CN113971144B - Dynamic mixed lottery method for multiprocessor priority arbitration - Google Patents

Dynamic mixed lottery method for multiprocessor priority arbitration Download PDF

Info

Publication number
CN113971144B
CN113971144B CN202111254497.6A CN202111254497A CN113971144B CN 113971144 B CN113971144 B CN 113971144B CN 202111254497 A CN202111254497 A CN 202111254497A CN 113971144 B CN113971144 B CN 113971144B
Authority
CN
China
Prior art keywords
master device
lottery
master
time
request
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN202111254497.6A
Other languages
Chinese (zh)
Other versions
CN113971144A (en
Inventor
程知
蔡佳楠
何立新
张新
颜士喆
朱家兵
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University
Huainan Normal University
Original Assignee
Hefei University
Huainan Normal University
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 Hefei University, Huainan Normal University filed Critical Hefei University
Priority to CN202111254497.6A priority Critical patent/CN113971144B/en
Publication of CN113971144A publication Critical patent/CN113971144A/en
Application granted granted Critical
Publication of CN113971144B publication Critical patent/CN113971144B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14—Handling requests for interconnection or transfer
    • G06F13/16—Handling requests for interconnection or transfer for access to memory bus
    • G06F13/1605—Handling requests for interconnection or transfer for access to memory bus based on arbitration
    • G06F13/1652—Handling requests for interconnection or transfer for access to memory bus based on arbitration in a multiprocessor architecture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14—Handling requests for interconnection or transfer
    • G06F13/16—Handling requests for interconnection or transfer for access to memory bus
    • G06F13/18—Handling requests for interconnection or transfer for access to memory bus based on priority control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
    • G06F7/58—Random or pseudo-random number generators
    • G06F7/588—Random number generators, i.e. based on natural stochastic processes
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C15/00—Generating random numbers; Lottery apparatus
    • G07C15/006—Generating random numbers; Lottery apparatus electronically

Abstract

The invention relates to a dynamic hybrid lottery method for multiprocessor priority arbitration, which comprises the following steps: the lottery ticket generator dynamically generates an initial ticket number of each main device, and the initial ticket number takes any value in each clock period in a set interval range; calculating the time weight f (t) of each master device according to the sequence of the request time; calculating the total number of lottery ticket holding of each master device after considering the time weight; the random number generator generates a random number, compares the random number with the total number of lottery held by each master device, and selects the corresponding master device for output. The invention not only considers that the master equipment with higher priority is endowed with a larger number of lottery tickets, but also considers that the lottery ticket number obtained by the master equipment is influenced by the time of the master equipment sending out the request, and the earlier the request time is, the larger the lottery ticket number is; the time weight function is provided to correct the existing lottery calculation method, so that the elements of the request time are fully considered in the calculation of the lottery number.

Description

Dynamic mixed lottery method for multiprocessor priority arbitration
Technical Field
The invention relates to the technical field of integrated circuit design, in particular to a dynamic hybrid lottery method for multiprocessor priority arbitration.
Background
For a board-level system with multiple processors, when multiple master devices request to contend for the same transmission channel, the lottery priority can give different devices different weights, and the higher priority devices have more lottery numbers and have wider application. However, the conventional lottery priority, whether static lottery priority or dynamic lottery priority, does not consider the time of message request, and decides which device to respond to only according to the number of generated tickets, which results in some requests that may not be responded to all the time, although early.
In some specific situations, where load balancing is required for information transfer between the master and slave devices, this may result in some devices being in a "busy" state and others being in an "idle" state, which obviously is detrimental to balanced distribution of information flows between the devices. On the premise of considering the priority of the equipment, the factor of the sequence of the request time can be properly considered, and the problem of how to embody the variable of the request time on the basis of the original priority is urgently needed to be solved.
Disclosure of Invention
The invention aims to provide a dynamic hybrid lottery method for multiprocessor priority arbitration, which is used for taking lottery priority and first come first serve arbitration into consideration and modulates the lottery priority by using request time.
In order to achieve the above purpose, the present invention adopts the following technical scheme: a dynamic hybrid lottery method for multiprocessor priority arbitration, the method comprising the sequential steps of:
the method comprises the following steps in sequence:
(1) The lottery ticket generator dynamically generates an initial ticket number of each main device, and the initial ticket number takes any value in each clock period in a set interval range;
(2) Calculating the time weight f (t) of each master device according to the sequence of the request time;
(3) Calculating the total number of lottery ticket holding of each master device after considering the time weight;
(4) The random number generator generates a random number, compares the random number with the total number of lottery ticket holding of each master device obtained in the step (3), and selects the corresponding master device for output.
In step (2), the calculation formula of the time weight f (t) of each master device is as follows:
f(t)＝(n-i+1)/n(1)
wherein n is the total number of the main devices; i is the order of sending out the request, and the order corresponds to the order of the request time;
the weight is distributed according to the request time, and the earlier the request time is, the larger the weight is; assuming that the total number of the master devices is n, the order of the master device requests is i, the time weight of the master device requests is (n-i+1)/n, and thus, the device with the earliest request time, i.e., the 1 st requested device, has a weight of 1, and sequentially shifts back with time, and the time weight of the master device is (n-1)/n, (n-2)/n,...
The step (3) specifically refers to: adding the time weight to the number of the lottery generated by the lottery generator to obtain the total number of the lottery held by each master device after considering the time weight:
assuming that n master devices exist, the total number of lottery ticket holds generated by each master device is s 0 ＝r 0 *b 0 f 0 (t)，s 1 ＝r 1 *b 1 f 1 (t)+s 0 ，s 2 ＝r 2 *b 2 f 2 (t)+s 1 ，...，s n-1 ＝r n-1 *b n-1 f n-1 (t)+s n-2 ，s n ＝r n *b n f n (t)+s n-1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein r is 0 ,r 1 ,r 2 ,r n-1 ,r n The initial distribution ticket numbers of master device 0, master device 1, master device 2, master device n-1 and master device n, respectively, which dynamically change over time in a dynamic lottery priority method; b 0 ，b 1 ，b 2 ，...，b n-1 ，b n Respectively representing a master device 0, a master device 1, a master device 2, whether a request is issued by a master device n-1 and a master device n, and if the request is issued, the value is 1, otherwise, the value is 0; f (f) 0 ，f 1 ，f 2 ，...，f n-1 ，f n Representing the time weights of master 0, master 1, master 2, master n-1 and master n, respectively.
The step (4) specifically refers to: generating a random number C by a random number generator, if the value of C is within (0, s 0 ) Selecting a first master device within the range; if the value of C is within(s) 0 ,s 1 ) Selecting a second master device within the range; if the value of C is within(s) 1 ,s 2 ) Selecting a third master device within the range; if the value of C is within(s) n-1 ,s n ) Selecting an nth master device within the range; in the case where the request signal is valid and the permission signal is valid, the selected device is output as the highest priority device.
According to the technical scheme, the beneficial effects of the invention are as follows: firstly, the invention not only considers that the master equipment with higher priority is endowed with a larger number of lottery tickets, but also considers that the lottery ticket number obtained by the master equipment is influenced by the time of the master equipment sending out the request, and under the condition that other conditions are unchanged, the earlier the time of the request is, the more the lottery ticket number is; secondly, the invention provides a function expression of time weight, the weight function is in direct proportion to the sequence of the request time, and the earlier the request time is, the larger the value of the weight function is; third, the present invention corrects the existing lottery calculation method using the proposed time weight function, so that the lottery number calculation fully considers the time factor.
Drawings
FIG. 1 is a schematic diagram of time weight distribution;
FIG. 2 is a schematic diagram of a dynamic hybrid lottery arbitration priority for request time modulation;
FIG. 3 is a schematic diagram of time weights of four master devices in a lottery priority arbitration method;
fig. 4 is a flow chart of the method of the present invention.
Detailed Description
As shown in fig. 4, a dynamic hybrid lottery method for multiprocessor priority arbitration, the method comprising the following sequential steps:
the method comprises the following steps in sequence:
(1) The lottery ticket generator dynamically generates an initial ticket number of each main device, and the initial ticket number takes any value in each clock period in a set interval range; the lottery ticket generator may be implemented as a random number generator;
(2) Calculating the time weight f (t) of each master device according to the sequence of the request time;
(3) Calculating the total number of lottery ticket holding of each master device after considering the time weight;
(4) The random number generator generates a random number, compares the random number with the total number of lottery ticket holding of each master device obtained in the step (3), and selects the corresponding master device for output.
In step (2), the calculation formula of the time weight f (t) of each master device is as follows:
f(t)＝(n-i+1)/n(1)
wherein n is the total number of the main devices; i is the order of sending out the request, and the order corresponds to the order of the request time;
the weight is distributed according to the request time, and the earlier the request time is, the larger the weight is; assuming that the total number of the master devices is n, the order of the master device requests is i, the time weight of the master device requests is (n-i+1)/n, and thus, the device with the earliest request time, i.e., the 1 st requested device, has a weight of 1, and sequentially shifts back with time, and the time weight of the master device is (n-1)/n, (n-2)/n,...
The step (3) specifically refers to: adding the time weight to the number of the lottery generated by the lottery generator to obtain the total number of the lottery held by each master device after considering the time weight:
assuming that n master devices exist, the total number of lottery ticket holds generated by each master device is s 0 ＝r 0 *b 0 f 0 (t)，s 1 ＝r 1 *b 1 f 1 (t)+s 0 ，s 2 ＝r 2 *b 2 f 2 (t)+s 1 ，...，s n-1 ＝r n-1 *b n-1 f n-1 (t)+s n-2 ，s n ＝r n *b n f n (t)+s n-1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein r is 0 ,r 1 ,r 2 ,r n-1 ,r n The initial distribution ticket numbers of master device 0, master device 1, master device 2, master device n-1 and master device n, respectively, which dynamically change over time in a dynamic lottery priority method; b 0 ，b 1 ，b 2 ，...，b n-1 ，b n Respectively representing a master device 0, a master device 1, a master device 2, whether a request is issued by a master device n-1 and a master device n, and if the request is issued, the value is 1, otherwise, the value is 0; f (f) 0 ，f 1 ，f 2 ，...，f n-1 ，f n Representing the time weights of master 0, master 1, master 2, master n-1 and master n, respectively.
The step (4) specifically refers to: generating a random number C by a random number generator, if the value of C is within (0, s 0 ) Selecting a first master device within the range; if the value of C is within(s) 0 ,s 1 ) Selecting a second master device within the range; if the value of C is within(s) 1 ,s 2 ) Selecting a third master device within the range; if the value of C is within(s) n-1 ,s n ) Selecting an nth master device within the range; in the case where the request signal is valid and the permission signal is valid, the selected device is output as the highest priority device. The request signal here refers to b sent by the master device 0 、b 1 、b 2 And b 3 When the request signal is valid, the value is 1, and when the request signal is invalid, the value is 0; the permission signal refers to g sent by the answering device 0 、g 1 、g 2 、g 3 The value is 1 when the enable signal is valid, and 0 when the enable signal is invalid.
The invention is further described below in connection with fig. 1 to 4.
As shown in fig. 1, master X, master Y,..master K, master Q,..master C, master D, sequentially enter the circular queue in chronological order. Assuming that the total number of master devices is n, the request time allocation weights are as follows:
assuming that the total number of the master devices is n, the order of the master device requests is i, the time weight of the master device requests is (n-i+1)/n, and thus, the device with the earliest request time, i.e., the 1 st requested device, has a weight of 1, and sequentially shifts back with time, and the time weight of the master device is (n-1)/n, (n-2)/n,...
As shown in fig. 2, a variable of request time is added to the number of generated tickets based on the conventional dynamic ticket priority arbitration. In the figure r 0 ,r 1 ,r 2 And r 3 The ticket numbers are initially distributed to the master device 0, the master device 1, the master device 2 and the master device 3 respectively, and the ticket numbers corresponding to different master devices in each clock period are dynamically changed. The traditional master demand signal is a time independent boolean variable, which is set to 1 when requested; when there is no request, the demand signal is set to 0. Unlike conventional dynamic lottery generation, the present invention proposes to set the demand signal as a time-dependent function, i.e. b 0 f 0 (t)、b 1 f 1 (t)、b 2 f 2 (t) and b 3 f 3 (t) wherein b 0 、b 1 、b 2 And b 3 The boolean variable indicates whether a request signal is issued, and the value is only 0 or 1.f (f) 0 (t)、f 1 (t)、f 2 (t) and f 3 (t) is a time-dependent weight function. At r 0 ,r 1 ,r 2 And r 3 And b 0 f 0 (t)、b 1 f 1 (t)、b 2 f 2 (t) and b 3 f 3 Under the combined action of (t), the number of generated lottery tickets is s respectively 0 ＝r 0 *b 0 f 0 (t)、s 1 ＝r 1 *b 1 f 1 (t)+s 0 、s 2 ＝r 2 *b 2 f 2 (t)+s 1 Sum s 3 ＝r 3 *b 3 f 3 (t)+s 2 . At this time, the random number generated by the random number generator is equal to s 0 、s 1 、s 2 S 3 In comparison, generate a priority order, g 0 、g 1 、g 2 、g 3 And representing the permission signal, and sequentially selecting the master device with the highest priority to respond when the permission signal is valid.
As shown in fig. 3, if the master device 0, the master device 1, the master device 2, and the master device 3 are respectively enqueued 2, enqueued 4, enqueued 3, and enqueued 1 according to the order of time sequencing. Then according to the formula (n-i+1)/n (where n is the total number of devices, i is the enqueue order, the earlier the order means the earlier the request time), the time weights of master 0, master 1, master 2, master 3 are respectively: 3/4,1/4,2/4,1.
Based on the calculated time weights in FIG. 3, the total number of ticket holds per master device s generated by the ticket generator in FIG. 2 0 ，s 1 ，s 2 Sum s 3 Can be written as formula (1)
In the method, in the process of the invention,represents an upward rounding, r 0 ,r 1 ,r 2 And r 3 Initial lottery hold number for each master device, b 0 、b 1 、b 2 And b 3 Indicating whether a request signal is sent, if so, the value is 1, otherwise, the value is 0.
In summary, the invention considers not only giving a larger number of lottery numbers to the master device with higher priority, but also considering that the lottery numbers obtained by the master device are influenced by the time of the master device sending out the request, and under the condition that other conditions are unchanged, the earlier the request time is, the larger the lottery numbers are; secondly, the invention provides a function expression of time weight, the weight function is in direct proportion to the sequence of the request time, and the earlier the request time is, the larger the value of the weight function is; third, the present invention corrects the existing lottery calculation method using the proposed time weight function, so that the lottery number calculation fully considers the time factor.

Claims (3)

1. A dynamic hybrid lottery method for multiprocessor priority arbitration, characterized by: the method comprises the following steps in sequence:
(1) The lottery ticket generator dynamically generates an initial ticket number of each main device, and the initial ticket number takes any value in each clock period in a set interval range;
(2) Calculating the time weight f (t) of each master device according to the sequence of the request time;
(3) Calculating the total number of lottery ticket holding of each master device after considering the time weight;
(4) The random number generator generates a random number, compares the random number with the lottery holding total number of each main device obtained in the step (3), and selects the corresponding main device for outputting;
in step (2), the calculation formula of the time weight f (t) of each master device is as follows:
f(t)＝(n-i+1)/n (1)
wherein n is the total number of the main devices; i is the order of sending out the request, and the order corresponds to the order of the request time;
the weight is distributed according to the request time, and the earlier the request time is, the larger the weight is; assuming that the total number of the master devices is n, the order of the master device requests is i, the time weight of the master device requests is (n-i+1)/n, and thus, the device with the earliest request time, i.e., the 1 st requested device, has a weight of 1, and sequentially shifts back with time, and the time weight of the master device is (n-1)/n, (n-2)/n,...
2. The dynamic hybrid lottery method for multi-processor priority arbitration of claim 1, wherein: the step (3) specifically refers to: adding the time weight to the number of the lottery generated by the lottery generator to obtain the total number of the lottery held by each master device after considering the time weight:
assuming that n master devices exist, the total number of lottery ticket holds generated by each master device is s 0 ＝r 0 *b 0 f 0 (t)，s 1 ＝r 1 *b 1 f 1 (t)+s 0 ，s 2 ＝r 2 *b 2 f 2 (t)+s 1 ，...，s n-1 ＝r n-1 *b n-1 f n-1 (t)+s n-2 ，s n ＝r n *b n f n (t)+s n-1 The method comprises the steps of carrying out a first treatment on the surface of the Wherein r is 0 ,r 1 ,r 2 ,r n-1 ,r n The initial distribution ticket numbers of master device 0, master device 1, master device 2, master device n-1 and master device n, respectively, which dynamically change over time in a dynamic lottery priority method; b 0 ，b 1 ，b 2 ，...，b n-1 ，b n Respectively representing a master device 0, a master device 1, a master device 2, whether a request is issued by a master device n-1 and a master device n, and if the request is issued, the value is 1, otherwise, the value is 0; f (f) 0 ，f 1 ，f 2 ，...，f n-1 ，f n Representing the time weights of master 0, master 1, master 2, master n-1 and master n, respectively.
3. The dynamic hybrid lottery method for multi-processor priority arbitration of claim 1, wherein: the step (4) specifically refers to: generating a random number C by a random number generator, if the value of C is within (0, s 0 ) Selecting a first master device within the range; if the value of C is within(s) 0 ,s 1 ) Within the range ofSelecting a second master device; if the value of C is within(s) 1 ,s 2 ) Selecting a third master device within the range; if the value of C is within(s) n-1 ,s n ) Selecting an nth master device within the range; in the case where the request signal is valid and the permission signal is valid, the selected device is output as the highest priority device.
CN202111254497.6A 2021-10-27 2021-10-27 Dynamic mixed lottery method for multiprocessor priority arbitration Active CN113971144B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111254497.6A CN113971144B (en) 2021-10-27 2021-10-27 Dynamic mixed lottery method for multiprocessor priority arbitration

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111254497.6A CN113971144B (en) 2021-10-27 2021-10-27 Dynamic mixed lottery method for multiprocessor priority arbitration

Publications (2)

Publication Number Publication Date
CN113971144A CN113971144A (en) 2022-01-25
CN113971144B true CN113971144B (en) 2024-04-16

Family

ID=79588574

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111254497.6A Active CN113971144B (en) 2021-10-27 2021-10-27 Dynamic mixed lottery method for multiprocessor priority arbitration

Country Status (1)

Country Link
CN (1) CN113971144B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115580585B (en) * 2022-11-17 2023-05-12 沐曦集成电路(南京)有限公司 Balanced arbitration method based on arbiter

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5463624A (en) * 1994-04-15 1995-10-31 Dsc Communications Corporation Bus arbitration method for telecommunications switching
US5956493A (en) * 1996-03-08 1999-09-21 Advanced Micro Devices, Inc. Bus arbiter including programmable request latency counters for varying arbitration priority
CN102576350A (en) * 2009-08-21 2012-07-11 谷歌公司 System and method of caching information
CN103299271A (en) * 2011-01-11 2013-09-11 惠普发展公司，有限责任合伙企业 Concurrent request scheduling
CN107301546A (en) * 2017-05-03 2017-10-27 北京众享比特科技有限公司 Generating random number and methods for using them and device
JP2021082103A (en) * 2019-11-21 2021-05-27 ルネサスエレクトロニクス株式会社 Arbitration circuit, data forwarding system, and arbitration method by arbitration circuit

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6877053B2 (en) * 2001-01-03 2005-04-05 Nec Corporation High performance communication architecture for circuit designs using probabilistic allocation of resources
US11538513B2 (en) * 2019-08-16 2022-12-27 Micron Technology, Inc. Memory element for weight update in a neural network
DE102019128651A1 (en) * 2019-10-23 2021-04-29 Infineon Technologies Ag Device for a single wire digital bus, master device, sensor and method for assigning addresses to multiple devices on a single wire digital bus

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5463624A (en) * 1994-04-15 1995-10-31 Dsc Communications Corporation Bus arbitration method for telecommunications switching
US5956493A (en) * 1996-03-08 1999-09-21 Advanced Micro Devices, Inc. Bus arbiter including programmable request latency counters for varying arbitration priority
CN102576350A (en) * 2009-08-21 2012-07-11 谷歌公司 System and method of caching information
CN103299271A (en) * 2011-01-11 2013-09-11 惠普发展公司，有限责任合伙企业 Concurrent request scheduling
CN107301546A (en) * 2017-05-03 2017-10-27 北京众享比特科技有限公司 Generating random number and methods for using them and device
JP2021082103A (en) * 2019-11-21 2021-05-27 ルネサスエレクトロニクス株式会社 Arbitration circuit, data forwarding system, and arbitration method by arbitration circuit

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
An efficient memory arbitration algorithm for a single chip MPEG2 AV decoder;T. Takizawa etal;《2001 INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS, DIGEST OF TECHNICAL PAPERS》;20010721;182-183 *
Design a low latency Arbiter for on chip Communication Architecture;Khanam, R etal;《 2015 INTERNATIONAL CONFERENCE ON COMPUTING, COMMUNICATION & AUTOMATION (ICCCA)》;20160922;1421-1426 *
多路总线仲裁算法优化研究;赵琳;《电子设计工程》;20180405;第26卷(第07期);14-17 *
面向抖动优化的任务静态优先级指派算法;檀明 等;《计算机工程》;20121020;第38卷(第20期);282-285 *

Also Published As

Publication number Publication date
CN113971144A (en) 2022-01-25

Similar Documents

Publication Publication Date Title
WO2022105589A1 (en) Resource scheduling method and apparatus, electronic device and computer readable medium
US6877053B2 (en) High performance communication architecture for circuit designs using probabilistic allocation of resources
US20080209093A1 (en) Fine-grained bandwidth control arbiter and the method thereof
CN113971144B (en) Dynamic mixed lottery method for multiprocessor priority arbitration
WO2014094495A1 (en) Cloud scheduling method and system thereof
CN107562545A (en) A kind of container dispatching method based on Docker technologies
CN108762687B (en) I/O service quality control method, device, equipment and storage medium
CN100514316C (en) Dynamic self-adaptive bus arbiter based on microprocessor-on-chip
CN109548161A (en) A kind of method, apparatus and terminal device of wireless resource scheduling
CN113283778A (en) Layered convergence federated learning method based on security evaluation
CN105320565A (en) Computer resource scheduling method for various application software
CN109783236A (en) Method and apparatus for output information
Kumar et al. ON THE M [supx]/G/1 RETRIAL QUEUE WITH BERNOULLI SCHEDULES AND GENERAL RETRIAL TIMES.
US5898847A (en) Bus arbitration method and appparatus for use in a multiprocessor system
CN111045804B (en) Task allocation method, device, medium and electronic equipment
CN113703945A (en) Scheduling method, device, equipment and storage medium of micro-service cluster
CN110149341A (en) Cloud system user access control method based on suspend mode
WO2020177249A1 (en) Virtual channel-based operation unit sharing method and system
CN110234167A (en) A kind of method for channel allocation, channel dividing arrangement and electronic equipment
TWI332615B (en) A fine-grained bandwidth control arbiter and the method thereof
CN115249074B (en) Distributed federal learning method based on Spark cluster and Ring-AllReduce architecture
KR20200144755A (en) Apparatus and method for consensus based on blockchain
CN113645324B (en) Queue-based IP distribution method and system
CN116342075B (en) Intelligent building integrated cooperative big data processing method
CN117119600A (en) CCE allocation method, device, network equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant