent by all the hosts and the network device in the cloud computing system. The number of connections between hosts in the system.
The device according to claim 21 or 22, wherein the obtaining module is configured to determine, according to a pre-configured connection relationship between all hosts and network devices in the cloud computing system, the cloud computing system The number of connections between hosts.
The apparatus of claim 16 wherein said communication topology information comprises Network domain information of the host in the cloud computing system;

The first determining unit is specifically configured to determine the affinity according to network domain information of a host in the first host set and network domain information of all hosts in the second host set.
The apparatus according to claim 25, wherein the level of the affinity is the same as the network in which the host in the first host set is located and the network in which all hosts in the second host set are located. The size of the difference in the domain is inversely proportional.
The device according to claim 25 or 26, wherein the obtaining module is specifically configured to acquire network domain information of a host in the pre-configured cloud computing system.
The apparatus according to any one of claims 15 to 27, wherein the deployment module comprises:

a second determining unit, configured to determine, according to the communication topology information, a host and a third host set in the first host set, when the host of the virtual machine of the software system is not deployed in the cloud computing system Affinity of all hosts in the set, wherein the third set of hosts is a set of hosts of virtual machines in the cloud computing system that have deployed other software systems;

And a second deployment unit, configured to select, from the first set of hosts, a virtual machine that has low affinity with all hosts in the third host set to deploy a virtual machine of the software system.