WO2016115956A1 - 基于云计算的业务系统的调度方法及调度装置 - Google Patents
基于云计算的业务系统的调度方法及调度装置 Download PDFInfo
- Publication number
- WO2016115956A1 WO2016115956A1 PCT/CN2015/098308 CN2015098308W WO2016115956A1 WO 2016115956 A1 WO2016115956 A1 WO 2016115956A1 CN 2015098308 W CN2015098308 W CN 2015098308W WO 2016115956 A1 WO2016115956 A1 WO 2016115956A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- scheduling
- virtual machine
- mapping relationship
- service system
- physical server
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
Definitions
- the present invention relates to a cloud computing resource scheduling method, and in particular to a cloud computing based service system scheduling method and scheduling device.
- Value-added service systems of telecom operators such as SMS, MMS, and WAP gateways
- each set of value-added business systems uses proprietary server resources. For example, the server used by the MMS center is exclusive to the MMS center.
- the server used by the SMS Center is exclusive to the SMS Center.
- An obvious disadvantage of this independent chimney construction method is the serious waste of physical server resources.
- the independent, chimney-based construction model of value-added service systems has gradually been replaced by cloud computing models.
- the resources are first pooled, that is, the virtual server is virtualized into multiple virtual machines, and the business system is carried on the virtual machine.
- a layer of virtualization is added to decouple the coupling between the business system and the physical server, making it possible for multiple service systems to reuse a group of physical servers.
- This multi-service cloud system built with cloud computing technology not only improves the utilization of resources, but also provides the possibility of saving energy and reducing consumption and operating costs.
- the telecommunication operator's business system has obvious tidal characteristics. For example, during the daytime, the load of the business system peaks due to the strong communication demand of people, and the load of the business system will drop significantly at night due to the decrease of communication requirements of people. Therefore, a multi-service cloud system adopting cloud computing technology often uses intelligent migration characteristics to perform intelligent virtual machine scheduling to reduce power consumption and reduce costs.
- a typical solution of the prior art is that, during the daytime, all the virtual machines carrying the services are evenly distributed in each server of the resource pool, and the system carries the peak of the service in an optimal performance state.
- the virtual machine that carries the service is compressed and migrated to the least server, and the idle physical server that does not run the virtual machine is powered off or hibernated to reduce the power consumption of the servers. Achieve the purpose of reducing costs.
- the technical problem to be solved by the embodiments of the present invention is to provide a scheduling method and a scheduling apparatus for a cloud computing-based service system, which are used to reduce the complexity of the scheduling algorithm and reduce the scheduling jitter caused by the traditional scheduling algorithm.
- a scheduling method of a cloud computing-based business system including:
- mapping relationship table between the virtual machine and the physical server of the service system in the scenario mode saved in advance is obtained, where the service system is recorded in the mapping relationship table.
- the virtual machine of the service system is scheduled to the corresponding physical server according to the mapping relationship in the mapping relationship table.
- the service system when the service system is switched to the scenario mode, if the mapping relationship table in the scenario mode is not saved, the resource usage of each physical server in the service system is periodically collected. And performing the migration scheduling of the virtual machine according to the predetermined virtual machine scheduling algorithm; and recording and saving the mapping relationship table in the scene mode according to the mapping relationship between the virtual machine and the physical server after the migration.
- the scheduling the virtual machine of the service system to the corresponding physical server according to the mapping relationship in the mapping relationship table comprises: waking up the physical server recorded in the mapping relationship table, and according to the a mapping relationship in the mapping relationship table, scheduling the virtual machine recorded in the mapping relationship table to a corresponding physical server; and, remaining the physical system except the physical server recorded in the mapping relationship table Physical server, powering down or sleeping.
- the method further includes: periodically collecting the service system The resource occupation of each physical server, and performing the migration scheduling of the virtual machine according to the predetermined virtual machine scheduling algorithm according to the resource occupation situation; and, after performing the migration scheduling of the virtual machine, according to the service The mapping relationship between the current virtual machine and the physical server of the system is updated.
- the working time period of the service system includes a pre-set and periodically presented busy hour period and a idle time period; and according to the resource occupation situation, performing virtual according to a predetermined virtual machine scheduling algorithm
- the migration scheduling of the machine includes: when the scene mode is a busy time scenario corresponding to the busy time period, according to the load balancing algorithm, part of the virtual machine on the physical server whose resource occupancy exceeds a predetermined threshold is migrated to The physical server in the service system has the lowest resource occupancy rate; when the scene mode is a idle time scenario corresponding to the idle time period, the virtual machine of the service system is migrated to a minimum amount according to a knapsack algorithm. On the physical server.
- a scheduling apparatus for a cloud computing-based service system including:
- an obtaining unit configured to: when the service system is switched to a scene mode, obtain a mapping relationship table between the virtual machine and the physical server of the service system in the scenario mode saved in advance, where the mapping relationship is recorded A mapping relationship between the virtual machine and the physical server that is scheduled by the service system according to a predetermined virtual machine scheduling algorithm in the scenario mode, where the scenario mode corresponds to a periodic period of the service system. ;
- the first scheduling unit is configured to schedule the virtual machine of the service system to the corresponding physical server according to the mapping relationship in the mapping relationship table.
- the foregoing scheduling apparatus further includes:
- a second scheduling unit configured to periodically collect the physical relationship between the physical servers in the service system, if the mapping relationship table in the scenario mode has not been saved when the service system is switched to the scenario mode
- the resource occupancy situation is performed, and the migration scheduling of the virtual machine is performed according to a predetermined virtual machine scheduling algorithm
- the first update unit is configured to record and save the mapping relationship table in the scene mode according to a mapping relationship between the virtual machine and the physical server after the migration.
- the first scheduling unit includes:
- a migrating unit configured to wake up the physical server recorded in the mapping relationship table, and schedule the virtual machine recorded in the mapping relationship table to the corresponding physical server according to the mapping relationship in the mapping relationship table;
- the power saving unit is configured to perform power-off or sleep processing on the remaining physical servers except the physical server recorded in the mapping relationship table in the service system.
- the foregoing scheduling apparatus further includes:
- the third scheduling unit is configured to periodically collect resource usage of each physical server in the service system, and perform migration scheduling of the virtual machine according to the predetermined virtual machine scheduling algorithm according to the resource occupation situation; and After the migration scheduling of the virtual machine is performed, the mapping relationship table is updated according to the mapping relationship between the current virtual machine and the physical server of the service system.
- the working time period of the service system includes a pre-set and periodically presented busy hour period and idle time period;
- the second scheduling unit or the third scheduling unit is further configured to: when the scene mode is a busy time scenario corresponding to the busy time period, according to a load balancing algorithm, a physical server whose resource occupancy exceeds a predetermined threshold The part of the virtual machine is migrated to the physical server with the lowest resource occupancy rate in the service system; and when the scene mode is the idle time scene corresponding to the idle time period, according to the backpack algorithm, Business department The virtual machine is migrated to the least number of physical servers.
- the scheduling method and the scheduling device of the cloud computing-based service system can simplify the scheduling algorithm of the service cloud system, can directly and efficiently schedule the service system, and reduce the scheduling algorithm.
- the complexity and the scheduling "jitter" problem caused by the traditional scheduling algorithm are reduced, and the performance degradation of the service system caused by unnecessary virtual machine migration during peak traffic is reduced or avoided, and the stability of the service system is improved.
- FIG. 1 is a schematic flowchart of a service system-based scheduling method according to an embodiment of the present invention
- FIG. 2 is a schematic structural diagram of a service system-based scheduling apparatus according to an embodiment of the present invention.
- the multi-service cloud system built with cloud computing technology not only improves the utilization of resources, but also provides energy saving and reducing operating and maintenance costs.
- the telecommunication operator's business system has obvious tidal characteristics.
- One scheduling scheme adopted for the above tidal characteristics is as follows: during the day, all servers are powered up as the load increases. Then, all the virtual machines that carry the services use the load balancing scheduling principle to migrate the virtual machines to each server in the system. At night, an energy-efficient scheduling algorithm scheme is used, typically using a backpack algorithm to compress the virtual machine to a minimum of servers.
- the above-mentioned scheduling scheme lacks global considerations, and only considers the resource load situation at the time of scheduling, and does not take into account the difference in the load of the service system, which may cause scheduling scheduling problems in the middle of the system to generate scheduling jitter. For example, if the business system spreads the virtual machine to all servers according to the principle of CPU balancing, the CPU load considered is only the CPU load at the scheduling time. Over time, the peak load of each business system will gradually come.
- the load of each service system is different, for example, a virtual machine that carries a multimedia message center, the CPU usage at the peak time reaches 45%; and the virtual machine that carries the SMS center has a CPU usage of 55% at the peak, then
- the virtual machines of the two SMS centers are scheduled to run on a physical server, after the peak of the traffic load arrives, it is likely to trigger the re-scheduling to schedule a virtual machine running in it, for example, to migrate the MMS virtual machine to other physics.
- the server that is, scheduling jitter occurs. Scheduling jitter It is difficult to avoid in the scheduling algorithm of the related art.
- the scheduling of the multi-service cloud system of the traditional carrier does not take into account the difference of the load of each service system, nor does it predict the load of the service system. Therefore, when the service peaks easily, the virtual machine carrying the service system is overloaded by the physical server. High, the cloud management system performs scheduled migration. In the case of a virtual machine migration that occurs during peak business hours, the performance of the entire service system is degraded due to the performance degradation of the virtual machine during migration, resulting in instability of the entire business system.
- the embodiment of the present invention proposes a new scheduling algorithm to reduce or avoid the occurrence of the scheduling jitter phenomenon.
- the specific idea is to predict the future business situation by analyzing the historical load, thereby performing virtualization in a scientific and efficient manner.
- Machine scheduling The scheduling method of the embodiment of the present invention can simplify the scheduling algorithm of the multi-service cloud system, more directly and efficiently schedule the multi-service cloud system, reduce the complexity of the scheduling algorithm, and reduce the scheduling jitter caused by the traditional scheduling algorithm. "Problem, avoiding the performance degradation of business systems caused by unnecessary virtual machine migration during peak business.
- the energy-saving algorithm including the backpack algorithm is used to compress the virtual machine to run on as few physical servers as possible. After the compression is completed, a steady state is reached, and the virtual machine at this time is mapped with the physical server running thereby, and the mapping table at this time is recorded, corresponding to the night mode.
- the scheduler When the business system is switched from the night mode to the day mode, the scheduler directly migrates the virtual machine to the physical server to which it belongs according to the mapping relationship between the virtual machine and the physical server in the previously recorded daytime mode. Such scheduling is equivalent to restoring the system to the last steady state. Due to the existing various business systems including telecommunication service systems, the load is continuous and has a certain periodicity and regularity for a period of time, so it is usually possible to predict yesterday's steady state, and today it is also Steady state, or maximum probability steady state, is the most basic principle of the scheduling method of the embodiment of the present invention.
- the virtual machine migration scheduling can also be performed according to the mapping table in the steady state at the previously recorded night mode.
- the scheduling method proposed by the embodiment of the present invention predicts the load and virtual machine distribution of the next cycle according to the load of the previous cycle and the distribution of the virtual machine. Considering the historical load situation, the virtual machine can be directly transferred to the virtual machine as soon as possible. Steady state, its scheduling algorithm is extremely simple, and the scheduling effect is accurate and efficient.
- a scheduling method of a cloud computing-based service system provided by an embodiment of the present invention can be applied to a multi-service service.
- the system includes a plurality of physical servers and a plurality of virtual machines, where the virtual machines are used to process different services in the service system, and the scheduling method includes the following steps:
- Step 11 When the service system is switched to a scene mode, obtain a mapping relationship table between the virtual machine and the physical server of the service system in the scenario mode saved in advance, where the mapping relationship table records The mapping relationship between the virtual machine and the physical server that is scheduled by the service system according to the predetermined virtual machine scheduling algorithm in the scenario mode.
- the scenario mode corresponds to a periodic period of the service system.
- the working time period of the business system includes a pre-set and periodically presented busy hour period and idle time period.
- the scene corresponding to the busy time period is a busy time scene
- the scene corresponding to the idle time period is a idle time scene.
- the scene mode in the foregoing step 11 may be the daytime mode mentioned in the foregoing, that is, the busy time scene mode corresponding to the busy time period of the service system, or the night mode, that is, the service system corresponds to the idle time period. Free time scene mode.
- Step 12 The virtual machine of the service system is scheduled to the corresponding physical server according to the mapping relationship in the mapping relationship table.
- performing the scheduling process according to the mapping relationship in the mapping relationship table may include: waking up the physical server recorded in the mapping relationship table, and directly executing the physical server if the physical server is already in the awake state. Next, if the physical server is in a power-off or hibernation state, the physical server is woken up; then, the virtual machine recorded in the mapping relationship table is scheduled to the corresponding physical server according to the mapping relationship in the mapping relationship table. After the scheduling is completed, the remaining physical servers except the physical server recorded in the mapping relationship table in the service system are powered off or dormant to save energy.
- the virtual machine and the physical server of the service system are quickly obtained according to the mapping relationship table in the last other scene mode recorded in advance.
- the mapping relationship is restored to the most recent situation in the scene mode, that is, a steady state is reached quickly, so that the processing of the scheduling algorithm can be simplified, and the scheduling jitter steady state can be reduced or avoided.
- the mapping relationship table in the saved scene mode is not found when the service system is switched to the scene mode, it may indicate that the scene mode may be entered for the first time.
- the affiliation between the current virtual machine and the physical server in the service system periodically collects the resource usage of each physical server in the service system, and performs the migration scheduling of the virtual machine according to the predetermined virtual machine scheduling algorithm. A steady state.
- the mapping relationship table in the scenario mode is recorded and saved according to the mapping relationship between the virtual machine and the physical server after the migration, and if the mapping relationship table already exists, The mapping relationship table is updated according to the mapping relationship between the virtual machine and the physical server after the migration is scheduled.
- the embodiment of the present invention may further include the following steps: periodically collecting resource usage of each physical server in the service system, and scheduling according to the predetermined virtual machine according to the resource occupation situation.
- the algorithm performs a migration scheduling of the virtual machine.
- the mapping relationship table is updated according to a mapping relationship between the current virtual machine and the physical server of the service system.
- part of the virtual machines on the physical server whose resource occupancy exceeds a predetermined threshold may be migrated to resources in the service system according to a load balancing algorithm.
- the virtual machine of the service system may be migrated to a minimum number of physical servers according to a knapsack algorithm.
- the embodiment of the present invention can simplify the migration scheduling of the virtual machine of the service system, reduce or avoid the steady state of the scheduling jitter, and improve the performance and stability of the service system.
- a scheduling apparatus for a cloud computing-based service system includes:
- the obtaining unit 21 is configured to: when the service system is switched to a scene mode, obtain a mapping relationship table between the virtual machine and the physical server of the service system in the scenario mode saved in advance, where the mapping relationship table is Recording a mapping relationship between the virtual machine and the physical server that is scheduled by the service system according to a predetermined virtual machine scheduling algorithm in the scenario mode, and the scenario mode corresponds to a periodic time of the service system. segment;
- the first scheduling unit 22 is configured to schedule the virtual machine of the service system to the corresponding physical server according to the mapping relationship in the mapping relationship table.
- scheduling apparatus may further include:
- a second scheduling unit configured to periodically collect the physical relationship between the physical servers in the service system, if the mapping relationship table in the scenario mode has not been saved when the service system is switched to the scenario mode
- the resource occupancy situation is performed, and the migration scheduling of the virtual machine is performed according to a predetermined virtual machine scheduling algorithm
- the first update unit is configured to record and save the mapping relationship table in the scene mode according to a mapping relationship between the virtual machine and the physical server after the migration.
- the foregoing first scheduling unit 22 may include:
- a migrating unit configured to wake up the physical server recorded in the mapping relationship table, and schedule the virtual machine recorded in the mapping relationship table to the corresponding physical server according to the mapping relationship in the mapping relationship table;
- the power saving unit is configured to perform power-off or sleep processing on the remaining physical servers except the physical server recorded in the mapping relationship table in the service system.
- the scheduling device may further include:
- the third scheduling unit is configured to periodically collect resource usage of each physical server in the service system, and perform migration scheduling of the virtual machine according to the predetermined virtual machine scheduling algorithm according to the resource occupation situation; and After the migration scheduling of the virtual machine is performed, the mapping relationship table is updated according to the mapping relationship between the current virtual machine and the physical server of the service system.
- the working time period of the service system may include a pre-set and periodically presented busy hour period and idle time period.
- the second scheduling unit or the third scheduling unit is further configured to: when the scene mode is a busy time scenario corresponding to the busy time period, according to a load balancing algorithm, the resource occupancy rate exceeds a predetermined threshold. Part of the virtual machine on the physical server is migrated to the physical server with the lowest resource occupancy rate in the service system; and when the scene mode is the idle time scene corresponding to the idle time period, according to the backpack algorithm The virtual machine of the business system is migrated to a minimum number of physical servers.
- the scheduling method is divided into four phases, namely an initial daytime phase, an initial daytime phase, a normal daytime phase, and a normal night phase.
- Step 1 As an initial state, the service system randomly distributes the virtual machine (VM) to each physical server host (Host) according to the load balancing scheduling algorithm principle (an optional algorithm is a random spread algorithm), and establishes a VM and a Host.
- the mapping table for the daytime phase is a simple algorithm.
- Step 2 Dynamic scheduling according to the load situation.
- the strategy of the telecom operator is generally that the CPU usage of the host exceeds 70%, and VM scheduling is required. Therefore, the cloud management scheduling device detects the CPU usage of each host in real time. If a host with a CPU usage exceeding 70% selects the VM with the highest CPU and migrates to the host with the lightest load. At the same time, the mapping table between the VM and the host during the daytime phase is updated.
- Step 1 As an initial state, the service system compresses the VM (virtual machine) to the least Host (physical host) according to the principle of the energy-saving scheduling algorithm (an optional algorithm is a knapsack algorithm), and establishes a mapping of the night phase of the VM and the Host. Relational tables.
- VM virtual machine
- Host physical host
- an optional algorithm is a knapsack algorithm
- Step 1 When switching from the night mode to the day mode, the VM is quickly scheduled to the corresponding Host according to the previously established mapping relationship between the VM and the host daytime stage.
- Step 2 Real-time detection of the CPU usage of each host. If there is a host with a CPU usage exceeding 70%, select the VM with the highest CPU, migrate to the least loaded host, and update the VM and Host during the daytime. Mapping table.
- step two is performed periodically until the day and night mode transition, and the scheduling policy is switched.
- Step 1 According to the previously established mapping relationship between the VM and the night phase of the Host, the VM is quickly scheduled to the corresponding Host, and the idle Host is put into sleep or power-off operation to save power and air conditioning refrigeration consumption.
- the scheduling algorithm and the scheduling device provided by the embodiments of the present invention can effectively schedule the multi-service cloud system of the operator, and can simplify the scheduling algorithm, realize energy saving, and provide services on the basis of ensuring system performance. System performance and reliability, specific high practicality.
- the scheduling method and the scheduling device of the cloud computing-based service system can simplify the scheduling algorithm of the service cloud system, can directly and efficiently schedule the service system, reduce the complexity of the scheduling algorithm, and reduce the complexity.
- the scheduling "jitter" problem caused by the traditional scheduling algorithm reduces or avoids the performance degradation of the service system caused by unnecessary virtual machine migration during peak traffic, and improves the stability of the service system.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Power Sources (AREA)
Abstract
一种基于云计算的业务系统的调度方法及调度装置。通过对历史负载的分析，来预测未来的业务情况，从而科学高效地进行虚拟机调度。可以简化多业务云系统的调度算法，更加直接高效地对多业务云系统进行调度，降低了调度算法的复杂度，并且减少传统调度算法造成的调度"抖动"问题，避免业务高峰时不必要的虚拟机迁移造成的业务系统性能下降问题。
Description
本发明涉及云计算资源调度方法，具体而言，涉及一种基于云计算的业务系统的调度方法及调度装置。
电信运营商的增值业务系统，例如短信中心、彩信中心、WAP网关等业务系统，是运营商收入的重要来源，因此增值业务系统的建设一直备受运营商重视。在传统建设方案中，每一套增值业务系统，都使用专有的服务器资源。例如，彩信中心使用的服务器，由彩信中心独享。短信中心所使用的服务器，由短信中心独享。这种独立烟囱式的建设方式，一个显而易见的缺点是物理服务器资源浪费严重。
随着虚拟化、云计算技术的发展，增值业务系统这种独立的、烟囱式的建设模式，逐渐被云计算模式取代。在云计算模式下，首先将资源池化，即通过虚拟化技术，将物理服务器虚拟化成多个虚拟机，业务系统承载在虚拟机上。这样将物理服务器资源池化后，添加了一层虚拟化层，解耦了业务系统与物理服务器之间的耦合关系，使得多个业务系统复用一组物理服务器成为可能。这种使用云计算技术建设的多业务云系统，不但提高了资源的利用率，也为节能降耗，降低运维成本提供了可能。
电信运营商的业务系统，存在着明显的潮汐特性，例如白天，由于人们的通信需求旺盛使得业务系统的负载呈现峰值，夜晚由于人们的通信需求降低，业务系统的负载将会非常显著地下降。所以，采用云计算技术的多业务云系统，往往借助虚拟化的迁移特性，进行智能化的虚拟机调度，以达到减少电力消耗，降低成本的效果。现有技术的一种典型方案是，在白天，将所有承载业务的虚拟机平均分布在资源池的每一台服务器中，系统以最佳的性能状态承载业务高峰的到来。在夜晚低负载时，将承载业务的虚拟机，压缩迁移到最少的服务器中运行，同时将没有运行虚拟机的空闲物理服务器，进行下电或休眠操作，以减少这些服务器对电能的消耗，从而达到降低成本的目的。
然而，现行的调度算法的问题是，缺乏全局考虑，只考虑在调度时刻的资源负载情况，没有考虑到业务系统不同业务负载之间的差异性，会带来系统中间运行过程中再次发生调度从而产生调度抖动问题。
发明内容
本发明实施例要解决的技术问题是提供一种基于云计算的业务系统的调度方法及调度装置，用以降低调度算法的复杂度，并减少传统调度算法造成的调度抖动问题。
根据本发明的一个方面，提供了一种基于云计算的业务系统的调度方法，包括：
在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段；
按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
可选地，在所述业务系统切换至所述场景模式时，若尚未保存有所述场景模式下的所述映射关系表，则周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表。
可选地，所述按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器，包括：将所述映射关系表中记录的物理服务器唤醒，并按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；以及，将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理。
可选地，在所述按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器的步骤之后，所述方法还包括：周期性的采集所述业务系统中各个物理服务器的资源占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
可选地，所述业务系统的工作时间段包括预先设定的、且周期性呈现的忙时时段和闲时时段；所述根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度，包括：在所述场景模式为对应于所述忙时时段时的忙时场景时，按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；在所述场景模式为对应于所述闲时时段时的闲时场景时，按照背包算法，将所述业务系统的虚拟机迁移至最少数量的物理服务器上。
根据本发明实施例的另一个方面，还提供了一种基于云计算的业务系统的调度装置，包括：
获取单元，设置为在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段；
第一调度单元，设置为按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
可选地，上述调度装置中还包括：
第二调度单元，设置为在所述业务系统切换至所述场景模式时，若尚未保存有所述场景模式下的所述映射关系表，则周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度；
第一更新单元，设置为根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表。
可选地，上述调度装置中，所述第一调度单元包括：
迁移单元，设置为将所述映射关系表中记录的物理服务器唤醒，并按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；以及，
节能单元，设置为将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理。
可选地，上述调度装置中，还包括：
第三调度单元，设置为周期性的采集所述业务系统中各个物理服务器的资源占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
可选地，上述调度装置中，所述业务系统的工作时间段包括预先设定的、且周期性呈现的忙时时段和闲时时段；
所述第二调度单元或第三调度单元，还设置为在所述场景模式为对应于所述忙时时段时的忙时场景时，按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；以及，在所述场景模式为对应于所述闲时时段时的闲时场景时，按照背包算法，将所述业务系
统的虚拟机迁移至最少数量的物理服务器上。
与相关技术相比，本发明实施例提供的基于云计算的业务系统的调度方法及调度装置，可以简化业务云系统的调度算法，能够更加直接高效地对业务系统进行调度，降低了调度算法的复杂度，并且减少了传统调度算法造成的调度“抖动”问题，减少或避免了在业务高峰时不必要的虚拟机迁移所造成的业务系统性能下降问题，提高了业务系统的稳定性。
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1为本发明实施例提供的基于业务系统的调度方法的流程示意图；
图2为本发明实施例提供的基于业务系统的调度装置的结构示意图。
为使本发明要解决的技术问题、技术方案和优点更加清楚，下面将结合附图及具体实施例进行详细描述。
使用云计算技术建设的多业务云系统，不但提高了资源的利用率，也为节能降耗，降低运维成本提供了可能。电信运营商的业务系统，存在着明显的潮汐特性，针对上述潮汐特性所采用的一种调度方案如下，即在白天，随着负载增加，将所有服务器加电。然后将所有承载业务的虚拟机，使用负载均衡的调度原则，将虚拟机平均迁移到系统中的每一台服务器上。在夜晚，使用节能调度算法方案，典型地是采用背包算法，将虚拟机压缩到最少的服务器上运行。
上述调度方案缺乏全局考虑，只考虑在调度时刻的资源负载情况，没有考虑到业务系统负载的差异性，会带来系统中间运行过程中再次发生调度从而产生调度抖动问题。举例说明，假设业务系统在按照CPU均衡的原则，将虚拟机散开到所有服务器上时，考虑到的CPU负载只是调度时刻点的CPU负载。随着时间的推移，各个业务系统的负载高峰将会逐渐到来。由于每一种业务系统的负载是有差异的，例如承载彩信中心的虚拟机，高峰时的CPU使用率达到45％；承载短信中心的虚拟机，高峰时的CPU使用率达到55％，那么如果将两个短信中心的虚拟机调度到一台物理服务器上运行时，在业务负载高峰到来后，很可能触发再次调度，将其中运行的一个虚拟机调度走，例如将彩信虚拟机迁移到其他物理服务器，即发生调度抖动。这种调度抖动现象
在相关技术的调度算法中难以避免。
传统运营商多业务云系统的调度，由于没有考虑到各个业务系统负载的差异性，也没有对业务系统的负载进行预测，因此容易造成业务高峰时，承载业务系统的虚拟机由于物理服务器负载过高，被云管理系统执行调度迁移。而在业务高峰时发生的虚拟机迁移，则会因为虚拟机在迁移时本身的性能下降而造成整个业务系统的性能下降，从而导致整个业务系统的不稳定。
为解决上述问题，本发明实施例提出一种新的调度算法，减少或避免上述调度抖动现象发生，其具体思路是通过对历史负载的分析，来预测未来的业务情况，从而科学高效地进行虚拟机调度。采用本发明实施例所述调度方法，可以简化多业务云系统的调度算法，更加直接高效地对多业务云系统进行调度，降低了调度算法的复杂度，并且减少传统调度算法造成的调度“抖动”问题，避免业务高峰时不必要的虚拟机迁移造成的业务系统性能下降问题。
在对现有的业务系统进行研究后发现，在业务系统运行时，如果白天采用负载均衡类的调度算法进行调度，最终会达到一个稳态，即虚拟机不会再由于业务系统的负载变化情况而造成虚拟机的迁移，此时，称之为稳态。将此时所有物理服务器与其上运行的所有虚拟机之间的关系，按照隶属关系进行映射。形成一个稳态的映射关系表，对应于白天模式。
同理，在夜晚，采用包括背包算法在内的节能算法，将虚拟机压缩到尽量少的物理服务器上运行。在压缩完毕后，达到一个稳态，也将此时的虚拟机和其运行的物理服务器进行映射，同时记下此时的映射关系表，对应于夜晚模式。
在业务系统由夜晚模式向白天模式切换时，调度器按照之前记录的白天模式下的虚拟机与物理服务器的映射关系表，直接将虚拟机迁移到所归属的物理服务器上。这样的调度，相当于将系统恢复到上一次的稳态状态。由于现有的包括电信业务系统在内的各种业务系统，其负载在一段时间内是连续的、且具有特定的周期性和规律性，所以通常可以预测昨天的稳态，在今天来说也是稳态的，或者说是最大概率稳态的，这是本发明实施例调度方法的最基本原理。
同理，对于白天模式向夜晚模式切换时，也可以依据之前记录的夜晚模式下的稳态时的映射关系表，进行虚拟机的迁移调度。
本发明实施例提出的调度方法，根据上一个周期的负载与虚拟机分布，来预测下一个周期的负载与虚拟机分布，考虑到了历史负载情况，能够较为精确的直接将虚拟机尽快迁移至一稳态，其调度算法极为简单，并且调度的效果准确高效。请参照图1，本发明实施例提供的基于云计算的业务系统的调度方法，可以应用于一多业务的业务
系统，该业务系统包括有多个物理服务器和多个虚拟机，所述虚拟机用于处理业务系统中的不同业务，该调度方法包括以下步骤：
步骤11，在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段。
这里，所述业务系统的工作时间段包括预先设定的、且周期性呈现的忙时时段和闲时时段。对应于所述忙时时段时的场景为忙时场景，对应于所述闲时时段时的场景为闲时场景。具体的，上述步骤11中的场景模式可以是前文中提及的白天模式，即业务系统在忙时时间段对应的忙时场景模式，也可以是夜晚模式，即业务系统在闲时时间段对应的闲时场景模式。
步骤12，按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
这里，上述步骤12中，按照所述映射关系表中的映射关系执行调度处理具体可以包括：将所述映射关系表中记录的物理服务器唤醒，如果该物理服务器已经处于唤醒状态，则可以直接执行下一步，如果该物理服务器处于下电或休眠状态，则唤醒该物理服务器；然后，按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；在调度完成后，将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理，以节约能耗。
通过以上步骤，本实施例在业务系统从一个场景模式切换到另一场景模式时，根据预先记录的最近一次该另一场景模式下的映射关系表，快速的将业务系统的虚拟机与物理服务器的映射关系恢复为最近一次该场景模式下的情形，即快速达到一稳态，从而可以简化调度算法的处理，并且减少或避免调度抖动稳态。
以上步骤11中，如果在所述业务系统切换至所述场景模式时，未能够找到保存的所述场景模式下的所述映射关系表，则表明可能是初次进入该场景模式，此时可以基于业务系统中当前的虚拟机和物理服务器的隶属关系，周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度，使之到达一个稳态。并且，在每次发生迁移调度后，根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表，若已存在所述映射关系表，则根据迁移调度后的虚拟机与物理服务器之间的映射关系，更新所述映射关系表。
由于业务系统的负载并非一直不变的，在恢复为前一次场景模式下的稳态后，可
能还需要根据业务负载的当前变化情况，继续对虚拟机进行调度迁移，以适应新的变化。此时，本发明实施例在上述步骤12之后，还可以包括以下步骤：周期性的采集所述业务系统中各个物理服务器的资源占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；然后，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
本发明实施例中，在根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度时，针对不同的场景可以有不同的处理，例如：
在所述场景模式为对应于所述忙时时段时的忙时场景时，可以按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；
在所述场景模式为对应于所述闲时时段时的闲时场景时，可以按照背包算法，将所述业务系统的虚拟机迁移至最少数量的物理服务器上。
通过以上方法，本发明实施例可以简化业务系统的虚拟机的迁移调度，减少或避免调度抖动稳态，提高业务系统的性能和稳定性。
基于以上方法，本发明实施例还提供了一种用以实施上述方法的装置。请参照图2所示，本发明实施例提供的基于云计算的业务系统的调度装置，包括：
获取单元21，设置为在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段；
第一调度单元22，设置为按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
进一步的，上述调度装置还可以包括：
第二调度单元，设置为在所述业务系统切换至所述场景模式时，若尚未保存有所述场景模式下的所述映射关系表，则周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度；
第一更新单元，设置为根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表。
可选的，上述第一调度单元22可以包括：
迁移单元，设置为将所述映射关系表中记录的物理服务器唤醒，并按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；以及，
节能单元，设置为将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理。
为适应业务系统的业务负载的变化，上述调度装置还可以包括：
第三调度单元，设置为周期性的采集所述业务系统中各个物理服务器的资源占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
可选的，所述业务系统的工作时间段可以包括预先设定的、且周期性呈现的忙时时段和闲时时段。此时，所述第二调度单元或第三调度单元，进一步用于在所述场景模式为对应于所述忙时时段时的忙时场景时，按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；以及，在所述场景模式为对应于所述闲时时段时的闲时场景时，按照背包算法，将所述业务系统的虚拟机迁移至最少数量的物理服务器上。
下面，再通过更为具体的一个调度方法的示例，对本发明作更为详细的描述。本调度方法共分四个阶段，即初始白天阶段、初始夜晚阶段、常态白天阶段和常态夜晚阶段。
初始白天阶段：
步骤一：作为初始状态，业务系统根据负载均衡调度算法原则(一个可选的算法是随机散开算法)，将虚拟机(VM)随机分散到各个物理服务器主机(Host)上，建立VM与Host的白天阶段的映射关系表。
步骤二：根据负载情况进行动态调度。电信运营商的策略，一般是Host的CPU使用率超过70％，就需要进行VM调度。因此，云管理调度装置实时侦测每台Host的CPU使用率，如有CPU使用率超过70％的Host，选择其上占用CPU最高的VM，迁移到负载最轻的Host上。同时更新VM与Host白天阶段的映射关系表。
反复执行步骤二，直到白天模式转换为夜晚模式为止。
初始夜晚阶段：
步骤一：作为初始状态，业务系统根据节能调度算法原则(一个可选的算法是背包算法)，将VM(虚拟机)压缩到最少Host(物理主机)上，建立VM与Host的夜晚阶段的映射关系表。
常态白天阶段
步骤一：当由夜晚模式切换到白天模式时，根据之前已建立的VM与Host白天阶段的映射关系表，快速将VM调度到对应的Host上。
步骤二：实时侦测每台Host的CPU占用率，如有CPU使用率超过70％的Host，选择其上占用CPU最高的VM，迁移到负载最轻的Host上，同时更新VM与Host白天阶段的映射关系表。
周期性执行上述步骤二，直到白天夜晚模式转换，调度策略切换为止。
常态夜晚阶段：
步骤一：根据之前建立的VM与Host的夜晚阶段的映射关系表，快速将VM调度到对应的Host上，将空闲的Host进行休眠或下电操作，以节省电力和空调制冷消耗。
综上所述，本发明实施例提供的调度算法和调度装置，能够有效地对运营商多业务云系统进行调度，在保证系统性能的基础上，能够简化调度算法，实现节能降耗，提供业务系统的性能和可靠性，具体较高的实用性。
以上所述是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。
本发明实施例提供的基于云计算的业务系统的调度方法及调度装置，可以简化业务云系统的调度算法，能够更加直接高效地对业务系统进行调度，降低了调度算法的复杂度，并且减少了传统调度算法造成的调度“抖动”问题，减少或避免了在业务高峰时不必要的虚拟机迁移所造成的业务系统性能下降问题，提高了业务系统的稳定性。
Claims (10)
- 一种基于云计算的业务系统的调度方法，包括：在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段；按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
- 如权利要求1所述的方法，其中，在所述业务系统切换至所述场景模式时，若尚未保存有所述场景模式下的所述映射关系表，则周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表。
- 如权利要求1所述的方法，其中，所述按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器，包括：将所述映射关系表中记录的物理服务器唤醒，并按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；以及，将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理。
- 如权利要求1所述的方法，其中，在所述按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器的步骤之后，所述方法还包括：周期性的采集所述业务系统中各个物理服务器的资源占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
- 如权利要求2或4所述的方法，其中，所述业务系统的工作时间段包括预先设定的、且周期性呈现的忙时时段和闲时时段；所述根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度，包括：在所述场景模式为对应于所述忙时时段时的忙时场景时，按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；在所述场景模式为对应于所述闲时时段时的闲时场景时，按照背包算法，将所述业务系统的虚拟机迁移至最少数量的物理服务器上。
- 一种基于云计算的业务系统的调度装置，包括：获取单元，设置为在所述业务系统切换至一场景模式时，获取预先保存的所述场景模式下所述业务系统的虚拟机与物理服务器之间的映射关系表，所述映射关系表中记录有所述业务系统最近一次在所述场景模式下按照预定的虚拟机调度算法调度得到的虚拟机与物理服务器的映射关系，所述场景模式对应于所述业务系统的呈周期性的一时间段；第一调度单元，设置为按照所述映射关系表中的映射关系，将所述业务系统的虚拟机调度至对应的物理服务器。
- 如权利要求6所述的调度装置，其中，还包括：第二调度单元，设置为在所述业务系统切换至所述场景模式时，若尚未保存有所述场景模式下的所述映射关系表，则周期性的采集所述业务系统中各个物理服务器的资源占用情况，并按照预定的虚拟机调度算法，执行虚拟机的迁移调度；第一更新单元，设置为根据迁移调度后的虚拟机与物理服务器之间的映射关系，记录并保存所述场景模式下的所述映射关系表。
- 如权利要求6所述的调度装置，其中，所述第一调度单元包括：迁移单元，设置为将所述映射关系表中记录的物理服务器唤醒，并按照所述映射关系表中的映射关系，将所述映射关系表中记录的虚拟机调度至对应的物理服务器；以及，节能单元，设置为将所述业务系统中除所述映射关系表中记录的物理服务器之外的剩余物理服务器，进行下电或休眠处理。
- 如权利要求6所述的调度装置，其中，还包括：第三调度单元，设置为周期性的采集所述业务系统中各个物理服务器的资源 占用情况，并根据所述资源占用情况，按照预定的虚拟机调度算法，执行虚拟机的迁移调度；以及，在执行完所述虚拟机的迁移调度后，根据所述业务系统当前的虚拟机和物理服务器之间的映射关系，更新所述映射关系表。
- 如权利要求7或9所述的调度装置，其中，所述业务系统的工作时间段包括预先设定的、且周期性呈现的忙时时段和闲时时段；所述第二调度单元或第三调度单元，还设置为在所述场景模式为对应于所述忙时时段时的忙时场景时，按照负载均衡算法，将资源占用率超出预定门限的物理服务器上的部分虚拟机，迁移至所述业务系统中资源占用率最低的物理服务器上；以及，在所述场景模式为对应于所述闲时时段时的闲时场景时，按照背包算法，将所述业务系统的虚拟机迁移至最少数量的物理服务器上。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510035570.9A CN105868004B (zh) | 2015-01-23 | 2015-01-23 | 一种基于云计算的业务系统的调度方法及调度装置 |
| CN201510035570.9 | 2015-01-23 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016115956A1 true WO2016115956A1 (zh) | 2016-07-28 |
Family
ID=56416401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/098308 WO2016115956A1 (zh) | 2015-01-23 | 2015-12-22 | 基于云计算的业务系统的调度方法及调度装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105868004B (zh) |
| WO (1) | WO2016115956A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107145216A (zh) * | 2017-05-05 | 2017-09-08 | 北京景行锐创软件有限公司 | 一种调度方法 |
| US10417035B2 (en) | 2017-12-20 | 2019-09-17 | At&T Intellectual Property I, L.P. | Virtual redundancy for active-standby cloud applications |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247617B (zh) * | 2017-05-17 | 2020-11-24 | 北京神州数码云科信息技术有限公司 | 虚拟机资源的调配方法、试用平台及可读存储介质 |
| CN110609747A (zh) * | 2019-08-29 | 2019-12-24 | 联想(北京)有限公司 | 信息处理方法及电子设备 |
| CN112631744A (zh) * | 2019-09-24 | 2021-04-09 | 阿里巴巴集团控股有限公司 | 进程处理方法、装置、电子设备及计算机可读存储介质 |
| CN111104203B (zh) * | 2019-12-13 | 2023-04-28 | 广东省华南技术转移中心有限公司 | 虚拟机分散调度方法、装置以及电子设备、存储介质 |
| CN112667392B (zh) * | 2020-12-09 | 2024-01-23 | 南方电网数字电网研究院有限公司 | 云计算资源分配方法、装置、计算机设备和存储介质 |
| CN112866131B (zh) * | 2020-12-30 | 2023-04-28 | 神州绿盟成都科技有限公司 | 一种流量负载均衡方法、装置、设备及介质 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100070784A1 (en) * | 2008-09-15 | 2010-03-18 | Vmware, Inc. | Reducing Power Consumption in a Server Cluster |
| CN102546700A (zh) * | 2010-12-23 | 2012-07-04 | 中国移动通信集团公司 | 一种资源调度以及资源迁移的方法及设备 |
| US20120266163A1 (en) * | 2011-04-13 | 2012-10-18 | International Business Machines Corporation | Virtual Machine Migration |
| CN102890554A (zh) * | 2011-07-22 | 2013-01-23 | 鸿富锦精密工业（深圳）有限公司 | 电源管理系统及方法 |
| CN103905303A (zh) * | 2012-12-28 | 2014-07-02 | 中国移动通信集团公司 | 一种虚拟机vm跨网迁移后的数据处理方法、装置及系统 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8850434B1 (en) * | 2012-09-14 | 2014-09-30 | Adaptive Computing Enterprises, Inc. | System and method of constraining auto live migration of virtual machines using group tags |
| CN104077171B (zh) * | 2013-03-28 | 2017-12-15 | 华为技术有限公司 | 调度虚拟机时的处理方法和设备 |
| JP2015011569A (ja) * | 2013-06-28 | 2015-01-19 | 株式会社東芝 | 仮想マシン管理装置、仮想マシン管理方法、及び仮想マシン管理プログラム |
| CN103747085A (zh) * | 2014-01-10 | 2014-04-23 | 浪潮电子信息产业股份有限公司 | 云计算操作系统下一种存储资源调度算法 |
| CN103957231B (zh) * | 2014-03-18 | 2015-08-26 | 成都盛思睿信息技术有限公司 | 一种云计算平台下的虚拟机分布式任务调度方法 |
-
2015
- 2015-01-23 CN CN201510035570.9A patent/CN105868004B/zh active Active
- 2015-12-22 WO PCT/CN2015/098308 patent/WO2016115956A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100070784A1 (en) * | 2008-09-15 | 2010-03-18 | Vmware, Inc. | Reducing Power Consumption in a Server Cluster |
| CN102546700A (zh) * | 2010-12-23 | 2012-07-04 | 中国移动通信集团公司 | 一种资源调度以及资源迁移的方法及设备 |
| US20120266163A1 (en) * | 2011-04-13 | 2012-10-18 | International Business Machines Corporation | Virtual Machine Migration |
| CN102890554A (zh) * | 2011-07-22 | 2013-01-23 | 鸿富锦精密工业（深圳）有限公司 | 电源管理系统及方法 |
| CN103905303A (zh) * | 2012-12-28 | 2014-07-02 | 中国移动通信集团公司 | 一种虚拟机vm跨网迁移后的数据处理方法、装置及系统 |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107145216A (zh) * | 2017-05-05 | 2017-09-08 | 北京景行锐创软件有限公司 | 一种调度方法 |
| US10417035B2 (en) | 2017-12-20 | 2019-09-17 | At&T Intellectual Property I, L.P. | Virtual redundancy for active-standby cloud applications |
| US10990435B2 (en) | 2017-12-20 | 2021-04-27 | At&T Intellectual Property I, L.P. | Virtual redundancy for active-standby cloud applications |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105868004B (zh) | 2020-10-16 |
| CN105868004A (zh) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016115956A1 (zh) | 基于云计算的业务系统的调度方法及调度装置 | |
| CN110888714B (zh) | 容器的调度方法、装置和计算机可读存储介质 | |
| CA2741088C (en) | Methods of achieving cognizant power management | |
| KR101624765B1 (ko) | 에너지-인식 서버 관리 | |
| WO2016041468A1 (zh) | 一种唤醒方法、装置及终端 | |
| Gu et al. | Energy efficient scheduling of servers with multi-sleep modes for cloud data center | |
| CN108023958B (zh) | 一种基于云平台资源监视的资源调度系统 | |
| Enokido et al. | An extended power consumption model for distributed applications | |
| US9274585B2 (en) | Combined dynamic and static power and performance optimization on data centers | |
| CN102929720A (zh) | 一种节能作业调度系统 | |
| CN111625080B (zh) | 一种服务器节能方法、装置及电子设备和存储介质 | |
| Mao et al. | A multi-resource task scheduling algorithm for energy-performance trade-offs in green clouds | |
| CN112954707B (zh) | 基站的节能方法、装置、基站和计算机可读存储介质 | |
| CN110633152A (zh) | 用于实现业务集群水平伸缩的方法和装置 | |
| CN103108039A (zh) | 一种低能耗集群环境下的服务质量保证方法 | |
| CN103414784B (zh) | 支持应急模式的云计算资源调度方法 | |
| Orgerie et al. | ERIDIS: energy-efficient reservation infrastructure for large-scale distributed systems | |
| CN114327023B (zh) | 一种Kubernetes集群的节能方法、系统、计算机介质和电子设备 | |
| WO2011113321A1 (zh) | 一种能耗控制方法及装置 | |
| CN117251044A (zh) | 一种基于arima技术的云服务器动态能耗管理方法和系统 | |
| CN104883725B (zh) | 一种长期演进网络中基于站点实际负载的网络节能方法 | |
| CN113992687B (zh) | 智能业务集群调度方法、装置、电子设备及存储介质 | |
| CN110806918A (zh) | 基于深度学习神经网络的虚拟机运行方法和装置 | |
| Zhang et al. | An energy-aware task scheduling algorithm for a heterogeneous data center | |
| CN111246549A (zh) | 一种节点休眠、唤醒时间提供的方法及装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15878624 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15878624 Country of ref document: EP Kind code of ref document: A1 |