WO2017162034A1 - 一种负载方法和系统 - Google Patents

一种负载方法和系统 Download PDF

Info

Publication number
WO2017162034A1
WO2017162034A1 PCT/CN2017/076039 CN2017076039W WO2017162034A1 WO 2017162034 A1 WO2017162034 A1 WO 2017162034A1 CN 2017076039 W CN2017076039 W CN 2017076039W WO 2017162034 A1 WO2017162034 A1 WO 2017162034A1
Authority
WO
WIPO (PCT)
Prior art keywords
load
standby
load device
temporary
service
Prior art date
Application number
PCT/CN2017/076039
Other languages
English (en)
French (fr)
Inventor
蒋海涛
林洁
Original Assignee
阿里巴巴集团控股有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2017162034A1 publication Critical patent/WO2017162034A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/40Support for services or applications
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • H04L67/1001Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004Server selection for load balancing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • H04L67/1001Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers

Definitions

  • the present invention relates to the field of network technologies, and in particular, to a load method and system.
  • the capacity expansion scheme in the prior art is horizontal expansion, that is, starting more expansion servers to support the current traffic load.
  • the waiting time for expanding a new server is usually a few minutes or longer. If the business load continues to rise during this period, once the processing capacity of the online server is exceeded, the service will be abnormal, unavailable, or even suspended.
  • the present invention provides a load method.
  • the temporary load device is invoked to perform a service load, and the standby load device is started.
  • the standby load device is started, the service of the temporary load device is started.
  • the load is switched to the standby load device.
  • the invention avoids the service abnormality, unavailability or even suspension phenomenon caused by the continuous increase of the service load during the startup of the standby load and exceeding the service load capacity of the primary load device.
  • the present invention provides a load method applied to a system including a primary load device, a temporary load device, and a backup load device, wherein the primary load device is a device currently carrying a traffic load, and the temporary load device A device that can perform a temporary service load at any time, and the standby device is a device that can perform a service load after being started, and the method includes:
  • the temporary load device is invoked to perform a traffic load, and the standby load device is started;
  • the service load of the temporary load device is switched to the standby load device, so that the standby load device performs a traffic load.
  • the load capacity of the backup load device is different, and the load capacity of the standby load device is different, and the startup time of the standby load device is higher.
  • the standby load device is started, and specifically includes:
  • the standby load device corresponding to the load capacity is started according to the growth rate and the startup time of each of the standby load devices.
  • the standby load device is multiple, and each of the standby load devices has the same load capacity
  • the standby load device is started, and specifically includes:
  • a corresponding number of the standby load devices are activated according to the growth rate and the startup time of the standby load device.
  • the method further includes:
  • the standby load device When the traffic load of the standby load device is zero, the standby load device is turned off.
  • a system includes a primary load device, a temporary load device, and a backup load device, wherein the primary load device is a device that is currently carrying a traffic load, and the temporary load device is capable of performing temporary traffic load at any time.
  • the standby device is a device that can perform a service load after being started, and the system includes:
  • a determining module configured to determine whether a service load value of the primary load device exceeds a threshold
  • Calling a module if the threshold is exceeded, used to invoke the temporary load device to perform a service load; and a startup module, configured to start the standby load device;
  • the switching module is configured to switch the service load of the temporary load device to the standby load device after the startup of the standby load device is completed, so that the standby load device performs a traffic load.
  • the load capacity of the backup load device is different, and the load capacity of the standby load device is different, and the startup time of the standby load device is higher.
  • the startup module is specifically configured to:
  • the standby load device corresponding to the load capacity is started according to the growth rate and the startup time of each of the standby load devices.
  • the standby load device is multiple, and each of the standby load devices has the same load capacity
  • the startup module is specifically configured to:
  • a corresponding number of the standby load devices are activated according to the growth rate and the startup time of the standby load device.
  • the switching module is specifically configured to:
  • the system further includes:
  • the shutdown module is configured to shut down the standby load device after the traffic load of the temporary load device is switched to the standby load device.
  • the temporary load device when the service load value of the primary load device exceeds the threshold, the temporary load device is invoked to perform a traffic load, and the standby load device is started, and when the standby load device is started, the traffic load of the temporary load device is switched to the On the standby load device.
  • the invention avoids the service abnormality, unavailability or even suspension phenomenon caused by the continuous increase of the service load during the startup of the standby load and exceeding the service load capacity of the primary load device.
  • FIG. 1 is a loading method according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a location of an online server, a capacity expansion server, and a buffer server according to an embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of a system according to an embodiment of the present invention.
  • the present invention provides a load method.
  • the specific method is as shown in FIG. 1 , and the method is applied to a system including a primary load device, a temporary load device, and a backup load device, where
  • the primary load device is a device that is currently carrying a service load
  • the temporary load device is a device that can perform a temporary service load at any time
  • the standby device is a device that can perform a service load after being started, and the method includes the following steps. :
  • Step 101 Determine whether the service load value of the primary load device exceeds a threshold. If yes, execute step 102. If not, end the process.
  • the primary load device and the backup load device are capable of providing stable load services and providing users with stable, high-quality services, stability of the primary load device and the backup load device, and service load. Capability, ability to provide services, and hardware configuration and hardware performance are higher than the temporary load device, the primary load device and the standby load device are laterally initiated, that is, the load capacity of the primary load device exceeds a threshold The standby load device is only started, and further, the method can be specifically used in a scenario of lateral startup.
  • the temporary load device is in a state of being on, that is, in a state in which a service load can be performed at any time, and when the main load device exceeds a threshold, the temporary load device is used to perform a traffic load, so that the temporary load device is the
  • the primary load device shares the load, and the cost consumed by the temporary load device when it is in the on state is much lower than the cost consumed when the primary load device and the backup load device are in an open state, the cost including Such as: maintenance costs, consumption costs, and resource occupation costs.
  • the service load value of the primary load device When the service load value of the primary load device is close to the threshold, it indicates that the service load capacity that the primary load device can provide is already small. If the traffic load continues to grow, the service load of the primary load device is exceeded. Capability, the standby load device needs to be started laterally.
  • Step 102 Call the temporary load device to perform a traffic load, and start the standby load device.
  • the temporary load device is invoked due to The temporary load device is always in an open state, so the temporary load device can share the traffic load for the primary load device at any time when the primary load device exceeds its load capacity, in the process of starting the standby load device.
  • the temporary load device can provide stable load service and provide users with stable, high quality services.
  • the backup load device has multiple load capacities, and the load capacities of the multiple backup load devices may be the same or different, and the backup load device with a higher load capacity has a longer startup time.
  • the standby load device corresponding to the load capacity is started according to the growth rate and the startup time of each of the standby load devices.
  • the growth rate and the startup of each of the standby load devices are required according to the growth rate.
  • Time to start the standby load device which can save system resources and costs.
  • the standby load device with a traffic load capacity of 10 needs to be started; if the traffic load growth rate is 3, and if the service load capacity exceeds 10, The threshold of the standby load device, but does not exceed the threshold of the standby load device with a traffic load capacity of 15, then the standby load device with a traffic load capacity of 15 needs to be started, if the standby with a service load capacity of 10 is not exceeded The threshold of the load device, then the standby load set with a service load capacity of 10 needs to be started. .
  • the standby load device that determines which level of traffic load is used in the system can be determined according to actual conditions.
  • the standby load device which specifically includes:
  • a corresponding number of the standby load devices are activated according to the growth rate and the startup time of the standby load device.
  • the startup time of the multiple standby load devices is the same, any one of them can be started at the time of startup, but the backup load device can be loaded to exceed the service load of the primary load device, and the startup is guaranteed.
  • a reasonable number of the standby load devices are required to start the number of the standby load devices according to the growth rate and the startup time of each of the standby load devices.
  • the backup load devices have three, and the service load is The capability is 10 and the startup time is 3 minutes. If the current traffic load growth rate is 1, then the number of startups needs to be 1; if the current traffic load growth rate is 3, and if one of the backups is exceeded If the threshold of the load device is 2, the number of devices to be started is 2. If the threshold of the standby load device is not exceeded, the number of devices to be started is 1. If the current service load growth rate is 4, then you need to enable The number of moves is two.
  • the standby load device that determines which level of traffic load is used in the system can be determined according to actual conditions.
  • one of the plurality of the standby load devices may be arbitrarily activated. If the load capacity of the standby load device that is started cannot meet the traffic load exceeding the load capacity of the primary load device, then the other backup load device is arbitrarily activated until the load capacity of the standby load device that is started can meet the excess The traffic load of the primary load device load capacity.
  • the standby load device may also be one.
  • the load capacity of the backup load device is determined according to the daily traffic load, that is, the service load capability of the backup load device is sufficiently strong, and the service of the primary load device is After the load value exceeds the threshold, the temporary load device temporarily performs the traffic load. After the standby load device is started, the service on the temporary load device is switched to the standby load device to perform the traffic load.
  • Step 103 After the startup of the standby load device is completed, the service load of the temporary load device is switched to the standby load device, so that the standby load device performs a traffic load.
  • the service load of the temporary load device needs to be switched to the standby. Load the device, and then let the standby load device perform a traffic load.
  • the temporary load device performs a service load by using an application running by the device, and when the standby load device is started, only the application needs to be closed, the service load is released, and the system releases the service.
  • the load is switched to the activated standby load device, so that the effect of releasing the traffic load can be achieved without shutting down the temporary load device, and the method for shutting down the temporary load device is more reasonable and saves time, and is called at the same time.
  • the temporary load device When the temporary load device is activated, the temporary load device may be invoked by starting the application in the temporary load device, or after the temporary load device turns off the application to release a service load, the temporary load device
  • the application can be actively loaded for immediate use when the temporary load device is invoked.
  • the temporary load device may be invoked to perform a traffic load according to the above method, and the other backup load devices are started.
  • the method further includes:
  • the standby load device When the traffic load of the standby load device is zero, the standby load device is turned off.
  • the backup needs to be closed.
  • the load device when the primary load device is exceeded again, may invoke the temporary load device to perform a traffic load according to the foregoing method, and start the standby load device.
  • the system may be a cloud service system, where the cloud service system includes an online server, a capacity expansion server, and a buffer server.
  • the online server corresponds to the primary load device and the capacity expansion server in this embodiment.
  • the buffer server corresponds to the temporary load device in this embodiment, and may of course be other systems.
  • the temporary load device when the service load value of the primary load device exceeds the threshold, the temporary load device is invoked to perform a traffic load, and the standby load device is started, and when the standby load device is started, the traffic load of the temporary load device is switched to the On the standby load device.
  • the invention avoids the service abnormality, unavailability and even suspension caused by the service load continuously rising in the process of starting the standby load, and the service load of the main load device is exceeded, and the zero-waiting and zero-risk business load is realized.
  • the cloud server includes an online server, a capacity expansion server, and a buffer server, as shown in FIG. 2, where
  • the online server is a server that is currently carrying a service load
  • the buffer server is a server that can perform a temporary service load at any time
  • the expansion server is a server that can perform a business load after being expanded, wherein the buffer server is placed in a cache pool. If the current service load exceeds the load threshold of the online server in the cloud service, and the capacity expansion server needs to be expanded horizontally, the method includes:
  • Expanding the capacity expansion server If there are multiple expansion servers in the cloud server, and the service load capacity of the multiple expansion servers is different, the current service load growth rate and the expansion time of each expansion server are expanded.
  • the application of the buffer server is performed to perform the business load, and the buffer server is put back into the buffer pool, and the buffer server placed in the buffer pool can automatically start the application for the service load, but The traffic load is performed, and after the buffer server is invoked, the buffer server loads the application to load the traffic load.
  • the expansion server When the service load of the expansion server is zero, the expansion server is closed. When the service load capacity of the online server is exceeded, the buffer server may be invoked again to perform a service load according to the foregoing method, and the other expansion servers are started.
  • the present invention proposes a system, as shown in FIG. 3, the system includes a primary load device, a temporary load device, and a backup load device, wherein the primary load device is currently being A device that performs a service load, and the temporary load device is a device that can perform a temporary service load at any time, and the standby device is a device that can perform a service load after being started, and the system includes:
  • the determining module 31 is configured to determine whether the service load value of the primary load device exceeds a threshold
  • the calling temporary load device is used to perform a traffic load;
  • the startup module 33 is configured to start the standby load device;
  • the switching module 34 is configured to switch the service load of the temporary load device to the standby load device after the startup of the standby load device is completed, so that the standby load device performs a traffic load.
  • the load capacity of the backup load device is different, and the load capacity of the standby load device is different, and the startup time of the standby load device is higher.
  • the startup module is specifically configured to:
  • the standby load device corresponding to the load capacity is started according to the growth rate and the startup time of each of the standby load devices.
  • the standby load device is multiple, and each of the standby load devices has the same load capacity
  • the startup module is specifically configured to:
  • the switching module is specifically configured to:
  • the system further includes:
  • the shutdown module is configured to shut down the standby load device after the traffic load of the temporary load device is switched to the standby load device.
  • the temporary load device when the service load value of the primary load device exceeds the threshold, the temporary load device is invoked to perform a traffic load, and the standby load device is started, and when the standby load device is started, the traffic load of the temporary load device is switched to the On the standby load device.
  • the invention avoids the service abnormality, unavailability and even suspension caused by the service load continuously rising in the process of starting the standby load, and the service load of the main load device is exceeded, and the zero-waiting and zero-risk business load is realized.
  • the aforementioned program can be stored in a computer readable storage medium.
  • the program when executed, performs the steps including the foregoing method embodiments; and the foregoing storage medium includes various media that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.
  • the device embodiments described above are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, ie may be located in one place. Or it can be distributed to at least two network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. Those of ordinary skill in the art can understand and implement without deliberate labor.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)
  • Hardware Redundancy (AREA)

Abstract

本发明提供了一种负载方法和系统,其中,所述方法包括:判断所述主负载设备的业务负载值是否超过阈值;如果超过,调用所述临时负载设备进行业务负载,并启动所述备用负载设备;当对所述备用负载设备启动完毕后,将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。本发明避免了在备用负载启动的过程中由于业务负载持续上升,超过主负载设备的业务负载能力时而引起的服务异常、不可用,甚至挂起现象。

Description

一种负载方法和系统
本申请要求2016年03月22日递交的申请号为201610165378.6、发明名称为“一种负载方法和系统”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明涉及网络技术领域,尤其涉及一种负载方法和系统。
背景技术
云服务中线上服务器业务负载上升达到阈值的时候,就需要扩容出新的服务器来应对新增负载。
现有技术中的扩容方案是横向扩容,也就是启动更多的扩容服务器来支撑当前业务负载。
在实现本发明的过程中,发明人发现现有技术至少存在如下问题:
扩容新服务器的等待时间通常是几分钟或者更长,如果在此期间业务负载持续上升,一旦超出线上服务器的处理能力,就会造成服务异常、不可用,甚至是服务挂起的现象。
发明内容
本发明提供一种负载方法,当主负载设备的业务负载值超过阈值时,调用临时负载设备进行业务负载,并启动备用负载设备,当所述备用负载设备启动完毕后将所述临时负载设备的业务负载切换到所述备用负载设备上。本发明避免了在备用负载启动的过程中由于业务负载持续上升,超过主负载设备的业务负载能力时而引起的服务异常、不可用,甚至挂起现象。
本发明提供一种负载方法,所述方法应用于包括主负载设备、临时负载设备和备用负载设备的系统中,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述方法包括:
判断所述主负载设备的业务负载值是否超过阈值;
如果超过,调用所述临时负载设备进行业务负载,并启动所述备用负载设备;
当对所述备用负载设备启动完毕后,将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同,其中,业务负载能力越高的所述备用负载设备的启动时间越长;
启动所述备用负载设备,具体包括:
确定当前业务负载的增长速率;
根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同;
启动所述备用负载设备,具体包括:
确定当前业务负载的增长速率;
根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设备。
将所述临时负载设备的业务负载切换到所述备用负载设备上,具体包括:
关闭所述临时负载设备进行业务负载的应用;
将所述临时负载设备的业务负载切换到所述备用负载设备上。
在将所述临时负载设备的业务负载切换到所述备用负载设备上后,所述方法还包括:
当所述备用负载设备的业务负载为零时,关闭所述备用负载设备。
一种系统,所述系统中包括主负载设备、临时负载设备和备用负载设备,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述系统包括:
判断模块,用于判断所述主负载设备的业务负载值是否超过阈值;
调用模块,如果超过阈值,用于调用所述临时负载设备进行业务负载;启动模块,用于启动所述备用负载设备;
切换模块,当对所述备用负载设备启动完毕后,用于将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同,其中,业务负载能力越高的所述备用负载设备的启动时间越长;
所述启动模块,具体用于:
确定当前业务负载的增长速率;
根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同;
所述启动模块,具体用于:
确定当前业务负载的增长速率;
根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设备。
所述切换模块,具体用于:
关闭所述临时负载设备进行业务负载的应用;
将所述临时负载设备的业务负载切换到所述备用负载设备上。
所述系统,还包括:
关闭模块,在将所述临时负载设备的业务负载切换到所述备用负载设备上后,当所述备用负载设备的业务负载为零时,用于关闭所述备用负载设备。
本发明中当主负载设备的业务负载值超过阈值时,调用临时负载设备进行业务负载,并启动备用负载设备,当所述备用负载设备启动完毕后将所述临时负载设备的业务负载切换到所述备用负载设备上。本发明避免了在备用负载启动的过程中由于业务负载持续上升,超过主负载设备的业务负载能力时而引起的服务异常、不可用,甚至挂起现象。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为本发明实施例中的一种负载方法;
图2为本发明实施例中在线服务器、扩容服务器和缓冲服务器位置结构示意图;
图3为本发明实施例中的一种系统的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范 围。
为了解决现有技术中提出的问题,本发明提出了一种负载方法,具体方法如图1所示,所述方法应用于包括主负载设备、临时负载设备和备用负载设备的系统中,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述方法包括以下步骤:
步骤101,判断所述主负载设备的业务负载值是否超过阈值,如果超过,则执行步骤102,如果没有超过,则结束流程。
具体的,所述主负载设备和所述备用负载设备能够提供稳定的负载业务,并为用户提供稳定的、高品质的服务,所述主负载设备和所述备用负载设备的稳定性、业务负载能力、提供业务的能力,以及硬件配置和硬件性能要高于所述临时负载设备,所述主负载设备和所述备用负载设备是横向启动,也就是说,在主负载设备的负载能力超过阈值时,才会启动所述备用负载设备,进一步的,所述方法具体可用于横向启动的场景里。
所述临时负载设备为时刻开启状态,即处于随时可以进行业务负载的状态,在所述主负载设备超过阈值时立即使用所述临时负载设备进行业务负载,以使所述临时负载设备为所述主负载设备分担负载,并且所述临时负载设备处于开启状态时其所消耗的成本相对于所述主负载设备和所述备用负载设备处于开启状态时所消耗的成本要低很多,所述成本包括如:维护成本、消耗成本,以及资源占用成本等。
当所述主负载设备的业务负载值接近阈值时,表示所述主负载设备能够提供的业务负载能力已经很少,如果所述业务负载在持续增长,那么会超过所述主负载设备的业务负载能力,需要横向启动所述备用负载设备。
步骤102,调用所述临时负载设备进行业务负载,并启动所述备用负载设备。
具体的,由于启动所述备用负载设备需要一定时间,为了避免在所述备用负载设备的过程中,当前的业务负载超过所述主负载设备的业务负载能力,因此调用所述临时负载设备,由于所述临时负载设备时刻处于开启状态,因此所述临时负载设备可以随时在所述主负载设备超过其负载能力时为所述主负载设备分担业务负载,在所述备用负载设备启动的过程中起到一个过渡作用,在此期间,所述临时负载设备能够为稳定的负载业务,并为用户提供稳定的、高品质的服务。
其中,所述备用负载设备为多个,并且多个所述备用负载设备的负载能力可以相同,也可以不同,并且负载能力越高的所述备用负载设备启动时间也越长。
当所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同时,启动所述备用负载设备,具体包括:
确定当前业务负载的增长速率;
根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
具体的,为了保证启动的所述备用负载设备能够承担超出所述主负载设备的业务负载,并且保证合理的启动所述备用负载设备,需要根据所述增长速率和各个所述备用负载设备的启动时间来启动所述备用负载设备,这样可以节省系统的资源和成本,例如:所述备用负载设备有3个,业务负载能力分别为:10、15、20,启动时间分别为3分钟、4分钟、5分钟,如果当前业务负载的增长速率为1,那么需要启动业务负载能力为10的所述备用负载设备;如果当业务负载的增长速率为3,并且如果超过业务负载能力为10的所述备用负载设备的阈值,但是没有超过业务负载能力为15的所述备用负载设备的阈值,那么需要启动业务负载能力为15的所述备用负载设备,如果未超过业务负载能力为10的所述备用负载设备的阈值,那么需要启动业务负载能力为10的所述备用负载设备。
其中,在确定系统中使用何种级别的业务负载量的所述备用负载设备可以根据实际情况确定。
当所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同时,启动所述备用负载设备,具体包括:
确定当前业务负载的增长速率;
根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设备。
具体的,由于多个所述备用负载设备的启动时间相同,在启动时启动任何一个都可以,但是为了保证启动的所述备用负载设备能够负载超过所述主负载设备的业务负载,并且保证启动合理个数的所述备用负载设备,需要根据所述增长速率和各个所述备用负载设备的启动时间来启动所述备用负载设备的个数,例如:所述备用负载设备有3个,业务负载能力都为10,启动时间都为3分钟,如果当前业务负载的增长速率为1,那么需要启动的个数为1个;如果当前业务负载的增长速率为3,并且如果超过了一个所述备用负载设备的阈值,那么需要启动的个数为2个,如果未超过一个所述备用负载设备的阈值,那么需要启动的个数为1个;如果当前业务负载的增长速率为4,那么需要启 动的个数为2个。
其中,在确定系统中使用何种级别的业务负载量的所述备用负载设备可以根据实际情况确定。
在所述备用负载设备为多个的情况下,当所述主负载设备的业务负载值超过阈值后,启动所述备用负载设备时,也可以任意启动多个所述备用负载设备中的一个,如果启动的所述备用负载设备的负载能力不能满足超出所述主负载设备负载能力的业务负载,那么任意启动另一个所述备用负载设备,直到启动的所述备用负载设备的负载能力能够满足超出所述主负载设备负载能力的业务负载。当然所述备用负载设备也可以为一个,所述备用负载设备的负载能力是根据日常的业务负载量确定的,即所述备用负载设备的业务负载能力足够强大,在所述主负载设备的业务负载值超过阈值后,由所述临时负载设备暂时进行业务负载,当所述备用负载设备启动完成后,将所述临时负载设备上负载的业务切换到所述备用负载设备上进行业务负载。
步骤103,当对所述备用负载设备启动完毕后,将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
具体的,为了保证为能够保证更好的进行业务负载,以及为用户提供更好的服务,当对所述备用负载设备启动完毕后,需要将所述临时负载设备的业务负载切换到所述备用负载设备上,进而让所述备用负载设备进行业务负载。
将所述临时负载设备的业务负载切换到所述备用负载设备上,具体包括:
关闭所述临时负载设备进行业务负载的应用;
将所述临时负载设备的业务负载切换到所述备用负载设备上。
具体的,所述临时负载设备是通过其运行的应用来进行业务负载的,当对所述备用负载设备启动完毕后只需要将所述应用关闭,就会释放业务负载,系统会将释放的业务负载切换到启动了的所述备用负载设备上,这样不用关闭所述临时负载设备就能达到释放业务负载的效果,并且相对于关闭所述临时负载设备的方法更加合理和节省时间,同时在调用所述临时负载设备时,通过启动所述临时负载设备中的所述应用就可以调用所述临时负载设备了,或者在所述临时负载设备关闭所述应用释放业务负载后,所述临时负载设备可以主动加载所述应用,以在调用所述临时负载设备时可以立即使用。当超出所述备用负载设备的负载能力时,可以按照上述方法再调用所述临时负载设备进行业务负载,并启动其他所述备用负载设备。
在将所述临时负载设备的业务负载切换到所述备用负载设备上后,所述方法还包括:
当所述备用负载设备的业务负载为零时,关闭所述备用负载设备。
具体的,为了节约成本,以及避免不必要的资源浪费,当所述备用负载设备的业务负载为零时,即所述备用负载设备不在进行业务负载和为用户提供服务时,需要关闭所述备用负载设备,当再次超出所述主负载设备时,可以按照上述方法再调用所述临时负载设备进行业务负载,并启动所述备用负载设备。
所述系统可以为云服务系统,所述云服务系统中包括在线服务器、扩容服务器和缓冲服务器,其中,在线服务器对应本实施例中的主负载设备、扩容服务器对应本实施例中的备用负载设备、缓冲服务器对应本实施例中的临时负载设备,当然也可以是其他系统,凡是在主负载设备超过阈值需要横向启动备用服务器,且启动备用服务器的过程中会造成服务异常、不可用,甚至是服务挂起的现象的系统均属于本发明的保护范围。
本发明中当主负载设备的业务负载值超过阈值时,调用临时负载设备进行业务负载,并启动备用负载设备,当所述备用负载设备启动完毕后将所述临时负载设备的业务负载切换到所述备用负载设备上。本发明避免了在备用负载启动的过程中由于业务负载持续上升,超过主负载设备的业务负载能力时而引起的服务异常、不可用,甚至挂起现象,实现了零等待、零风险的业务负载。
为了进一步阐述本发明的技术思想,现结合具体的应用场景,对本申请的技术方案进行说明,具体的,云服务器包括在线服务器、扩容服务器和缓冲服务器,具体如图2所示,其中,所述在线服务器为当前正在进行业务负载的服务器,所述缓冲服务器为随时可以进行临时业务负载的服务器,所述扩容服务器是进行扩容后才能进行业务负载的服务器,其中,所述缓冲服务器放于缓存池中,如果当前业务负载超过云服务中在线服务器的负载阈值,需要横向对扩容服务器进行扩容时,所述方法包括:
1、从缓存池中调用所述缓冲服务器,以快速对超出所述在线服务器负载能力的业务负载进行负载,这样可以在超出所述在线服务器的负载能力时所述云服务仍可以为用户提供服务,实现零等待、零风险的目的。
2、对扩容服务器进行扩容,如果云服务器中有多个所述扩容服务器,且多个所述扩容服务器的业务负载能力不同,根据当前的业务负载增长速率和各个所述扩容服务器的扩容时间扩容对应负载能力的所述扩容服务器;如果云服务器中有多个所述扩容服务器,且多个所述扩容服务器的业务负载能力相同,根据当前的业务负载增长速率和所述扩容服务器的扩容时间扩容对应个数的所述扩容服务器,具体方法已在上述进行了详细说明,在此不再一一赘述。
3、当对所述扩容服务器扩容完毕后,立即让所述扩容服务器上线。
4、关闭所述缓冲服务器进行业务负载的应用,并将所述缓冲服务器放回所述缓存池中,放入所述缓存池中的所述缓冲服务器可以自动启动进行业务负载的应用,但是不进行业务负载,也可以在调用所述缓冲服务器后,所述缓冲服务器在加载所述应用,以对业务负载进行负载。
5、将所述缓冲服务器负载的业务负载切换到所述扩容服务器中进行业务负载。当超出所述扩容服务器的业务负载能力时,可以按照上述方法再次调用所述缓冲服务器,并启动其他所述扩容服务器。
6、当所述扩容服务器的业务负载为零时,关闭所述扩容服务器。当超出所述在线服务器的业务负载能力时,可以按照上述方法再次调用所述缓冲服务器进行业务负载,并启动其他所述扩容服务器。
基于与上述方法同样的申请构思,本发明提出了一种系统,如图3所示,所述系统中包括主负载设备、临时负载设备和备用负载设备,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述系统包括:
判断模块31,用于判断所述主负载设备的业务负载值是否超过阈值;
调用模块32,如果超过阈值,用于调用所述临时负载设备进行业务负载;启动模块33,用于启动所述备用负载设备;
切换模块34,当对所述备用负载设备启动完毕后,用于将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同,其中,业务负载能力越高的所述备用负载设备的启动时间越长;
所述启动模块,具体用于:
确定当前业务负载的增长速率;
根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同;
所述启动模块,具体用于:
确定当前业务负载的增长速率;
根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设 备。
所述切换模块,具体用于:
关闭所述临时负载设备进行业务负载的应用;
将所述临时负载设备的业务负载切换到所述备用负载设备上。
所述系统,还包括:
关闭模块,在将所述临时负载设备的业务负载切换到所述备用负载设备上后,当所述备用负载设备的业务负载为零时,用于关闭所述备用负载设备。
本发明中当主负载设备的业务负载值超过阈值时,调用临时负载设备进行业务负载,并启动备用负载设备,当所述备用负载设备启动完毕后将所述临时负载设备的业务负载切换到所述备用负载设备上。本发明避免了在备用负载启动的过程中由于业务负载持续上升,超过主负载设备的业务负载能力时而引起的服务异常、不可用,甚至挂起现象,实现了零等待、零风险的业务负载。
本领域普通技术人员可以理解:实现上述各方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取存储介质中。该程序在执行时,执行包括上述各方法实施例的步骤;而前述的存储介质包括:ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所描述的装置实施例仅仅是示意性的,其中作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到至少两个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下,即可以理解并实施。
最后应说明的是:以上各实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述各实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分或者全部技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims (10)

  1. 一种负载方法,其特征在于,所述方法应用于包括主负载设备、临时负载设备和备用负载设备的系统中,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述方法包括:
    判断所述主负载设备的业务负载值是否超过阈值;
    如果超过,调用所述临时负载设备进行业务负载,并启动所述备用负载设备;
    当对所述备用负载设备启动完毕后,将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
  2. 如权利要求1所述方法,其特征在于,所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同,其中,业务负载能力越高的所述备用负载设备的启动时间越长;
    启动所述备用负载设备,具体包括:
    确定当前业务负载的增长速率;
    根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
  3. 如权利要求1所述方法,其特征在于,所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同;
    启动所述备用负载设备,具体包括:
    确定当前业务负载的增长速率;
    根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设备。
  4. 如权利要求1所述方法,其特征在于,将所述临时负载设备的业务负载切换到所述备用负载设备上,具体包括:
    关闭所述临时负载设备进行业务负载的应用;
    将所述临时负载设备的业务负载切换到所述备用负载设备上。
  5. 如权利要求1所述方法,其特征在于,在将所述临时负载设备的业务负载切换到所述备用负载设备上后,所述方法还包括:
    当所述备用负载设备的业务负载为零时,关闭所述备用负载设备。
  6. 一种系统,其特征在于,所述系统中包括主负载设备、临时负载设备和备用负 载设备,其中,所述主负载设备为当前正在进行业务负载的设备,所述临时负载设备为随时可以进行临时业务负载的设备,所述备用设备是进行启动后才能进行业务负载的设备,所述系统包括:
    判断模块,用于判断所述主负载设备的业务负载值是否超过阈值;
    调用模块,如果超过阈值,用于调用所述临时负载设备进行业务负载;启动模块,用于启动所述备用负载设备;
    切换模块,当对所述备用负载设备启动完毕后,用于将所述临时负载设备的业务负载切换到所述备用负载设备上,以使所述备用负载设备进行业务负载。
  7. 如权利要求6所述系统,其特征在于,所述备用负载设备为多个,且不同的所述备用负载设备的负载能力不同,其中,业务负载能力越高的所述备用负载设备的启动时间越长;
    所述启动模块,具体用于:
    确定当前业务负载的增长速率;
    根据所述增长速率和各个所述备用负载设备的启动时间启动对应负载能力的所述备用负载设备。
  8. 如权利要求6所述系统,其特征在于,所述备用负载设备为多个,且各个所述备用负载设备的负载能力相同;
    所述启动模块,具体用于:
    确定当前业务负载的增长速率;
    根据所述增长速率和所述备用负载设备的启动时间启动对应数量的所述备用负载设备。
  9. 如权利要求6所述系统,其特征在于,所述切换模块,具体用于:
    关闭所述临时负载设备进行业务负载的应用;
    将所述临时负载设备的业务负载切换到所述备用负载设备上。
  10. 如权利要求6所述系统,其特征在于,所述系统,还包括:
    关闭模块,在将所述临时负载设备的业务负载切换到所述备用负载设备上后,当所述备用负载设备的业务负载为零时,用于关闭所述备用负载设备。
PCT/CN2017/076039 2016-03-22 2017-03-09 一种负载方法和系统 WO2017162034A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610165378.6A CN107222514B (zh) 2016-03-22 2016-03-22 一种负载方法和系统
CN201610165378.6 2016-03-22

Publications (1)

Publication Number Publication Date
WO2017162034A1 true WO2017162034A1 (zh) 2017-09-28

Family

ID=59899187

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/076039 WO2017162034A1 (zh) 2016-03-22 2017-03-09 一种负载方法和系统

Country Status (3)

Country Link
CN (1) CN107222514B (zh)
TW (1) TWI746513B (zh)
WO (1) WO2017162034A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114302350A (zh) * 2021-12-30 2022-04-08 胜斗士(上海)科技技术发展有限公司 业务提供方故障切换方法、装置、电子设备和存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108540567B (zh) * 2018-04-19 2019-10-11 北京百度网讯科技有限公司 用于推送信息的方法和装置
CN113254224A (zh) * 2021-07-15 2021-08-13 中电金信软件有限公司 计算资源扩容方法、装置、电子设备和可读存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030079154A1 (en) * 2001-10-23 2003-04-24 Kie Jin Park Mothed and apparatus for improving software availability of cluster computer system
CN102868744A (zh) * 2012-09-10 2013-01-09 北京用友政务软件有限公司 一种实现SaaS与IaaS自动化集成管理的方法
CN103856512A (zh) * 2012-11-30 2014-06-11 华为技术有限公司 云计算的管理服务器、工作和闲置主机以及资源调度方法
CN105024842A (zh) * 2014-04-25 2015-11-04 深圳市腾讯计算机系统有限公司 服务器的扩容方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6578068B1 (en) * 1999-08-31 2003-06-10 Accenture Llp Load balancer in environment services patterns
CN101834831A (zh) * 2009-03-13 2010-09-15 华为技术有限公司 一种实现nat设备冗余备份的方法、装置和系统
US8190564B2 (en) * 2010-02-25 2012-05-29 Ebay, Inc. Temporary session data storage
CN102571527A (zh) * 2010-12-09 2012-07-11 上海杉达学院 基于虚拟路由冗余的备份及负载均衡系统
WO2011144116A2 (zh) * 2011-05-27 2011-11-24 华为技术有限公司 控制备用设备的方法、主用设备和备用设备
US10104166B2 (en) * 2014-05-20 2018-10-16 Citrix Systems, Inc. Systems and methods for providing load balancing as a service
US10511694B2 (en) * 2014-07-23 2019-12-17 Citrix Systems, Inc. Systems and methods for application specific load balancing

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030079154A1 (en) * 2001-10-23 2003-04-24 Kie Jin Park Mothed and apparatus for improving software availability of cluster computer system
CN102868744A (zh) * 2012-09-10 2013-01-09 北京用友政务软件有限公司 一种实现SaaS与IaaS自动化集成管理的方法
CN103856512A (zh) * 2012-11-30 2014-06-11 华为技术有限公司 云计算的管理服务器、工作和闲置主机以及资源调度方法
CN105024842A (zh) * 2014-04-25 2015-11-04 深圳市腾讯计算机系统有限公司 服务器的扩容方法及装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114302350A (zh) * 2021-12-30 2022-04-08 胜斗士(上海)科技技术发展有限公司 业务提供方故障切换方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
TWI746513B (zh) 2021-11-21
CN107222514B (zh) 2020-01-14
TW201734829A (zh) 2017-10-01
CN107222514A (zh) 2017-09-29

Similar Documents

Publication Publication Date Title
WO2017162034A1 (zh) 一种负载方法和系统
KR101739216B1 (ko) 중앙집중식 태스크 스케줄링
US9513955B2 (en) Application management method and terminal
CN106899649B (zh) 一种任务请求处理方法、装置和用户设备
CN105792335B (zh) 一种提高续航时间的方法和终端设备
TWI629850B (zh) 備用電源供應器支援之系統和方法及相關的非暫時性機器可讀取媒體
CN107844187A (zh) 功耗管理方法、装置及电子设备
US10146294B2 (en) Method and network units for handling states
CN107577563A (zh) 一种系统升级异常断电的保护方法及终端设备
WO2017031866A1 (zh) 一种适于无线扩容的业务处理方法及装置
CN106293885B (zh) 任务创建、挂起和恢复方法
US9665163B2 (en) Distributed power management with partial suspend mode for distributed storage systems
WO2018228323A1 (zh) 在线服务系统的服务级别控制方法与系统、可读存储介质
CN107872480B (zh) 大数据集群数据平衡方法和装置
CN105306514A (zh) 基于Zookeeper的录音文件分布存储方法和系统
US20150169367A1 (en) System and method for supporting adaptive busy wait in a computing environment
JP2009223368A (ja) クラスタリング制御装置、制御システム、制御方法及び制御プログラム
WO2019076082A1 (zh) 一种应用自启动的控制方法及装置
CN102129284B (zh) 降低业务系统功耗的方法、装置及系统
WO2023093323A1 (zh) 一种版本升级方法及装置
JP2009223687A (ja) 情報処理システムおよびその制御方法
JP6070282B2 (ja) 仮想マシン管理装置、方法及びプログラム
JP2009086758A (ja) コンピュータ・システム及びシステム管理プログラム
US20150253843A1 (en) Dynamic energy savings for digital signal processor modules using plural energy savings states
CN104375889A (zh) 一种Web层切换系统及方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17769311

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17769311

Country of ref document: EP

Kind code of ref document: A1