CN106293917A - The optimization method of a kind of I O scheduling cfq algorithm and system - Google Patents
The optimization method of a kind of I O scheduling cfq algorithm and system Download PDFInfo
- Publication number
- CN106293917A CN106293917A CN201610648207.9A CN201610648207A CN106293917A CN 106293917 A CN106293917 A CN 106293917A CN 201610648207 A CN201610648207 A CN 201610648207A CN 106293917 A CN106293917 A CN 106293917A
- Authority
- CN
- China
- Prior art keywords
- request
- ratio
- cfq
- scheduler
- requests
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a kind of optimization method based on linux block device cfq dispatching algorithm and system, including: add up the I/O Request number of each IO process respectively；Using IO processes most for the request number of IO process as the first process, using the IO process of the request number more than second of IO process as the second process；Calculating the ratio of the number of requests of described first process and the number of requests of described second process, when described ratio is more than predetermined threshold, the request number that I/O scheduler performs described first process adjusts according to pre-defined rule；Visible, in the present embodiment, by when the process requested number ratio of the most process of number of requests and number of requests more than second is more than predetermined threshold, by the number of the most process of regulation I/O scheduler execution number of requests, accelerate request processing speed, it is to avoid I O access is delayed.
Description
Technical field
The present invention relates to linux block device I O scheduling technical field, more particularly, it relates to one sets based on linux block
The optimization method of standby I O scheduling cfq algorithm and system.
Background technology
The overall goal of the I/O scheduler of Linux is desirable to magnetic head can always be moved toward a direction, moves to the end
Walking toward opposite direction, this is exactly exactly real-life elevator model, so I/O scheduler is also designated as elevator again
(elevator) algorithm is the most just called elevator algorithm and accordingly.And the elevator algorithm of I O scheduling includes as in Linux
(Anticipatory), cfq (Complete Fairness Queueing), deadline, noop (No Operation),
In up-to-date kernel version and release, all select CFQ as the I/O scheduler of acquiescence, be also for general server
Good selection.CFQ attempts to be evenly distributed the access to I/O bandwidth, it is to avoid relatively low delay is died of hunger and realized to process, is
The compromise of deadline and as scheduler.CFQ is best choosing for multimedia application (video, audio) and desktop system
Select.CFQ gives I/O and asks a priority, and I/O priority request is independent of process priority, the process of high priority
High I/O priority can not be automatically inherited in read-write.CFQ is each process/thread, and one queue of independent establishment manages this and enters
The produced request of journey, say, that one queue of each process, the scheduling between individual queue uses timeslice to dispatch, with this
Ensure that each process can be well assigned to I/O bandwidth.I/O scheduler performs 4 requests of a process every time.Mesh
Before, the dispatching algorithm that linux kernel block io accesses is cfq algorithm, fair-queueing between cfq dispatching algorithm process, uses timeslice
The mode of round carries out I O scheduling, and when some process needs intensive IO, the dispatching algorithm of round robin can make the IO of this process
Queue is piled up long, causes I O access delayed.
Therefore, the problem that the delayed those skilled in the art of being of I O access need to solve how is avoided.
Summary of the invention
It is an object of the invention to provide a kind of optimization method based on linux block device I O scheduling cfq algorithm and system,
To realize avoiding I O access delayed.
For achieving the above object, following technical scheme is embodiments provided:
A kind of optimization method based on linux block device cfq dispatching algorithm, including:
Add up the I/O Request number of each IO process respectively；
Using IO processes most for the request number of IO process as the first process, by the request number more than second of IO process
IO process is as the second process；
Calculate the ratio of the number of requests of described first process and the number of requests of described second process, when described ratio is big
When predetermined threshold, the request number that I/O scheduler performs described first process adjusts according to pre-defined rule.
Wherein, the described I/O Request number adding up each IO process respectively, including:
The I/O Request number of each IO process is added up respectively by statistics process.
Wherein, described using the request most IO process of number of IO process as the first process, by the number of request of IO process
The IO process of mesh more than second as the second process, including:
According to each IO process untreated request number, it is ranked up generating IO process according to descending by all IO processes
Sequencing table；
Using IO processes most for request number in described IO process sequencing table as the first process, will request number more than second
IO process as the second process.
Wherein, the described request number that I/O scheduler performs described first process adjusts according to pre-defined rule, including:
Obtain default number, determine double coefficient according to described ratio；
Utilize described default number and described double coefficient, determine the number after adjustment.
Wherein, also include:
When described ratio is more than predetermined threshold, the cfqFlag of described first process is designated as true；
During the request of the first process described in CFQ scheduler dispatches, whether the cfqFlag inquiring about described first process is
true；The number of the most described CFQ scheduler dispatches request request is transmitted according to the number after adjusting；If it is not,
The number of the most described CFQ scheduler dispatches request request is default number.
Wherein, also include:
If detecting, described ratio is not more than described predetermined threshold, then by described, I/O scheduler performs described first process
Request number recover to described default number.
A kind of optimization system based on linux block device cfq dispatching algorithm, including:
Statistical module, for adding up the I/O Request number of each IO process respectively；
Determination module, is used for IO processes most for the request number of IO process as the first process, asking IO process
Ask the IO process of number more than second as the second process；
Computing module, for calculating the number of requests of described first process and the ratio of the number of requests of described second process
Value；
Adjusting module, for when described ratio is more than predetermined threshold, performs asking of described first process by I/O scheduler
Number is asked to adjust according to pre-defined rule.
Wherein, described determination module is according to each IO process untreated request number, by all IO processes according to descending
It is ranked up generating IO process sequencing table；The IO process asking number most in described IO process sequencing table is entered as first
Journey, using the IO process of request number more than second as the second process.
Wherein, described adjusting module includes:
Acquiring unit, is used for obtaining default number；
Double factor determination unit, for determining double coefficient according to described ratio；
Determine unit, be used for utilizing described default number and described double coefficient, determine the number after adjustment.
Wherein, if described adjusting module detects that described ratio is not more than described predetermined threshold, then by described by I O scheduling
Device performs the request number of described first process and recovers to described default number.
By above scheme, it is a kind of based on linux block device cfq dispatching algorithm excellent that the embodiment of the present invention provides
Change method and system, including: add up the I/O Request number of each IO process respectively；IO most for the request number of IO process is entered
Cheng Zuowei the first process, using the IO process of the request number more than second of IO process as the second process；Calculate described first process
The ratio of number of requests of number of requests and described second process, when described ratio is more than predetermined threshold, by I/O scheduler
The request number performing described first process adjusts according to pre-defined rule；Visible, in the present embodiment, number of requests will be worked as most
The process requested number ratio of process and number of requests more than second more than predetermined threshold time, perform to ask by regulation I/O scheduler
Seek the number of the most process of quantity, accelerate request processing speed, it is to avoid I O access is delayed.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to
Other accompanying drawing is obtained according to these accompanying drawings.
Fig. 1 is a kind of optimization method flow process based on linux block device I O scheduling cfq algorithm disclosed in the embodiment of the present invention
Schematic diagram；
Fig. 2 is a kind of optimization system structure based on linux block device I O scheduling cfq algorithm disclosed in the embodiment of the present invention
Schematic diagram.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
The embodiment of the invention discloses a kind of optimization method based on linux block device I O scheduling cfq algorithm and system, with
Realization avoids I O access delayed.
See Fig. 1, a kind of based on linux block device cfq dispatching algorithm the optimization method that the embodiment of the present invention provides, bag
Include:
S101, add up the I/O Request number of each IO process respectively；
Wherein, the described I/O Request number adding up each IO process respectively, including:
The I/O Request number of each IO process is added up respectively by statistics process.
S102, using the request most IO process of number of IO process as the first process, by the request number of IO process the
IO process more than two is as the second process；
Wherein, described using the request most IO process of number of IO process as the first process, by the number of request of IO process
The IO process of mesh more than second as the second process, including:
According to each IO process untreated request number, it is ranked up generating IO process according to descending by all IO processes
Sequencing table；
Using IO processes most for request number in described IO process sequencing table as the first process, will request number more than second
IO process as the second process.
S103, calculate the ratio of the number of requests of described first process and the number of requests of described second process, when described
When ratio is more than predetermined threshold, the request number that I/O scheduler performs described first process adjusts according to pre-defined rule.
Wherein, the described request number that I/O scheduler performs described first process adjusts according to pre-defined rule, including:
Obtain default number, determine double coefficient according to described ratio；
Utilize described default number and described double coefficient, determine the number after adjustment.
Concrete, if detecting, described ratio is not more than described predetermined threshold, then by described by described for I/O scheduler execution
The request number of the first process recovers to described default number.
Concrete, in the present embodiment, when linux io uses cfq dispatching algorithm, except cfq, this is tieed up as each process
Outside the io queue protected, set up another a process and be responsible for the io request number of statistical analysis each io process, and a threshold value is set,
When the ratio of the io number of request of the io number of request of the io request most process of the number process many with io number of request contents reaches this
During threshold value, the request number that I/O scheduler performs this process every time is double.When ratio is less than threshold value, recover default number of times；If
Default number of times is 4 times, then the number of times after double can be 6 times, 8 times, and this can be according to the number of requests of the first process
Determine with the ratio of the number of requests of the second process.Wherein, whether the ratio of number of request reaches the mark (cfqFlag) of threshold value
And the queue id of the most at most io request queue represents with global variable form.
Based on technique scheme, also include:
When described ratio is more than predetermined threshold, the cfqFlag of described first process is designated as true；
During the request of the first process described in CFQ scheduler dispatches, whether the cfqFlag inquiring about described first process is
true；The number of the most described CFQ scheduler dispatches request request is transmitted according to the number after adjusting；If it is not,
The number of the most described CFQ scheduler dispatches request request is default number.
Concrete, in this programme, when I/O Request occurs, each process receiving io request is worked as to the report of statistics process
Front untreated io asks number, and statistics process receives the untreated io request number sequence side by side of each io process transmission and selects
Big io number of request purpose process and time maximum io number of request purpose process, the number obtained is divided by, and with threshold value compares, exceedes threshold
During value (acquiescence 2), labelling cfqFlag is true.
CFQ scheduler is broadly divided into following two step to bottom block device sending request:
1. standard CFQ dispatching algorithm selects a cfq_queue；
2. selecting request to be transmitted from cfq_queue, when inquiry cfqFlag is ture, sending request please
The number asked is double (8 times).
Concrete, the thinking selecting cfq_queue is as follows:
If the timeslice of the most current cfq_queue is not finished, then continue current cfq_queue；
If the timeslice consumption of the most current cfq_queue is over, then in corresponding prio_tree, preferentially select one
Difference between the sector that first access sector of individual cfq_queue, this cfq_queue and whole scheduler finally process must
Must be less than a threshold value, if OK just selects this cfq_queue；
3. if can not find such cfq_queue, then from service_tree, dispatch other queue.
The optimization system provided the embodiment of the present invention below is introduced, and optimization system described below is with described above
Optimization method can be cross-referenced.
See Fig. 2, a kind of based on linux block device cfq dispatching algorithm the optimization system that the embodiment of the present invention provides, bag
Include:
Statistical module 100, for adding up the I/O Request number of each IO process respectively；
Determination module 200, is used for IO processes most for the request number of IO process as the first process, by IO process
The IO process of request number more than second is as the second process；
Computing module 300, is used for the number of requests the calculating described first process number of requests with described second process
Ratio；
Adjusting module 400, for when described ratio is more than predetermined threshold, performing described first process by I/O scheduler
Request number adjusts according to pre-defined rule.
Based on technique scheme, described determination module is according to each IO process untreated request number, by all IO
Process is ranked up generating IO process sequencing table according to descending；By IO processes most for request number in described IO process sequencing table
As the first process, using the IO process of request number more than second as the second process.
Based on technique scheme, described adjusting module includes:
Acquiring unit, is used for obtaining default number；
Double factor determination unit, for determining double coefficient according to described ratio；
Determine unit, be used for utilizing described default number and described double coefficient, determine the number after adjustment.
Based on technique scheme, if described adjusting module detects that described ratio is not more than described predetermined threshold, then will
The described request number that I/O scheduler performs described first process recovers to described default number.
A kind of based on linux block device cfq dispatching algorithm the optimization method of embodiment of the present invention offer and system, bag
Include: add up the I/O Request number of each IO process respectively；Using IO processes most for the request number of IO process as the first process,
Using the IO process of the request number more than second of IO process as the second process；Calculate the number of requests of described first process with described
The ratio of the number of requests of the second process, when described ratio is more than predetermined threshold, performs described first process by I/O scheduler
Request number according to pre-defined rule adjust；Visible, in the present embodiment, by when the most process of number of requests and number of requests
When process requested number ratio more than second is more than predetermined threshold, perform, by regulation I/O scheduler, the process that number of requests is most
Number, accelerate request processing speed, be prevented effectively from the io delayed problem of request of io request intensive process.
In this specification, each embodiment uses the mode gone forward one by one to describe, and what each embodiment stressed is and other
The difference of embodiment, between each embodiment, identical similar portion sees mutually.
Described above to the disclosed embodiments, makes professional and technical personnel in the field be capable of or uses the present invention.
Multiple amendment to these embodiments will be apparent from for those skilled in the art, as defined herein
General Principle can realize without departing from the spirit or scope of the present invention in other embodiments.Therefore, the present invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and principles disclosed herein and features of novelty phase one
The widest scope caused.
Claims (10)
1. an optimization method based on linux block device cfq dispatching algorithm, it is characterised in that including:
Add up the I/O Request number of each IO process respectively；
Using IO processes most for the request number of IO process as the first process, the IO of the request number more than second of IO process is entered
Cheng Zuowei the second process；
Calculate the ratio of the number of requests of described first process and the number of requests of described second process, when described ratio is more than pre-
When determining threshold value, the request number that I/O scheduler performs described first process adjusts according to pre-defined rule.
Optimization method the most according to claim 1, it is characterised in that the described I/O Request number adding up each IO process respectively
Mesh, including:
The I/O Request number of each IO process is added up respectively by statistics process.
Optimization method the most according to claim 2, it is characterised in that described the request most IO of number of IO process is entered
Cheng Zuowei the first process, using the IO process of the request number more than second of IO process as the second process, including:
According to each IO process untreated request number, it is ranked up generating the sequence of IO process according to descending by all IO processes
Table；
Using IO processes most for request number in described IO process sequencing table as the first process, the IO of number more than second will be asked
Process is as the second process.
4. according to the optimization method described in any one in claim 1-3, it is characterised in that described I/O scheduler is performed institute
The request number stating the first process adjusts according to pre-defined rule, including:
Obtain default number, determine double coefficient according to described ratio；
Utilize described default number and described double coefficient, determine the number after adjustment.
Optimization method the most according to claim 4, it is characterised in that also include:
When described ratio is more than predetermined threshold, the cfqFlag of described first process is designated as true；
During the request of the first process described in CFQ scheduler dispatches, whether the cfqFlag inquiring about described first process is true；
The number of the most described CFQ scheduler dispatches request request is transmitted according to the number after adjusting；If it is not, it is then described
The number of CFQ scheduler dispatches request request is default number.
Optimization method the most according to claim 4, it is characterised in that also include:
If detecting, described ratio is not more than described predetermined threshold, then by described I/O scheduler performed described first process please
Number is asked to recover to described default number.
7. an optimization system based on linux block device cfq dispatching algorithm, it is characterised in that including:
Statistical module, for adding up the I/O Request number of each IO process respectively；
Determination module, is used for IO processes most for the request number of IO process as the first process, by the number of request of IO process
The IO process of mesh more than second is as the second process；
Computing module, is used for the ratio of the number of requests calculating described first process and the number of requests of described second process；
Adjusting module, for when described ratio is more than predetermined threshold, performing the number of request of described first process by I/O scheduler
Mesh adjusts according to pre-defined rule.
Optimization system the most according to claim 7, it is characterised in that
All IO processes, according to each IO process untreated request number, are ranked up raw by described determination module according to descending
Become IO process sequencing table；Using IO processes most for request number in described IO process sequencing table as the first process, by number of request
The IO process of mesh more than second is as the second process.
9. according to the optimization system described in any one in claim 7-8, it is characterised in that described adjusting module includes:
Acquiring unit, is used for obtaining default number；
Double factor determination unit, for determining double coefficient according to described ratio；
Determine unit, be used for utilizing described default number and described double coefficient, determine the number after adjustment.
Optimization system the most according to claim 9, it is characterised in that if described adjusting module detects described ratio not
More than described predetermined threshold, then the described request number that I/O scheduler performs described first process is recovered to described acquiescence number
Mesh.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610648207.9A CN106293917A (en) | 2016-08-09 | 2016-08-09 | The optimization method of a kind of I O scheduling cfq algorithm and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610648207.9A CN106293917A (en) | 2016-08-09 | 2016-08-09 | The optimization method of a kind of I O scheduling cfq algorithm and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106293917A true CN106293917A (en) | 2017-01-04 |
Family
ID=57667217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610648207.9A Pending CN106293917A (en) | 2016-08-09 | 2016-08-09 | The optimization method of a kind of I O scheduling cfq algorithm and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106293917A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107368263A (en) * | 2017-07-25 | 2017-11-21 | 郑州云海信息技术有限公司 | The method and system of priority scheduling in a kind of solid state hard disc |
| CN107562383A (en) * | 2017-09-05 | 2018-01-09 | 联想(北京)有限公司 | Information processing method, storage device and storage medium |
| CN107967181A (en) * | 2017-12-19 | 2018-04-27 | 北京小米移动软件有限公司 | The control method and device of critical zone |
| WO2018082570A1 (en) * | 2016-11-02 | 2018-05-11 | 华为技术有限公司 | I/o request scheduling method and device |
| CN114816687A (en) * | 2021-01-22 | 2022-07-29 | 京东方科技集团股份有限公司 | Cluster resource control method and device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103294548A (en) * | 2013-05-13 | 2013-09-11 | 华中科技大学 | Distributed file system based IO (input output) request dispatching method and system |
| CN103500143A (en) * | 2013-09-27 | 2014-01-08 | 华为技术有限公司 | Hard disc parameter adjusting method and device |
| CN105487987A (en) * | 2015-11-20 | 2016-04-13 | 深圳市迪菲特科技股份有限公司 | Method and device for processing concurrent sequential reading IO (Input/Output) |
-
2016
- 2016-08-09 CN CN201610648207.9A patent/CN106293917A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103294548A (en) * | 2013-05-13 | 2013-09-11 | 华中科技大学 | Distributed file system based IO (input output) request dispatching method and system |
| CN103500143A (en) * | 2013-09-27 | 2014-01-08 | 华为技术有限公司 | Hard disc parameter adjusting method and device |
| CN105487987A (en) * | 2015-11-20 | 2016-04-13 | 深圳市迪菲特科技股份有限公司 | Method and device for processing concurrent sequential reading IO (Input/Output) |
Non-Patent Citations (2)
| Title |
|---|
| 李雅芳: "基于服务感知的IO请求调度算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 赵梦茹: "移动终端Linux存储IO性能分析与优化", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018082570A1 (en) * | 2016-11-02 | 2018-05-11 | 华为技术有限公司 | I/o request scheduling method and device |
| US10628216B2 (en) | 2016-11-02 | 2020-04-21 | Huawei Technologies Co., Ltd. | I/O request scheduling method and apparatus by adjusting queue depth associated with storage device based on hige or low priority status |
| CN107368263A (en) * | 2017-07-25 | 2017-11-21 | 郑州云海信息技术有限公司 | The method and system of priority scheduling in a kind of solid state hard disc |
| CN107562383A (en) * | 2017-09-05 | 2018-01-09 | 联想(北京)有限公司 | Information processing method, storage device and storage medium |
| CN107562383B (en) * | 2017-09-05 | 2021-06-15 | 联想(北京)有限公司 | Information processing method, storage device, and storage medium |
| CN107967181A (en) * | 2017-12-19 | 2018-04-27 | 北京小米移动软件有限公司 | The control method and device of critical zone |
| CN114816687A (en) * | 2021-01-22 | 2022-07-29 | 京东方科技集团股份有限公司 | Cluster resource control method and device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108762896B (en) | Hadoop cluster-based task scheduling method and computer equipment | |
| CN106293917A (en) | The optimization method of a kind of I O scheduling cfq algorithm and system | |
| US9916183B2 (en) | Scheduling mapreduce jobs in a cluster of dynamically available servers | |
| US20190324819A1 (en) | Distributed-system task assignment method and apparatus | |
| US10908950B1 (en) | Robotic process automation system with queue orchestration and task prioritization | |
| EP2698711B1 (en) | Method for dispatching central processing unit of hotspot domain virtual machine and virtual machine system | |
| US7712103B2 (en) | Method and system for managing load balancing in data processing system | |
| US8799913B2 (en) | Computing system, method and computer-readable medium for managing a processing of tasks | |
| US20150295970A1 (en) | Method and device for augmenting and releasing capacity of computing resources in real-time stream computing system | |
| CN105389206B (en) | A kind of cloud computation data center resources of virtual machine quickly configuration method | |
| CN110474966B (en) | Method for processing cloud platform resource fragments and related equipment | |
| US20230229495A1 (en) | Task scheduling method and apparatus | |
| US8122449B2 (en) | Determining whether to retain or terminate a thread based on a minimum number of threads in a thread pool and a maximum number of threads allowed waiting on the channel | |
| US9104491B2 (en) | Batch scheduler management of speculative and non-speculative tasks based on conditions of tasks and compute resources | |
| US20130007386A1 (en) | Memory arbiter with latency guarantees for multiple ports | |
| CN107832143B (en) | Method and device for processing physical machine resources | |
| CN103927225A (en) | Multi-core framework Internet information processing and optimizing method | |
| CN112162835A (en) | Scheduling optimization method for real-time tasks in heterogeneous cloud environment | |
| WO2024016596A1 (en) | Container cluster scheduling method and apparatus, device, and storage medium | |
| WO2024021489A1 (en) | Task scheduling method and apparatus, and kubernetes scheduler | |
| CN105677744A (en) | Method and apparatus for increasing service quality in file system | |
| CN110493317B (en) | Method for processing cloud platform resource fragments and related equipment | |
| CN103218326B (en) | Comprehensive arbiter device | |
| CN110764887A (en) | Task rescheduling method and system, and related equipment and device | |
| CN115640113A (en) | Multi-plane flexible scheduling method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170104 |