WO2019174509A1 - 速率控制方法、装置及电子设备 - Google Patents

速率控制方法、装置及电子设备 Download PDF

Info

Publication number
WO2019174509A1
WO2019174509A1 PCT/CN2019/077267 CN2019077267W WO2019174509A1 WO 2019174509 A1 WO2019174509 A1 WO 2019174509A1 CN 2019077267 W CN2019077267 W CN 2019077267W WO 2019174509 A1 WO2019174509 A1 WO 2019174509A1
Authority
WO
WIPO (PCT)
Prior art keywords
sub
periods
period
values
value
Prior art date
Application number
PCT/CN2019/077267
Other languages
English (en)
French (fr)
Inventor
韩寅
张军
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 WO2019174509A1 publication Critical patent/WO2019174509A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/0816—Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5003—Managing SLA; Interaction between SLA and QoS
    • H04L41/5019—Ensuring fulfilment of SLA
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/25—Flow control; Congestion control with rate being modified by the source upon detecting a change of network conditions

Definitions

  • the present application relates to the field of communications technologies, and in particular, to a rate control method, apparatus, and electronic device.
  • the actual request may be processed centrally within the first 100 or 200 ms within 1 s.
  • the inventor has found that at least the following problems exist in the prior art: the traditional system performance evaluation scheme causes the centralized pulse processing of the request, so that the actual traffic distribution is uneven, resulting in the system at a certain moment. The transient pressure is too large.
  • the embodiment of the present application provides a rate control method, device, and electronic device to solve the defect of uneven distribution of traffic in the prior art and reduce the instantaneous pressure of the system.
  • the embodiment of the present application provides a rate control method, including:
  • the request is processed in each of the sub-periods in turn such that the number of requests that need to be processed is equal to the sub-value corresponding to the sub-period.
  • the embodiment of the present application further provides a rate control apparatus, including:
  • a first dividing module configured to divide the target QPS value into a plurality of sub-values, and the sum of the plurality of sub-values is the target QPS value
  • a second dividing module configured to divide a second into a plurality of sub-periods, the sum of the plurality of sub-periods being one second, the number of the sub-values being equal to the number of the sub-periods, each of the a sub-value corresponding to each of the sub-time segments;
  • an execution module configured to process the request sequentially in each of the sub-periods, such that the number of requests that need to be processed is equal to a sub-value corresponding to the sub-period.
  • the embodiment of the present application further provides an electronic device, including:
  • a processor for running the program stored in the memory for:
  • the request is processed in each of the sub-periods in turn such that the number of requests that need to be processed is equal to the sub-value corresponding to the sub-period.
  • the rate control method, device and electronic device provided by the embodiments of the present application allocate a target QPS value and a time period of one second, so that a large QPS value is distributed to each sub-time period for processing, and the traffic distribution is not solved.
  • the problem of the average greatly reduces the instantaneous pressure of the system.
  • FIG. 1 is a schematic diagram of the principle of a rate control method provided by the present application.
  • FIG. 2 is a system block diagram of a service system according to an embodiment of the present application.
  • FIG. 3 is a flowchart of an embodiment of a rate control method provided by the present application.
  • FIG. 5 is a schematic structural diagram of an embodiment of a rate control apparatus provided by the present application.
  • FIG. 6 is a schematic structural diagram of an embodiment of an electronic device provided by the present application.
  • the conventional system performance evaluation scheme achieves performance compliance when the performance is evaluated in units of seconds, that is, the number of requests processed per second can reach a preset target QPS value.
  • the actual traffic distribution in one second is uneven, resulting in excessive transient pressure of the system at a certain moment; at the same time, it will cause instantaneous resource preemption to the host machine, which will inevitably affect the execution of other tasks.
  • the present application provides a solution whose main principle is: assigning a target QPS value and a time period of one second, so that a large QPS value is dispersed to each sub-period to complete Solve the problem of uneven flow distribution, thus reducing the instantaneous pressure of the system.
  • FIG. 1 is a schematic diagram of the principle of the rate control method provided by the present application.
  • the target QPS value is 5000, that is, a total of 5000 executions per second, but since the tasks are all performed as much as possible, the task is concentrated in the first 200 ms (milliseconds), for example, the first 100 ms. 4800 times were completed, 200 times were completed in the second 100ms, and the flow distribution was extremely uneven.
  • the solution proposed by the present application is to split and match the target QPS value and the one-second time period, so that a large QPS value is dispersed to each sub-period to complete. As shown in FIG.
  • FIG. 2 is a system block diagram of a service system according to an embodiment of the present application.
  • the structure shown in FIG. 2 is only one example of a service system to which the technical solution of the present application can be applied.
  • a rate control device is included in the service system.
  • the rate control device is mainly used to perform the processing flow shown in FIG. 3 and FIG.
  • the sub-period the sum of the sub-periods is one second, the number of sub-values is equal to the number of sub-periods, and each sub-value corresponds to each sub-period sequentially; in each sub-period, the request is processed in turn, The number of requests that need to be processed is equal to the sub-value corresponding to the sub-period. Dispersing a large QPS value into each sub-period is completed, which can solve the problem of uneven distribution of flow and reduce the instantaneous pressure of the system.
  • FIG. 3 is a flowchart of an embodiment of a rate control method provided by the present application.
  • the execution body of the method may be the service system described in the foregoing embodiment.
  • the rate control method includes the following steps:
  • the target QPS value is divided into a plurality of sub-values, and the sum of the plurality of sub-values is a target QPS value.
  • the target QPS value refers to that the system requires that the number of requests must be reached every second.
  • One request in the embodiment of the present application is used to indicate a task performed by the system, and may also be used to represent a transaction (Transaction). ), that is, a program execution unit that accesses and possibly updates various data items in the database.
  • Transaction Transaction
  • the QPS value needs to be divided into several sub-values, and the sum of all the sub-values is the target QPS value.
  • the specific number of sub-values may be determined according to empirical values, or may be adaptively set and adjusted according to the performance of the business system.
  • the division manners that can be adopted are: target QPS value equalization, time division (ie, target QPS value and time are equally distributed, for example, 5,000 times equal to 10 500 times) , divide 1s into 10 100ms); target QPS values are not equally divided, time equal parts (for example, 5000 times are divided into 1000, 900, 800, 1100, and 1200 times, and 1s is divided equally.
  • target QPS value is equally divided, time is not equal (for example, 5,000 times is divided into 5 1000 times, 1s is divided into 200ms, 190ms, 150ms, 2100ms and 250ms); target QPS values are not equal
  • the time is not equally divided (for example, 5000 times are divided into 1000, 900, 800, 1100, and 1200 times, and 1s is sequentially divided into 200ms, 190ms, 150ms, 2100ms, and 250ms).
  • S303 Process the request in sequence in each sub-period, so that the number of requests that need to be processed is equal to the sub-value corresponding to the sub-period.
  • the requests for each sub-period are processed in chronological order, and in each sub-period, the number of requests that need to be processed is the same as the sub-value corresponding to the time period.
  • the division method for a certain business system is: divide 5000 times into 1000, 900, 800, 1100, and 1200, and divide 1s into 5 200ms. Then, in each 200 ms period, the number of requests processed in sequence is 1000, 900, 800, 1,100, and 1200, respectively.
  • the rate control method provided by the embodiment of the present application solves the problem of uneven distribution of traffic by dividing and matching the target QPS value and the one-second time period, so that a large QPS value is distributed to each sub-period. , greatly reducing the instantaneous pressure of the system.
  • the rate control method provided in this embodiment may further include the following steps:
  • the target QPS value is equally divided into a plurality of sub-values, and the sum of the plurality of sub-values is a target QPS value.
  • S402 divide one second into several sub-time segments, and the sum of the plurality of sub-time segments is one second, the number of sub-values is equal to the number of sub-time segments, and each sub-value corresponds to each sub-time segment.
  • a preferred division mode is adopted, that is, the target QPS value and the time are divided into equal parts, so that the traffic distribution is more uniform.
  • S403 Processing the request in each sub-period, in turn, when the number of processed requests is equal to the sub-value corresponding to the current sub-period, starting the blocking mechanism, so that the unprocessed request enters the waiting state until the next sub-period At the beginning, the blocking mechanism is turned off.
  • Blocking means suspending a process (unprocessed request) without obtaining a resource until a resource is obtained.
  • the suspended process goes to sleep and is removed by the scheduler's run queue until the wait condition (beginning of the next sub-period) is satisfied.
  • the remaining value and remaining time of the target QPS value are reallocated to keep the target QPS unchanged.
  • the division method is adopted: divide 5000 times into 10 500 times, and divide 1s into 10 100 ms.
  • the number of processed requests is 400 ( ⁇ 500)
  • the number of sub-values is an integer multiple (at least twice) of the number of sub-periods, each sub-period corresponding to multiple sub-values;
  • the requests are processed in parallel such that the total number of requests that need to be processed is equal to the sum of the sub-values corresponding to the sub-period.
  • the target QPS value and time may be reallocated. For example, suppose the original division method is: divide 5,000 times into 10 500 times, and divide 1s into 10 100ms. Then, it can be re-divided into: divide 5,000 times into 10 500 times, and divide 1s into 5 200ms.
  • the rate control method provided by the embodiment of the present application further divides and matches the target QPS value and the one-second time period, so that a large QPS value is evenly distributed to each sub-period to be processed, thereby further improving the traffic distribution. Uniformity reduces the instantaneous pressure of the system; meanwhile, when the task of a sub-period is not completed, the remaining value of the target QPS value can be re-allocated in the remaining time of the period, or the request can be processed in parallel, thereby ensuring The target QPS value is achieved within the period.
  • FIG. 5 is a schematic structural diagram of an embodiment of a rate control apparatus provided by the present application, which can be used to perform the method steps shown in FIG. 3 and FIG. 4.
  • the apparatus may include: a first dividing module 51, a second dividing module 52, and an executing module 53.
  • the first dividing module 51 is configured to divide the target QPS value into a plurality of sub-values, and the sum of the plurality of sub-values is a target QPS value;
  • the second dividing module 52 is configured to divide the one-second into a plurality of sub-periods, and sum the sub-periods For one second, the number of sub-values is equal to the number of sub-periods, and each sub-value corresponds to each sub-period sequentially;
  • the execution module 53 is configured to sequentially process the request in each sub-period so that the request needs to be processed. The number is equal to the sub-value corresponding to the sub-period.
  • the target QPS value refers to that the system requires that the number of requests must be reached every second.
  • One request in the embodiment of the present application is used to indicate a task performed by the system, and may also be used to represent a transaction (Transaction). ), that is, a program execution unit that accesses and possibly updates various data items in the database.
  • Transaction Transaction
  • the first dividing module 51 divides the QPS value into a plurality of sub-values, and the sum of all the sub-values is the target QPS value.
  • the specific number of sub-values may be determined according to empirical values, or may be adaptively set and adjusted according to the performance of the business system.
  • the second dividing module 52 divides the time of one second into a plurality of sub-time periods, and the sum of all the sub-time periods is one second.
  • the sub-period is the same as the number of sub-values described above, and each sub-value corresponds to the sub-period sequentially.
  • the execution order of the first dividing module 51 and the second dividing module 52 is performed in sequence, or may be performed simultaneously.
  • the first dividing module 51 and the second dividing module 52 divide and match the target QPS value and time, and may adopt an equal or unequal manner.
  • the division manners that can be adopted are: target QPS value equalization, time division (ie, target QPS value and time are equally distributed, for example, 5,000 times equal to 10 500 times) , divide 1s into 10 100ms); target QPS values are not equally divided, time equal parts (for example, 5000 times are divided into 1000, 900, 800, 1100, and 1200 times, and 1s is divided equally.
  • target QPS value is equally divided, time is not equal (for example, 5,000 times is divided into 5 1000 times, 1s is divided into 200ms, 190ms, 150ms, 2100ms and 250ms); target QPS values are not equal
  • the time is not equally divided (for example, 5000 times are divided into 1000, 900, 800, 1100, and 1200 times, and 1s is sequentially divided into 200ms, 190ms, 150ms, 2100ms, and 250ms).
  • the execution module 53 sequentially processes the requests of each sub-period in chronological order, and in each sub-period, the number of requests to be processed is the same as the sub-value corresponding to the time period.
  • the division method for a certain business system is: divide 5000 times into 1000, 900, 800, 1100, and 1200, and divide 1s into 5 200ms. Then, in each 200 ms period, the number of requests processed in sequence is 1000, 900, 800, 1,100, and 1200, respectively.
  • the executing module 53 may be further configured to process the request in each sub-period in sequence, and when the number of processed requests is equal to the sub-value corresponding to the current sub-period, the blocking mechanism is started to wait for the unprocessed request to enter. The state, until the beginning of the next sub-period, closes the blocking mechanism.
  • Blocking means suspending a process (unprocessed request) without obtaining a resource until a resource is obtained. The suspended process goes to sleep and is removed by the scheduler's run queue until the wait condition (beginning of the next sub-period) is satisfied.
  • first dividing module 51 can also be used to divide the target QPS value into several sub-values.
  • the second dividing module 52 can also be used to divide one second into several sub-time segments.
  • a preferred division manner may be adopted, that is, the target QPS value and the time are divided into equal parts, so that the traffic distribution is more uniform.
  • the rate control apparatus may further include: a first redistribution module 54.
  • the first redistribution module 54 can be configured to: when a certain sub-period ends, if the number of processed requests in the sub-period is less than the sub-value corresponding to the sub-period, the remaining value of the target QPS value and one second Re-allocate the remaining time.
  • the first redistribution module 54 when a sudden situation occurs, so that the task is not completed within a certain sub-period, the first redistribution module 54 reallocates the remaining value and the remaining time of the target QPS value to keep the target QPS unchanged.
  • the division method is adopted: divide 5000 times into 10 500 times, and divide 1s into 10 100 ms.
  • the number of processed requests is 400 ( ⁇ 500)
  • the rate control apparatus may further include: a second redistribution module 55.
  • the second redistribution module 55 can be configured to re-allocate the target QPS value and one second if the number of processed requests in the sub-period is less than the sub-value corresponding to the sub-period when the sub-period ends. So that the number of sub-values is an integer multiple of the number of sub-periods, each sub-period corresponds to a plurality of sub-values; the execution module 53 sequentially processes the requests in parallel in each sub-period, so that the request to be processed is The total is equal to the sum of the sub-values corresponding to the sub-period.
  • the rate control apparatus provided by the embodiment of the present application further divides and matches the target QPS value and the one-second time period, so that a large QPS value is evenly distributed to each sub-period to be processed, thereby further improving the traffic distribution. Uniformity reduces the instantaneous pressure of the system; meanwhile, when the task of a sub-period is not completed, the remaining value of the target QPS value can be re-allocated in the remaining time of the period, or the request can be processed in parallel, thereby ensuring The target QPS value is achieved within the period.
  • FIG. 6 is a schematic structural diagram of an embodiment of an electronic device provided by the present application. As shown in FIG. 6, the electronic device includes a memory 61 and a processor 62.
  • the memory 61 is for storing a program. In addition to the above described procedures, the memory 61 can also be configured to store various other data to support operation on the electronic device. Examples of such data include instructions for any application or method operating on an electronic device, contact data, phone book data, messages, pictures, videos, and the like.
  • the memory 61 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Disk or Optical Disk.
  • SRAM static random access memory
  • EEPROM electrically erasable programmable read only memory
  • EPROM erasable Programmable Read Only Memory
  • PROM Programmable Read Only Memory
  • ROM Read Only Memory
  • Magnetic Memory Flash Memory
  • Disk Disk or Optical Disk.
  • the processor 62 coupled to the memory 61, executes the program stored by the memory 61 for:
  • the target QPS value is divided into several sub-values, and the sum of the number of sub-values is the target QPS value; one second is divided into several sub-periods, the sum of the sub-periods is one second, the number of sub-values and the number of sub-time periods Equally, each sub-value corresponds to each sub-period sequentially; in each sub-period, the request is processed in sequence, so that the number of requests to be processed is equal to the sub-value corresponding to the sub-period.
  • the electronic device may further include: a communication component 63, a power component 64, an audio component 65, a display 66, and the like. Only some of the components are schematically illustrated in FIG. 6, and it is not meant that the electronic device includes only the components shown in FIG. 6.
  • Communication component 63 is configured to facilitate wired or wireless communication between the electronic device and other devices.
  • the electronic device can access a wireless network based on a communication standard such as WiFi, 2G or 3G, or a combination thereof.
  • the communication component 63 receives broadcast signals or broadcast associated information from an external broadcast management system via a broadcast channel.
  • the communication component 63 also includes a near field communication (NFC) module to facilitate short range communication.
  • NFC near field communication
  • the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
  • RFID radio frequency identification
  • IrDA infrared data association
  • UWB ultra-wideband
  • Bluetooth Bluetooth
  • a power supply component 64 provides power to various components of the electronic device.
  • Power component 64 can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the electronic device.
  • the audio component 65 is configured to output and/or input an audio signal.
  • the audio component 65 includes a microphone (MIC) that is configured to receive an external audio signal when the electronic device is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode.
  • the received audio signal may be further stored in the memory 61 or transmitted via the communication component 63.
  • the audio component 65 also includes a speaker for outputting an audio signal.
  • Display 66 includes a screen whose screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user.
  • the touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor may sense not only the boundary of the touch or sliding action, but also the duration and pressure associated with the touch or slide operation.
  • 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.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例提供了一种速率控制方法、装置及电子设备。该方法包括：将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。本申请实施例通过对目标QPS值和一秒的时间段进行分配，使得一个较大的QPS值分散到各个子时间段来处理，解决了流量分布不均的问题，极大地降低了***的瞬时压力。

Description

速率控制方法、装置及电子设备
本申请要求2018年03月15日递交的申请号为201810215023.2、发明名称为“速率控制方法、装置及电子设备”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
本申请涉及通信技术领域，尤其涉及一种速率控制方法、装置及电子设备。
背景技术
在计算机软件领域，涉及到流量、消息等速率控制时，***往往采用QPS(每秒请求数)或TPS(每秒事务数)进行性能评估，即，以秒(s)作为基本的单位，规定每秒内必须执行的请求/事务数。
传统的性能评估方案中，由于请求都是尽力处理的，导致在给定的目标QPS的情况下，实际请求可能在1s内的前100或200ms被集中处理。
发明人在实现本申请的过程中，发现现有技术至少存在如下问题：传统的***性能评价方案，会造成请求的集中式脉冲式处理，使得实际流量分布不均匀，导致***在某一时刻的瞬时压力过大。
发明内容
本申请实施例提供一种速率控制方法、装置及电子设备，以解决现有技术中流量分布不均的缺陷，降低***的瞬时压力。
为达到上述目的，本申请实施例提供了一种速率控制方法，包括：
将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
本申请实施例还提供了一种速率控制装置，包括：
第一划分模块，用于将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
第二划分模块，用于将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
执行模块，用于依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
本申请实施例还提供一种电子设备，包括：
存储器，用于存储程序；
处理器，用于运行所述存储器中存储的所述程序，以用于：
将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
本申请实施例提供的速率控制方法、装置及电子设备，通过对目标QPS值和一秒的时间段进行分配，使得一个较大的QPS值分散到各个子时间段来处理，解决了流量分布不均的问题，极大地降低了***的瞬时压力。
上述说明仅是本申请技术方案的概述，为了能够更清楚了解本申请的技术手段，而可依照说明书的内容予以实施，并且为了让本申请的上述和其它目的、特征和优点能够更明显易懂，以下特举本申请的具体实施方式。
附图说明
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本申请的限制。而且在整个附图中，用相同的参考符号表示相同的部件。在附图中：
图1为本申请提供的速率控制方法的原理示意图；
图2为本申请实施例提供的业务***的***框图；
图3为本申请提供的速率控制方法一个实施例的流程图；
图4为本申请提供的速率控制方法另一个实施例的流程图；
图5为本申请提供的速率控制装置实施例的结构示意图；
图6为本申请提供的电子设备实施例的结构示意图。
具体实施方式
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例，然而应当理解，可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反，提供这些实施例是为了能够更透彻地理解本公开，并且能够将本公开的范围完整的传达给本领域的技术人员。
在现有技术中，传统的***性能评价方案，在以秒为单位评估性能时，性能达标，即，每秒处理的请求数能够达到预设的目标QPS值。但是，在一秒内的实际流量分布不均匀，导致***在某一时刻的瞬时压力过大；同时，会对宿主机造成瞬间的资源抢占，必然影响其他任务的执行。针对现有技术的缺陷，本申请提供一种解决方案，其主要原理是：对目标QPS值和一秒的时间段进行分配，使得一个较大的QPS值分散到各个子时间段来完成，以解决流量分布不均的问题，从而降低***的瞬时压力。图1为本申请提供的速率控制方法的原理示意图。如图1所示，举例说明，假设目标QPS值为5000，即一秒总共执行5000次，但是，由于任务都是尽力执行，使得任务集中在前面200ms(毫秒)完成，如，第1个100ms完成4800次，第2个100ms完成200次，流量分布极不均匀。本申请提出的方案是，将目标QPS值和一秒的时间段进行拆分匹配，使得一个较大的QPS值分散到各个子时间段来完成。如图1所示，在本实例中，将5000次平均分为10个500次；同时将一秒平均分为10个100ms。使得，在每个子时间段(100ms)内，处理的任务数＝500次。这样做，保证了目标QPS没有变，一秒内执行的总任务数仍然是5000次，但是，平均分散在每段子时间段执行，使得流量分布均匀，极大地降低了***的瞬时压力。本申请实施例提供的方法可应用于任何具有速率控制逻辑的业务***，如待评估性能的***等。
图2为本申请实施例提供的业务***的***框图，图2所示的结构仅仅是本申请的技术方案可以应用的业务***的示例之一。如图2所述，业务***中包括速率控制装置。该速率控制装置主要用来执行下述图3和图4所示的处理流程，其主要用于，将目标QPS值分为数个子值，数个子值之和为目标QPS值；将一秒分为数个子时间段，数个子时间段之和为一秒，子值的个数与子时间段的个数相等，各子值与各子时间段依次对应；依次在每个子时间段内，处理请求，使得需要处理的请求数等于与子时间段对应的子值。将一个较大的QPS值分散到各个子时间段来完成，能够解决流量分布不均的问题，降低***的瞬时压力。
上述实施例是对本申请实施例的技术原理的说明，下面通过多个实施例来进一步对 本申请实施例具体技术方案进行详细描述。
实施例一
图3为本申请提供的速率控制方法一个实施例的流程图，该方法的执行主体可以为上述实施例所描述的业务***。如图3所示，该速率控制方法包括如下步骤：
S301，将目标QPS值分为数个子值，数个子值之和为目标QPS值。
在本申请实施例中，目标QPS值是指，***要求每秒钟必须要达到请求数，本申请实施例中的一个请求用于表示***执行的一个任务，也可以用于表示一个事务(Transaction)，即，访问并可能更新数据库中各种数据项的一个程序执行单元。当业务***给出目标QPS值时，需要将该QPS值分为数个子值，所有的子值之和为目标QPS值。子值的具体个数可以根据经验值而定，也可以根据业务***的性能自适应地设定和调整。
S302，将一秒分为数个子时间段，数个子时间段之和为一秒，子值的个数与子时间段的个数相等，各子值与各子时间段依次对应。
在本方案中，还需要将一秒的时间分为数个子时间段，所有的子时间段之和为一秒。子时间段与上述子值的个数相同，且各个子值与子时间段依次对应。在本申请实施例中，上述步骤S301和S302的执行顺序不分先后，也可以同时执行。
具体地，在采用上述两个步骤对目标QPS值和时间进行划分和匹配时，均可以采用等分或不等分的方式。例如，针对不同业务***的不同性能，可以采用的划分方式有：目标QPS值等分，时间等分(即，目标QPS值和时间都平均分配，例如，将5000次等分为10个500次，将1s等分为10个100ms)；目标QPS值不等分，时间等分(例如，将5000次依次分为1000次、900次、800次、1100次和1200次，将1s等分为5个200ms)；目标QPS值等分，时间不等分(例如，将5000次等分为5个1000次，将1s依次分为200ms、190ms、150ms、2100ms和250ms)；目标QPS值不等分，时间不等分(例如，将5000次依次分为1000次、900次、800次、1100次和1200次，将1s依次分为200ms、190ms、150ms、2100ms和250ms)。
S303，依次在每个子时间段内，处理请求，使得需要处理的请求数等于与该子时间段对应的子值。
在对目标QPS值和时间进行划分和匹配完成后。按时间顺序，依次处理每个子时间段的请求，在每个子时间段内，需要处理的请求数与该时间段对应的子值相同。假设， 针对某业务***采取的划分方式是：将5000次依次分为1000次、900次、800次、1100次和1200次，将1s等分为5个200ms。则，在每个200ms时段内，依次处理的请求数分别为1000次、900次、800次、1100次和1200次。
本申请实施例提供的速率控制方法，通过对目标QPS值和一秒的时间段进行划分和匹配，使得一个较大的QPS值分散到各个子时间段来处理，解决了流量分布不均的问题，极大地降低了***的瞬时压力。
实施例二
图4为本申请提供的速率控制方法另一个实施例的流程图。如图4所示，在上述图3所示实施例的基础上，本实施例提供的速率控制方法还可以包括以下步骤：
S401，将目标QPS值等分为数个子值，数个子值之和为目标QPS值。
S402，将一秒等分为数个子时间段，数个子时间段之和为一秒，子值的个数与子时间段的个数相等，各子值与各子时间段依次对应。
在本申请实施例中，采用一种较优的划分方式，即，目标QPS值和时间均采用等分的方式进行划分，使得流量分布更加均匀。
S403，依次在每个子时间段内，处理请求，当已处理的请求数等于与当前子时间段对应的子值时，启动阻塞机制，使未处理的请求进入等待状态，直到下一个子时间段开始时，关闭该阻塞机制。
在本申请实施例中，可以以阻塞等待的方式来保证每个子时间段内，已处理请求数＝对应的子值。阻塞是指没有获得资源则挂起进程(未处理的请求)，直到获得资源为止。被挂起的进程进入休眠状态，被调度器的运行队列移走，直到等待条件(下一个子时间段开始)被满足。
S404，当某个子时间段结束时，若该子时间段内已处理的请求数小于与子时间段对应的子值，对目标QPS值的剩余值和一秒的剩余时间重新进行分配。
当出现突发状况，使得某段子时间段内没有完成任务时，对目标QPS值的剩余值和剩余时间进行重新分配，以保持目标QPS不变。举例说明，假设在方案实施之初，采用的划分方式是：将5000次等分为10个500次，将1s等分为10个100ms。当第一个子时间段(100ms)结束时，已处理的请求数为400次(＜500次)，则需要对目标QPS值的剩余值(5000次-400次＝4600次)和一秒的剩余时间(1s-100ms＝900ms)重新进行分配。
进一步地，当某个子时间段结束时，若该子时间段内已处理的请求数小于与该子时间段对应的子值，也可以进行如下操作：
对目标QPS值和一秒重新进行分配，使得，子值的个数为子时间段的个数的整数倍(至少为两倍)，每个子时间段与多个子值相对应；依次在每个子时间段内，并行处理请求，使得需要处理的请求的总数等于与该子时间段对应的子值之和。
在本申请实施例中，如果某个子时间段没有完成任务，则可以对目标QPS值和时间进行重新分配。举例说明，假设原划分方式为：将5000次等分为10个500次，将1s等分为10个100ms。那么，可以重新划分为：将5000次等分为10个500次，将1s等分为5个200ms。则，每个200ms对应两个500次，在同一个子时间段(200ms)开启两个线程/进程，并行处理请求，使得需要处理的请求的总数等于与该子时间段对应的子值之和(500+500＝1000次)，从而保证周期内实现目标QPS值(5000次)。
本申请实施例提供的速率控制方法，通过对目标QPS值和一秒的时间段进行等分和匹配，使得一个较大的QPS值平均分散到各个子时间段来处理，进一步提高了流量分布的均匀度，降低了***的瞬时压力；同时，当某个子时间段的任务没有完成时，可以在该周期的剩余时间内对目标QPS值的剩余值重新进行分配，也可以并行处理请求，从而保证周期内实现目标QPS值。
实施例三
图5为本申请提供的速率控制装置实施例的结构示意图，可用于执行如图3和图4所示的方法步骤。如图5所示，该装置可以包括：第一划分模块51、第二划分模块52和执行模块53。
其中，第一划分模块51用于将目标QPS值分为数个子值，数个子值之和为目标QPS值；第二划分模块52用于将一秒分为数个子时间段，数个子时间段之和为一秒，子值的个数与子时间段的个数相等，各子值与各子时间段依次对应；执行模块53用于依次在每个子时间段内，处理请求，使得需要处理的请求数等于与该子时间段对应的子值。
在本申请实施例中，目标QPS值是指，***要求每秒钟必须要达到请求数，本申请实施例中的一个请求用于表示***执行的一个任务，也可以用于表示一个事务(Transaction)，即，访问并可能更新数据库中各种数据项的一个程序执行单元。当业务***给出目标QPS值时，第一划分模块51将该QPS值分为数个子值，所有的子值之和为目标QPS值。子值的具体个数可以根据经验值而定，也可以根据业务***的性能自 适应地设定和调整。第二划分模块52将一秒的时间分为数个子时间段，所有的子时间段之和为一秒。子时间段与上述子值的个数相同，且各个子值与子时间段依次对应。在本申请实施例中，第一划分模块51和第二划分模块52的执行顺序不分先后，也可以同时执行。
第一划分模块51和第二划分模块52对目标QPS值和时间进行划分和匹配，均可以采用等分或不等分的方式。例如，针对不同业务***的不同性能，可以采用的划分方式有：目标QPS值等分，时间等分(即，目标QPS值和时间都平均分配，例如，将5000次等分为10个500次，将1s等分为10个100ms)；目标QPS值不等分，时间等分(例如，将5000次依次分为1000次、900次、800次、1100次和1200次，将1s等分为5个200ms)；目标QPS值等分，时间不等分(例如，将5000次等分为5个1000次，将1s依次分为200ms、190ms、150ms、2100ms和250ms)；目标QPS值不等分，时间不等分(例如，将5000次依次分为1000次、900次、800次、1100次和1200次，将1s依次分为200ms、190ms、150ms、2100ms和250ms)。
在第一划分模块51和第二划分模块52对目标QPS值和时间进行划分和匹配完成后。执行模块53按时间顺序，依次处理每个子时间段的请求，在每个子时间段内，需要处理的请求数与该时间段对应的子值相同。假设，针对某业务***采取的划分方式是：将5000次依次分为1000次、900次、800次、1100次和1200次，将1s等分为5个200ms。则，在每个200ms时段内，依次处理的请求数分别为1000次、900次、800次、1100次和1200次。
具体地，执行模块53还可以用于依次在每个子时间段内，处理请求，当已处理的请求数等于与当前子时间段对应的子值时，启动阻塞机制，使未处理的请求进入等待状态，直到下一个子时间段开始时，关闭阻塞机制。
在本申请实施例中，执行模块53可以以阻塞等待的方式来保证每个子时间段内，已处理请求数＝对应的子值。阻塞是指没有获得资源则挂起进程(未处理的请求)，直到获得资源为止。被挂起的进程进入休眠状态，被调度器的运行队列移走，直到等待条件(下一个子时间段开始)被满足。
进一步地，第一划分模块51还可以用于将目标QPS值等分为数个子值。而第二划分模块52还可以用于将一秒等分为数个子时间段。
在本申请实施例中，可以采用一种较优的划分方式，即，目标QPS值和时间均采用等分的方式进行划分，使得流量分布更加均匀。
更进一步地，本申请实施例提供的速率控制装置，还可以包括：第一重分配模块54。该第一重分配模块54可以用于当某个子时间段结束时，若该子时间段内已处理的请求数小于与子时间段对应的子值，对目标QPS值的剩余值和一秒的剩余时间重新进行分配。
在本申请实施例中，当出现突发状况，使得某段子时间段内没有完成任务时，第一重分配模块54对目标QPS值的剩余值和剩余时间进行重新分配，以保持目标QPS不变。举例说明，假设在方案实施之初，采用的划分方式是：将5000次等分为10个500次，将1s等分为10个100ms。当第一个子时间段(100ms)结束时，已处理的请求数为400次(＜500次)，则第一重分配模块54对目标QPS值的剩余值(5000次-400次＝4600次)和一秒的剩余时间(1s-100ms＝900ms)重新进行分配。
再进一步地，本申请实施例提供的速率控制装置，还可以包括：第二重分配模块55。该第二重分配模块55可以用于当某个子时间段结束时，若该子时间段内已处理的请求数小于与子时间段对应的子值，对目标QPS值和一秒重新进行分配，使得，子值的个数为子时间段的个数的整数倍，每个子时间段与多个子值相对应；执行模块53依次在每个子时间段内，并行处理请求，使得需要处理的请求的总数等于与该子时间段对应的子值之和。
在本申请实施例中，如果某个子时间段没有完成任务，则第二重分配模块55可以对目标QPS值和时间进行重新分配。举例说明，假设原划分方式为：将5000次等分为10个500次，将1s等分为10个100ms。那么，第二重分配模块55可以重新划分为：将5000次等分为10个500次，将1s等分为5个200ms。则，每个200ms对应两个500次，执行模块53在同一个子时间段(200ms)开启两个线程/进程，并行处理请求，使得需要处理的请求的总数等于与该子时间段对应的子值之和(500+500＝1000次)，从而保证周期内实现目标QPS值(5000次)。
本申请实施例提供的速率控制装置，通过对目标QPS值和一秒的时间段进行等分和匹配，使得一个较大的QPS值平均分散到各个子时间段来处理，进一步提高了流量分布的均匀度，降低了***的瞬时压力；同时，当某个子时间段的任务没有完成时，可以在该周期的剩余时间内对目标QPS值的剩余值重新进行分配，也可以并行处理请求，从而保证周期内实现目标QPS值。
实施例四
以上描述了速率控制装置的内部功能和结构，该装置可实现为一种电子设备。图6 为本申请提供的电子设备实施例的结构示意图。如图6所示，该电子设备包括存储器61和处理器62。
存储器61，用于存储程序。除上述程序之外，存储器61还可被配置为存储其它各种数据以支持在电子设备上的操作。这些数据的示例包括用于在电子设备上操作的任何应用程序或方法的指令，联系人数据，电话簿数据，消息，图片，视频等。
存储器61可以由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。
处理器62，与存储器61耦合，执行存储器61所存储的程序，以用于：
将目标QPS值分为数个子值，数个子值之和为目标QPS值；将一秒分为数个子时间段，数个子时间段之和为一秒，子值的个数与子时间段的个数相等，各子值与各子时间段依次对应；依次在每个子时间段内，处理请求，使得需要处理的请求数等于与子时间段对应的子值。
进一步，如图6所示，电子设备还可以包括：通信组件63、电源组件64、音频组件65、显示器66等其它组件。图6中仅示意性给出部分组件，并不意味着电子设备只包括图6所示组件。
通信组件63被配置为便于电子设备和其他设备之间有线或无线方式的通信。电子设备可以接入基于通信标准的无线网络，如WiFi，2G或3G，或它们的组合。在一个示例性实施例中，通信组件63经由广播信道接收来自外部广播管理***的广播信号或广播相关信息。在一个示例性实施例中，所述通信组件63还包括近场通信(NFC)模块，以促进短程通信。例如，在NFC模块可基于射频识别(RFID)技术，红外数据协会(IrDA)技术，超宽带(UWB)技术，蓝牙(BT)技术和其他技术来实现。
电源组件64，为电子设备的各种组件提供电力。电源组件64可以包括电源管理***，一个或多个电源，及其他与为电子设备生成、管理和分配电力相关联的组件。
音频组件65被配置为输出和/或输入音频信号。例如，音频组件65包括一个麦克风(MIC)，当电子设备处于操作模式，如呼叫模式、记录模式和语音识别模式时，麦克风被配置为接收外部音频信号。所接收的音频信号可以被进一步存储在存储器61或经由通信组件63发送。在一些实施例中，音频组件65还包括一个扬声器，用于输出音频信号。
显示器66包括屏幕，其屏幕可以包括液晶显示器(LCD)和触摸面板(TP)。如果屏幕包括触摸面板，屏幕可以被实现为触摸屏，以接收来自用户的输入信号。触摸面板包括一个或多个触摸传感器以感测触摸、滑动和触摸面板上的手势。所述触摸传感器可以不仅感测触摸或滑动动作的边界，而且还检测与所述触摸或滑动操作相关的持续时间和压力。
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。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，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申请各实施例技术方案的范围。

Claims (10)

  1. 一种速率控制方法，其特征在于，包括：
    将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
    将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
    依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
  2. 根据权利要求1所述的速率控制方法，其特征在于，所述依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值，包括：
    依次在每个所述子时间段内，处理请求，当已处理的请求数等于与当前子时间段对应的子值时，启动阻塞机制，使未处理的请求进入等待状态，直到下一个所述子时间段开始时，关闭所述阻塞机制。
  3. 根据权利要求1所述的速率控制方法，其特征在于，所述将目标QPS值分为数个子值，包括：
    将所述目标QPS值等分为数个子值。
  4. 根据权利要求1所述的速率控制方法，其特征在于，所述将一秒分为数个子时间段，包括：
    将一秒等分为数个子时间段。
  5. 根据权利要求1至4中任一权利要求所述的速率控制方法，其特征在于，还包括：
    当某个所述子时间段结束时，若该子时间段内已处理的请求数小于与所述子时间段对应的子值，对所述目标QPS值的剩余值和所述一秒的剩余时间重新进行分配。
  6. 根据权利要求1至4中任一权利要求所述的速率控制方法，其特征在于，还包括：
    当某个所述子时间段结束时，若该子时间段内已处理的请求数小于与所述子时间段对应的子值，对所述目标QPS值和所述一秒重新进行分配，使得，所述子值的个数为所述子时间段的个数的整数倍，每个所述子时间段与多个所述子值相对应；
    依次在每个所述子时间段内，并行处理请求，使得需要处理的请求的总数等于与所述子时间段对应的子值之和。
  7. 一种速率控制装置，其特征在于，包括：
    第一划分模块，用于将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
    第二划分模块，用于将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
    执行模块，用于依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
  8. 根据权利要求7所述的速率控制装置，其特征在于，还包括：
    第一重分配模块，用于当某个所述子时间段结束时，若该子时间段内已处理的请求数小于与所述子时间段对应的子值，对所述目标QPS值的剩余值和所述一秒的剩余时间重新进行分配。
  9. 根据权利要求7所述的速率控制装置，其特征在于，还包括：
    第二重分配模块，用于当某个所述子时间段结束时，若该子时间段内已处理的请求数小于与所述子时间段对应的子值，对所述目标QPS值和所述一秒重新进行分配，使得，所述子值的个数为所述子时间段的个数的整数倍，每个所述子时间段与多个所述子值相对应；
    所述执行模块还用于，依次在每个所述子时间段内，并行处理请求，使得需要处理的请求的总数等于与所述子时间段对应的子值之和。
  10. 一种电子设备，其特征在于，包括：
    存储器，用于存储程序；
    处理器，用于运行所述存储器中存储的所述程序，以用于：
    将目标QPS值分为数个子值，数个所述子值之和为所述目标QPS值；
    将一秒分为数个子时间段，数个所述子时间段之和为一秒，所述子值的个数与所述子时间段的个数相等，各所述子值与各所述子时间段依次对应；
    依次在每个所述子时间段内，处理请求，使得需要处理的请求数等于与所述子时间段对应的子值。
PCT/CN2019/077267 2018-03-15 2019-03-07 速率控制方法、装置及电子设备 WO2019174509A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810215023.2 2018-03-15
CN201810215023.2A CN110278160B (zh) 2018-03-15 2018-03-15 速率控制方法、装置及电子设备

Publications (1)

Publication Number Publication Date
WO2019174509A1 true WO2019174509A1 (zh) 2019-09-19

Family

ID=67908517

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/077267 WO2019174509A1 (zh) 2018-03-15 2019-03-07 速率控制方法、装置及电子设备

Country Status (2)

Country Link
CN (1) CN110278160B (zh)
WO (1) WO2019174509A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023015869A1 (zh) * 2021-08-10 2023-02-16 ***股份有限公司 限流控制方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101764712A (zh) * 2010-01-21 2010-06-30 瑞斯康达科技发展股份有限公司 一种日志信息的传输***及方法
CN104079421A (zh) * 2013-03-27 2014-10-01 ***通信集团北京有限公司 一种域名***防护的方法和***
CN106603695A (zh) * 2016-12-28 2017-04-26 北京奇艺世纪科技有限公司 一种每秒查询率的调整方法及装置
US9641431B1 (en) * 2012-04-18 2017-05-02 Google Inc. System and methods for utilization-based balancing of traffic to an information retrieval system
CN106612238A (zh) * 2015-10-21 2017-05-03 阿里巴巴集团控股有限公司 流量控制方法、装置和***

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103842955B (zh) * 2013-11-28 2016-08-31 华为技术有限公司 一种业务流量控制方法、装置和***
CN107402810B (zh) * 2016-05-20 2020-07-14 阿里巴巴集团控股有限公司 线程分配方法及装置
CN107645456B (zh) * 2016-07-20 2021-04-30 菜鸟智能物流控股有限公司 流量控制方法和流量控制***

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101764712A (zh) * 2010-01-21 2010-06-30 瑞斯康达科技发展股份有限公司 一种日志信息的传输***及方法
US9641431B1 (en) * 2012-04-18 2017-05-02 Google Inc. System and methods for utilization-based balancing of traffic to an information retrieval system
CN104079421A (zh) * 2013-03-27 2014-10-01 ***通信集团北京有限公司 一种域名***防护的方法和***
CN106612238A (zh) * 2015-10-21 2017-05-03 阿里巴巴集团控股有限公司 流量控制方法、装置和***
CN106603695A (zh) * 2016-12-28 2017-04-26 北京奇艺世纪科技有限公司 一种每秒查询率的调整方法及装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023015869A1 (zh) * 2021-08-10 2023-02-16 ***股份有限公司 限流控制方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN110278160A (zh) 2019-09-24
CN110278160B (zh) 2023-02-28

Similar Documents

Publication Publication Date Title
US8706798B1 (en) Systems, methods, and devices for dynamic resource monitoring and allocation in a cluster system
EP3137965B1 (en) Cpu/gpu dcvs co-optimization for reducing power consumption in graphics frame processing
US10732853B2 (en) Dynamic memory management techniques
CN111406250B (zh) 无服务器计算环境中的使用经预取的数据的配设
US9104476B2 (en) Opportunistic multitasking of VOIP applications
CN109408205B (zh) 基于hadoop集群的任务调度方法和装置
WO2019075842A1 (zh) 任务分配方法、装置、存储介质和计算机设备
US9396028B2 (en) Scheduling workloads and making provision decisions of computer resources in a computing environment
EP2919115B1 (en) Task migration method and apparatus
US9983907B2 (en) Resource-aware backfill job scheduling
US9218207B1 (en) Configuring virtualization environments
EP3756093B1 (en) Dynamic processor power management
WO2019174509A1 (zh) 速率控制方法、装置及电子设备
CN115543965A (zh) 跨机房数据处理方法、设备、存储介质及程序产品
KR20130137503A (ko) 리소스 모니터링을 이용한 동적 데이터 처리 장치 및 그 방법
CN103997673B (zh) 一种事件处理方法和装置
WO2016019729A1 (zh) 一种动态调度方法、***及计算机存储介质
JP2024521963A (ja) データアクセス方法、装置及び非一時的なコンピュータ可読記憶媒体
US10318341B2 (en) Selecting and resizing currently executing job to accommodate execution of another job
US20210192467A1 (en) Recurring meeting scheduling
US20210109779A1 (en) Predictive scheduling of computing tasks
US12019909B2 (en) IO request pipeline processing device, method and system, and storage medium
US10007557B2 (en) Computing resource controller and control method for multiple engines to share a shared resource
CN110795318B (zh) 数据处理方法和装置以及电子设备
CN109753340A (zh) 虚拟机快照处理方法、装置及***

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: 19766694

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: 19766694

Country of ref document: EP

Kind code of ref document: A1