CN106250241A - A kind of multi-direction distribution data processing system resources is to the method for application program - Google Patents
A kind of multi-direction distribution data processing system resources is to the method for application program Download PDFInfo
- Publication number
- CN106250241A CN106250241A CN201610624977.XA CN201610624977A CN106250241A CN 106250241 A CN106250241 A CN 106250241A CN 201610624977 A CN201610624977 A CN 201610624977A CN 106250241 A CN106250241 A CN 106250241A
- Authority
- CN
- China
- Prior art keywords
- processing system
- data processing
- system resources
- little
- application program
- 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
- G06F9/50—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
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
-
- 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
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- 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
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/48—Indexing scheme relating to G06F9/48
- G06F2209/482—Application
Abstract
The invention discloses a kind of multi-direction distribution data processing system resources method to application program.Comprise the steps: to obtain one group of little code being configured to implement at least one task;Determine the dependency of little intersymbol described in described group;It is at least partially based on the described dependency of described little intersymbol availability based on data processing system resources, dynamically maps the given little code in described group to data processing system resources collection to perform described given little code.The present invention is by by multidirectional computer data, and dynamic distribution data processing system resources, to the method for application program, obtains one group of little code realizing some data processing task;Determine the dependency of these little intersymbols;With dependency based on little intersymbol availability based on data processing system resources and the cost used in connection with of various resource, use the resource of identification in data-oriented processing system, dynamically place and dispatch little code and perform.
Description
Technical field
The invention belongs to field of computer technology, particularly relate to a kind of multi-direction distribution data processing system resources to answering
By the method for program.
Background technology
Modern high end computer configuation comprises thousands of the most millions for the treatment of elements, a large amount of distributed memory,
And non local memorizer, networking component and the storage infrastructure of various rank.These systems are that execution is application consumed
Static state and the dynamic optimization of resource provide huge challenge.Traditionally, computer configuation makes great efforts to provide the most single, simply
The application in space, location, and offer itself is the most semantic to carry out code execution subsequently and data access.The normal form produced
For many years use good, but when by parallel processing rather than being distributed calculating and data and reality by faster clock rate
On realize all hardware accelerate time, described normal form hinders optimized allocation of resources.
A kind of solution for public domain is multiple stage computer to be divided into cluster so that they are each other by one
Individual network communicates.Every computer has the processor of optimum number, such as four, and the operating system of oneself.Often
When this machine data to being saved in another computer perform once-through operation, just set up once with this another computer
Communication.Time required for these communications result in (high-volume) in a large number with the demand operating identical data
Application program, the waiting time problem occurred in the distributed application program the most repeatedly communicated.Waiting time is for separating
The moment sending the request of an access memorizer and the time of the moment receiving the response to this request.
Patent CN98801271.5 (method for distributing memorizer in multi-processor data process system) is by reflecting
Penetrate and give at least one in the virtual address space that software application is associated at one with one there is Coutinuous store address
The region method of distributing physical memory cell, this application program running on a data handling system include one non-all
Even access memory cell, and utilize multiple virtual memory type, perform this mapping by one address correspondence table of scanning
Process, it is characterised in that the method at least includes that this given application software is selected by one with from one group of predefined rule
The step that fixed memorizer allocation rule is connected, and wherein, when this address correspondence table not comprising and this given application
During the item of the Coutinuous store address area in the virtual address that software is associated, step that the method includes generating abnormal information and
According to the subsequent step of one of these memorizer allocation rule distribution physical memory cell, the selection of rule is by given application
Virtual memory type profile used by software is carried out.
Summary of the invention
It is an object of the invention to provide a kind of multi-direction distribution data processing system resources method to application program, logical
Crossing multidirectional computer data, dynamic distribution data processing system resources, to the method for application program, solves calculating
Machine data distribution data processing system resources is to the application program efficiency problem such as slowly.
For solving above-mentioned technical problem, the present invention is achieved by the following technical solutions:
The present invention be a kind of multi-direction distribution data processing system resources to the method for application program, comprise the steps:
A1 obtains one group of little code being configured to implement at least one task;
A2 determines the dependency of little intersymbol described in described group;
A3 is at least partially based on the described dependency of described little intersymbol availability based on data processing system resources, dynamic
Given little code during state ground maps described group gives little code to data processing system resources collection so that execution is described.
Preferably, described mapping includes the function selected from the group consisted of: places, position, reorientate, move
And migration.
Preferably, described mapping includes the function selected from the group consisted of: determines and performs described given little code
Time started, and determine the position performing described given little code.
Preferably, described mapping includes performing mapping based at least one selected from the standard of group consisted of:
1) performance metric of application program is improved, 2) improve the utilization of described data processing system resources, and 3) maximize application program
Performance metric, observe given resource consumption object set simultaneously.
Preferably, described mapping to carry out the property of optimization application relative to the measurement selected from the group consisted of
Can:
The total run time of described program;The operation time of program described in specific part;Perform before specific instruction
Big time delay;The amount of processing unit used;Memory-aided amount;The use of register file;The use of cache memory;
The use of on-chip cache;The use of second level cache;The use of three grades of caches;The use of level-N cache, its
Middle N is positive number;The use of static RAM;The use of dynamic random access memory;The use of global storage;
The use of virtual memory;May be used in not for the amount processor performing described program;May be used in not for
Perform the amount of the memorizer of described program;Energy expenditure;Peak energy consumes;Calculate system lifetim cost;Required depositor
The amount updated;The amount of required core dump memory clear;The usefulness of security enforcement and the cost of security enforcement.
The method have the advantages that
The present invention is by by multidirectional computer data, and dynamic distribution data processing system resources is to application program
Method, obtains one group of little code realizing some data processing task;Determine the dependency of these little intersymbols;With based on little intersymbol
Dependency availability based on data processing system resources and the cost used in connection with of various resource, use the resource of identification to exist
In data-oriented processing system, dynamically place and dispatch little code and perform.
Certainly, the arbitrary product implementing the present invention it is not absolutely required to reach all the above advantage simultaneously.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, embodiment will be described required use below
Accompanying drawing is briefly described, it should be apparent that, the accompanying drawing in describing below is only some embodiments of the present invention, for ability
From the point of view of the those of ordinary skill of territory, on the premise of not paying creative work, it is also possible to obtain the attached of other according to these accompanying drawings
Figure.
Fig. 1 is a kind of multi-direction distribution data processing system resources flow chart to the method for application program of the present invention.
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, those of ordinary skill in the art obtained under not making creative work premise all other
Embodiment, broadly falls into the scope of protection of the invention.
The term of the present invention:
Little code: one group generally can be performed continuously after its input is available till instruction.
Little code collection: the little code character that can process as unit relative to dependency analysis or execution.
Dependency: the little code collection of expression one between two little code collection terminates another directed arc that could start.
Neighbouring: such as state close on storage space, the locality calculated in space, or time or dependency.
Referring to shown in Fig. 1, the present invention is a kind of multi-direction distribution data processing system resources method to application program,
Comprise the steps:
A1 obtains one group of little code being configured to implement at least one task;
A2 determines the dependency of little intersymbol described in described group;
A3 is at least partially based on the described dependency of described little intersymbol availability based on data processing system resources, dynamic
Given little code during state ground maps described group gives little code to data processing system resources collection so that execution is described.
Wherein, the function included selected from the group consisted of is mapped: place, position, reorientate, move and move
Move.Map the function of group included selected from consisting of: determine the time started performing described given little code, and determine and hold
The position of the described given little code of row.
Wherein, mapping includes performing mapping based at least one selected from the standard of the group consisted of: 1) improve
The performance metric of application program, 2) utilization of described data processing system resources, and 3 are improved) maximize the performance of application program
Tolerance, observes given resource consumption object set simultaneously.
Wherein, map to carry out the performance of optimization application relative to the measurement selected from the group consisted of: program
Total run time;The operation time of program described in specific part;Perform the maximum delay before specific instruction;Process used is single
The amount of unit;Memory-aided amount;The use of register file;The use of cache memory;Making of on-chip cache
With;The use of second level cache;The use of three grades of caches;The use of level-N cache, wherein N is positive number;Static with
The use of machine access memorizer;The use of dynamic random access memory;The use of global storage;Making of virtual memory
With;May be used in not for the amount processor performing described program;May be used in not for performing described program
The amount of memorizer;Energy expenditure;Peak energy consumes;Calculate system lifetim cost;The amount that required depositor updates;Needed for deposit
The amount that reservoir is removed;The usefulness of security enforcement and the cost of security enforcement.
It should be noted that in said system embodiment, included unit simply carries out drawing according to function logic
Point, but it is not limited to above-mentioned division, as long as being capable of corresponding function;It addition, each functional unit is concrete
Title also only to facilitate mutually distinguish, is not limited to protection scope of the present invention.
It addition, one of ordinary skill in the art will appreciate that all or part of step realizing in the various embodiments described above method
The program that can be by completes to instruct relevant hardware, and corresponding program can be stored in an embodied on computer readable storage and be situated between
In matter, described storage medium, such as ROM/RAM, disk or CD etc..
Present invention disclosed above preferred embodiment is only intended to help to illustrate the present invention.Preferred embodiment is the most detailed
Describe all of details, be also not intended to the detailed description of the invention that this invention is only described.Obviously, according to the content of this specification,
Can make many modifications and variations.These embodiments are chosen and specifically described to this specification, is to preferably explain the present invention
Principle and actual application so that skilled artisan can be best understood by and utilize the present invention.The present invention is only
Limited by claims and four corner thereof and equivalent.
Claims (5)
1. a multi-direction distribution data processing system resources is to the method for application program, it is characterised in that comprise the steps:
A1 obtains one group of little code being configured to implement at least one task;
A2 determines the dependency of little intersymbol described in described group;
A3 is at least partially based on the described dependency of described little intersymbol availability based on data processing system resources, dynamically
Map the given little code in described group to data processing system resources collection to perform described given little code.
A kind of multi-direction distribution data processing system resources the most according to claim 1 is to the method for application program, and it is special
Levying and be, described mapping includes the function selected from the group consisted of: places, position, reorientate, move and migrates.
A kind of multi-direction distribution data processing system resources the most according to claim 1 is to the method for application program, and it is special
Levying and be, described mapping includes the function selected from the group consisted of: determine the time started performing described given little code,
The described position giving little code is performed with determining.
A kind of multi-direction distribution data processing system resources the most according to claim 1 is to the method for application program, and it is special
Levying and be, described mapping includes performing mapping based at least one selected from the standard of the group consisted of: 1) improving should
With the performance metric of program, 2) utilization of described data processing system resources, and 3 are improved) maximize the performance degree of application program
Amount, observes given resource consumption object set simultaneously.
A kind of multi-direction distribution data processing system resources the most according to claim 1 is to the method for application program, and it is special
Levying and be, described mapping to carry out the performance of optimization application relative to the measurement selected from the group consisted of:
The total run time of described program;The operation time of program described in specific part;Perform the maximum before specific instruction to prolong
Time;The amount of processing unit used;Memory-aided amount;The use of register file;The use of cache memory;One-level
The use of cache;The use of second level cache;The use of three grades of caches;The use of level-N cache, wherein N
For positive number;The use of static RAM;The use of dynamic random access memory;The use of global storage;Empty
Intend the use of memorizer;May be used in not for the amount processor performing described program;May be used in not for holding
The amount of the memorizer of the described program of row;Energy expenditure;Peak energy consumes;Calculate system lifetim cost;Required depositor is more
New amount;The amount of required core dump memory clear;The usefulness of security enforcement and the cost of security enforcement.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610624977.XA CN106250241A (en) | 2016-08-02 | 2016-08-02 | A kind of multi-direction distribution data processing system resources is to the method for application program |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610624977.XA CN106250241A (en) | 2016-08-02 | 2016-08-02 | A kind of multi-direction distribution data processing system resources is to the method for application program |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106250241A true CN106250241A (en) | 2016-12-21 |
Family
ID=57606333
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610624977.XA Pending CN106250241A (en) | 2016-08-02 | 2016-08-02 | A kind of multi-direction distribution data processing system resources is to the method for application program |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106250241A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1237252A (en) * | 1997-09-04 | 1999-12-01 | 布尔有限公司 | Method for allocating memory in multiprocessor data processing system |
CN102741817A (en) * | 2009-09-04 | 2012-10-17 | 英特尔贝内卢克斯公司 | Method for generating a set of instruction compaction schemes, method for compacting a program according to the generated set, and programmable processor capable of executing a program thus compacted |
CN102934081A (en) * | 2010-04-13 | 2013-02-13 | Et国际有限公司 | Runspace method, system and apparatus |
CN103368698A (en) * | 2012-03-31 | 2013-10-23 | 华为技术有限公司 | Multi-in-put multi-out-put (MIMO) pre-coding method, equipment and system |
-
2016
- 2016-08-02 CN CN201610624977.XA patent/CN106250241A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1237252A (en) * | 1997-09-04 | 1999-12-01 | 布尔有限公司 | Method for allocating memory in multiprocessor data processing system |
CN102741817A (en) * | 2009-09-04 | 2012-10-17 | 英特尔贝内卢克斯公司 | Method for generating a set of instruction compaction schemes, method for compacting a program according to the generated set, and programmable processor capable of executing a program thus compacted |
CN102934081A (en) * | 2010-04-13 | 2013-02-13 | Et国际有限公司 | Runspace method, system and apparatus |
CN103368698A (en) * | 2012-03-31 | 2013-10-23 | 华为技术有限公司 | Multi-in-put multi-out-put (MIMO) pre-coding method, equipment and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101425022B (en) | Dynamic allocation of virtual machine devices | |
CN110096336B (en) | Data monitoring method, device, equipment and medium | |
US9229751B2 (en) | Apparatus and method for managing virtual memory | |
US8635626B2 (en) | Memory-aware scheduling for NUMA architectures | |
CN101986285A (en) | Virtual machine storage space management method, system and physical host | |
CN105577806B (en) | A kind of distributed caching method and system | |
CN101093455A (en) | Allocation method for internal memory pool in embedded type operation system | |
CN103049392B (en) | The implementation method of CACHE DIRECTORY and device | |
CN103064797A (en) | Data processing method and virtual machine management platform | |
CN105357322A (en) | Virtual machine distribution method based on topology partition | |
CN102609305A (en) | Method for sharing internal memory in server cluster | |
CN104301442A (en) | Method for achieving client of access object storage cluster based on fuse | |
CN103473115A (en) | Virtual machine placing method and device | |
CN112306415B (en) | GC flow control method and device, computer readable storage medium and electronic equipment | |
CN110399217A (en) | A kind of memory source distribution method, device and equipment | |
JP6974510B2 (en) | Methods, devices, devices and media for processing data | |
CN103488523A (en) | Page access method, page access device and server | |
CN111104219A (en) | Binding method, device, equipment and storage medium of virtual core and physical core | |
CN104346404A (en) | Method, equipment and system for accessing data | |
CN104506669A (en) | IP address distribution system and method facing distributed network simulation platform | |
CN112087332A (en) | Virtual network performance optimization system under cloud edge cooperation | |
CN112084173A (en) | Data migration method and device and storage medium | |
CN102970349B (en) | A kind of memory load equalization methods of DHT network | |
CN107608806A (en) | The system and method for quick access transport data between a kind of virtual machine | |
CN106155936B (en) | A kind of buffer replacing method and relevant apparatus |
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: 20161221 |