CN106940682A - A kind of embedded system optimization method based on programmable storage on piece - Google Patents
A kind of embedded system optimization method based on programmable storage on piece Download PDFInfo
- Publication number
- CN106940682A CN106940682A CN201710132267.XA CN201710132267A CN106940682A CN 106940682 A CN106940682 A CN 106940682A CN 201710132267 A CN201710132267 A CN 201710132267A CN 106940682 A CN106940682 A CN 106940682A
- Authority
- CN
- China
- Prior art keywords
- storage object
- storage
- piece
- programmable
- memory space
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
- G06F12/023—Free address space management
- G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Devices For Executing Special Programs (AREA)
- Memory System (AREA)
Abstract
The invention discloses a kind of embedded system optimization method based on programmable storage on piece, run time first to storage object is analyzed and marked, then the dispatching priority of storage object is calculated, the schedule sequences of storage object programmable storage on piece are generated based on dispatching priority again, the distribution of memory space is carried out finally according to the size that memory space is may be programmed on piece.This method is realized using the frequency of use of storage object, use time as parameter, is constructed the schedule sequences of storage object, is improved the utilization ratio of programmable storage on piece, improve the performance of embedded system, reduces the energy consumption of embedded system.
Description
Technical field
The invention belongs to embedded system technology field, it is more particularly to a kind of based on piece programmable storage it is embedded
System optimization method.
Background technology
Embedded system be according to application the need for, using computer technology, software and hardware is cut, so as to meet fixed
Make desired dedicated computer system.With continuing to develop for embedded system, performance, power consumption and real-time have become insertion
The major requirement of formula system design.In the design of embedded system, the design of storage hierarchy is extremely important；Storage hierarchy is designed
Quality the overall performance of embedded system, energy consumption and cost of implementation are had significant effect.Whole system performance impact is maximum
Be not processor execution speed, but the speed of memory.What is existed always between memory and processor is larger
Gaps between their growth rates, be " storage wall " (Memory Wall) problem.With the continuous improvement of processor computing capability, although deposit
The access speed of reservoir is also in increase, but this gap does not reduce not only, increasing on the contrary.Between processor and storage
Gaps between their growth rates turn into the main bottleneck that is further lifted of influence systematic function.
Therefore, in embedded system, storage subsystem is always to improve the bottleneck of systematic function.Moreover, in insertion
In formula system, storage subsystem is also the main bottleneck of system energy consumption.In embedded systems, the energy consumption of storage subsystem is often
The 50%-70% of whole system energy consumption is reached.The development of SoC technologies causes the speed difference that collection is reduced between processor and storage
Away from being possibly realized.Not only high performance bus on chip can be additionally provided with integrated memory on piece, can effectively improve storage
The speed of device, while reducing energy consumption.On SoC, memory will occupy more than the 50% of chip area.Use the memory on piece
System energy consumption can be effectively reduced, overall performance is improved.Therefore, all it is integrated with memory on many embeded processors.
In embedded system, DRAM and SRAM is two kinds of the most frequently used memories.SRAM speed is DRAM 10-100
Times, but price is also more than 20 times of DRAM.Therefore, in embedded systems, DRAM is typically used as the big accumulation layer of capacity
It is secondary, while there is provided less SRAM to store the most frequently used data to reduce run time.System with SRAM often compares
It is high by more than 20% using only DRAM systematic function.Due to SRAM annual raising 60%, DRAM speed can only be improved every year
7%, this performance gap will be increasing.In the Memory Hierarchy of embedded system, programmable storage is collection on piece
Into the SRAM on piece, DRAM can also be used under partial picture.Programmable storage is different from cache on piece, by software control
System, the control that programmer can be by programming realization to programmable storage on piece, or optimized when compiling is with operation.
Optimized by programmable storage on piece, its basic goal is by usability of program fragments (including code or number
According to) it is assigned to the address space of programmable storage on piece.All these usability of program fragments are referred to as storage object (Memory
Ojbect).Because the size of programmable storage on piece is limited, the storage on programmable storage on piece can be assigned to
The quantity of object is also limited.Optimization method needs the size according to programmable storage on piece, from the storage object of program
The most suitable distribution to complete programmable storage on piece of middle selection.In existing distribution method, storage object is generally concerned with
Frequency of use, is used as the basic basis for estimation of distribution according to the frequency of use size of storage object.But storage object makes
With mode except frequency of use is related, also there is correlation with the time factor such as the use time of storage object.
The content of the invention
It is an object of the invention to provide a kind of embedded system optimization method based on programmable storage on piece, this hair
Bright is that parameter calculates the dispatching priority of storage object from frequency of use, use time of storage object etc., then builds storage pair
The schedule sequences of elephant, and complete the optimization distribution of programmable storage on piece.
The technical solution adopted in the present invention is：A kind of embedded system optimization side based on programmable storage on piece
Method, comprises the following steps：
Step 1：Run time to storage object is analyzed and marked；
Each storage object is a fragment of program, this fragment code data or code and data
Mixture；
For a program P, all storage object collection generated after compiling are combined into MO (P)=(MO0,MO1,
MO2,…,MOn-1), have n storage object；For the storage object MO in MO (P)iFor, storage object MOiUsing frequency
Rate is designated as F (MOi), represent storage object MOiAccess times during program P is run；Storage object MOiUsed every time
Initial time is designated as Tj(MOi), representation program P jth time uses storage object MOiInitial time；
Step 2：Calculate the dispatching priority of storage object；
Step 3：Generate the schedule sequences of storage object programmable storage on piece；
Step 4：The distribution of memory space is may be programmed on piece.
The present invention is compared with background technology, and what is had has the advantages that：
The present invention is a kind of embedded system optimization method based on programmable storage on piece, and its major function is to journey
The run time of storage object in sequence is analyzed and marked, then the frequency of use and use time with storage object etc. is ginseng
Number calculates the dispatching priority of storage object, and the dispatching priority for being then based on storage object generates the scheduling sequence of storage object
Row, finally carry out may be programmed the optimization distribution of memory space on piece.
(1) high efficiency.This method realizes that the synthesis of many kinds of parameters such as the frequency of use based on storage object, use time is excellent
Change design, construct the schedule sequences of storage object, improve the utilization ratio of programmable storage on piece, improve embedded
The performance of system；
(2) it is low in energy consumption.This method realizes the efficient utilization of programmable storage on piece, effectively improves storage object
Allocative efficiency, reduce power consumption of embedded system.
Brief description of the drawings
Fig. 1 is the flow chart of the embodiment of the present invention.
Embodiment
Understand for the ease of those of ordinary skill in the art and implement the present invention, below in conjunction with the accompanying drawings and embodiment is to this hair
It is bright to be described in further detail, it will be appreciated that implementation example described herein is merely to illustrate and explain the present invention, not
For limiting the present invention.
A kind of embedded system optimization method based on programmable storage on piece provided see Fig. 1, the present invention, including
Following steps：
The first step, the run time to storage object are analyzed and marked；
Program generates storage object when being compiled, and storage object can be called in may be programmed on piece when being program operation and deposited
The memory object of reservoir.Each storage object is a fragment of program, and this fragment can be code, can be data,
Can be code and the mixture of data.For a program P, all storage object collection generated after compiling are combined into MO
(P)=(MO0,MO1,MO2,…,MOn-1), have n storage object.For the storage object MO in MO (P)iFor, storage pair
As MOiFrequency of use be designated as F (MOi), represent storage object MOiAccess times during program P is run；Storage object MOi
The initial time used every time is designated as Tj(MOi), representation program P jth time uses storage object MOiInitial time.
For program Pro, the storage object generated after compiling has 5, all storage object collection be combined into MO (Pro)=
(MO0,MO1,MO2,…,MO4).It is as shown in table 1 below for the storage object in MO (Pro)：
Table 1
The initial time that some storage object is used by program Pro every time, wherein a are represented in table 1 with the form of (a, b)
Which times of representation program Pro uses the storage object, and b represents the initial time that the storage object is used by program P.For example, right
In storage object MO0For, (2,4) represent storage object MO0Initial time when being used by program Pro for the 2nd time is 4；For depositing
Store up object MO4For, (2,5) represent storage object MO4Initial time when being used by program Pro for the 2nd time is 5.
Second step, the dispatching priority for calculating storage object；
For the storage object MO in MO (P)i, storage object MOiUsed m times by program P altogether, m is more than or equal to 2, meter
Calculate storage object MOiDispatching priority.Storage object MOiDispatching priority computational methods be：
In above-mentioned formula, Pri (MOi) represent storage object MOiDispatching priority.Pri(MOi) value it is bigger, represent
Storage object MOiDispatching priority it is higher.
Dispatching priority for each storage object in program Pro is as shown in table 2 below：
Table 2
| Storage object | Dispatching priority |
| MO0 | 0.600 |
| MO1 | 0.067 |
| MO2 | 0.036 |
| MO3 | 0.111 |
| MO4 | 0.208 |
3rd step, the schedule sequences for generating storage object programmable storage on piece；
According to the dispatching priority of the storage object calculated in second step, to all storage objects in MO (P) according to depositing
The dispatching priority of object is stored up, descending arrangement is carried out, storage object set A (P) is formed.It is all in storage object set A (P)
Storage object is arranged according to the descending of the dispatching priority of storage object, i.e.,：
A (P)=(MO '0,MO′1,MO′2,…,MO′n-1)
In storage object set A (P) descending arrangement is carried out by all storage objects according to the dispatching priority of storage object
The sequence formed is designated as Q (P), and Q (P) is the schedule sequences of storage object programmable storage on piece.
According to table 2, program Pro storage object carries out descending arrangement according to the dispatching priority of storage object, is formed
Storage object set A (Pro)=(MO0,MO4,MO3,MO1,MO2), Q (Pro)={ MO0,MO4,MO3,MO1,MO2}。
The distribution of memory space is may be programmed in 4th step, piece；
The piece that programmable storage S has on programmable storage S on piece for that can accommodate r storage object, piece
Upper programmable memory space is also r.It is S that memory space element number is may be programmed on the piece that programmable storage S has on piece0
To Sr-1.Program P storage object set A (P) is that the storage object in storage object set A (P) is distributed on piece operationally
Programmable memory space：
If r is more than or equal to n, show that memory space is may be programmed on programmable storage S has on piece piece to be held
Receive all storage objects in storage object set A (P).Therefore, if r is more than or equal to n, by institute in storage object set A (P)
Some storage objects are all stored and may be programmed onto piece in memory space.Being assigned for memory space is may be programmed on piece.
If r is less than n, shows to may be programmed memory space on programmable storage S has on piece piece to accommodate and deposit
Store up storage objects all in object set A (P).Therefore, if r is less than the method for salary distribution that memory space is may be programmed on n, piece
For：
A) it will be may be programmed in preceding r storage object storage in Q (P) to piece in memory space；
B) a certain moment t before completing is performed in program P, there is 1 storage object MO u By program P using completion,
No longer need to be stored in and memory space cell S is may be programmed on piecekIn, now since the r+1 storage object in Q (P), choosing
1 storage object is selected, this storage object meets condition C 1：With the n-r-1 since the r+1 storage object in Q (P)
The maximum dispatching priority initial time that last time is used by program P for the time being is more than t in individual storage object.If finding satisfaction
The storage object MO of the conditionv, then the storage object MO of the condition will be metvStore storage object MO u It can be compiled on the piece at place
Journey memory space cell SkIn.
C) in program P runnings, repeat step b), until in the absence of the storage object for meeting condition C 1.Can on piece
Program storage space is assigned.
Programmable storage S has on programmable storage S on piece for that can accommodate 5 storage objects, piece
It is also 5 that memory space is may be programmed on piece.Memory space element number is may be programmed on the piece that programmable storage S has on piece is
S0To S4.Program Pro storage object set A (Pro) is the storage object point in storage object set A (Pro) operationally
With programmable memory space on piece.Due to r=n=5, by MO0,MO4,MO3,MO1,MO2All storage is stored to programmable on piece
In space, being assigned for memory space is may be programmed on piece.
Programmable storage S has on programmable storage S on piece for that can accommodate 3 storage objects, piece
It is also 3 that memory space is may be programmed on piece.Memory space element number is may be programmed on the piece that programmable storage S has on piece is
S0To S2.Program Pro storage object set A (Pro) is the storage object point in storage object set A (Pro) operationally
With programmable memory space on piece.Because r for 3, n is that 5, r is less than n, showing can on programmable storage S has on piece piece
Program storage space can not accommodate storage objects all in storage object set A (Pro).Therefore, storage is may be programmed on piece empty
Between the method for salary distribution be：
A) by preceding 3 storage object MO in Q (Pro)0,MO4,MO3It is respectively stored on piece and may be programmed memory space unit
S0, S1And S2In；
B) a certain moment t=7 before completing is performed in program Pro, there is 1 storage object MO 0 Made by program Pro
With completion, it is no longer necessary to be stored on piece and may be programmed memory space cell S0In, now from the 4th storage object in Q (Pro)
Start, select 1 storage object, this storage object meets condition C 1：In MO1And MO2There is maximum in the two storage objects
Dispatching priority the initial time that is used by program Pro of last time is more than 7 for the time being.According to Tables 1 and 2, storage object MO1's
Dispatching priority is more than storage object MO2Dispatching priority, and storage object MO1Last time is by rising that program Pro is used
Moment beginning is 12, then finds and meet the storage object of the condition for storage object MO1, the storage object MO of the condition will be met1Deposit
Store up storage object MO 0 Memory space cell S is may be programmed on the piece at place0In.
C) a certain moment t=12 before completing is performed in program Pro, there is 1 storage object MO 1 By program Pro
Use completion, it is no longer necessary to be stored on piece and may be programmed memory space cell S0In, now from the 5th storage pair in Q (Pro)
As starting, 1 storage object is selected, this storage object meets condition C 1：In MO2There is maximum tune in this storage object
The degree initial time that last time is used by program Pro preferential for the time being is more than 12.According to Tables 1 and 2, now only storage object
MO1It is not stored on piece and may be programmed memory space, and storage object MO2The initial time that last time is used by program Pro
For 16, then it is storage object MO to find and meet the storage object of the condition2, the storage object MO of the condition will be met2Store
Store up object MO 0 Memory space cell S is may be programmed on the piece at place0In.
D) a certain moment t=14 before completing is performed in program Pro, there is 1 storage object MO 4 By program Pro
Use completion, it is no longer necessary to be stored on piece and may be programmed memory space cell S1In, now storage object MO0, MO1, MO4All
It is complete through being used up, storage object MO2And MO3Just it is being stored on piece and is may be programmed memory space cell S0And S2In, do not need point
It is fitted on the storage object that memory space is may be programmed on piece.Being assigned for memory space is may be programmed on piece.
It should be appreciated that the part that this specification is not elaborated belongs to prior art.
It should be appreciated that the above-mentioned description for preferred embodiment is more detailed, therefore it can not be considered to this
The limitation of invention patent protection scope, one of ordinary skill in the art is not departing from power of the present invention under the enlightenment of the present invention
Profit is required under protected ambit, can also be made replacement or be deformed, each fall within protection scope of the present invention, this hair
It is bright scope is claimed to be determined by the appended claims.
Claims (4)
1. a kind of embedded system optimization method based on programmable storage on piece, it is characterised in that comprise the following steps：
Step 1：Run time to storage object is analyzed and marked；
Each storage object is a fragment of program, this fragment code data or code and data it is mixed
It is fit；
For a program P, all storage object collection generated after compiling are combined into MO (P)=(MO0,MO1,MO2,…,
MOn-1), have n storage object；For the storage object MO in MO (P)iFor, storage object MOiFrequency of use be designated as F
(MOi), represent storage object MOiAccess times during program P is run；Storage object MOiThe initial time used every time
It is designated as Tj(MOi), representation program P jth time uses storage object MOiInitial time；
Step 2：Calculate the dispatching priority of storage object；
Step 3：Generate the schedule sequences of storage object programmable storage on piece；
Step 4：The distribution of memory space is may be programmed on piece.
2. the embedded system optimization method according to claim 1 based on programmable storage on piece, it is characterised in that：
In step 2, storage object MOiDispatching priority computational methods be：
Pri(MOi) represent storage object MOiDispatching priority, m represents storage object MOiAltogether by program P use time time,
m≥2；Pri(MOi) value it is bigger, represent storage object MOiDispatching priority it is higher.
3. the embedded system optimization method according to claim 1 based on programmable storage on piece, it is characterised in that：
In step 3, according to the dispatching priority of storage object, to dispatching priority of all storage objects in MO (P) according to storage object
Power, carries out descending arrangement, forms storage object set A (P)；All storage objects in storage object set A (P) are according to storage
The descending arrangement of the dispatching priority of object, i.e.,：
A (P)=MO '0,MO′1,MO′2,…,MO′n-1)
In storage object set A (P) descending arrangement institute shape is carried out by all storage objects according to the dispatching priority of storage object
Into sequence be designated as Q (P), Q (P) is the schedule sequences of storage object programmable storage on piece.
4. the embedded system optimization method according to claim 3 based on programmable storage on piece, it is characterised in that：
In step 4, programmable storage S on the piece for r storage object can be accommodated, the piece that programmable storage S has on piece
Upper programmable memory space is also r；It is S that memory space element number is may be programmed on the piece that programmable storage S has on piece0
To Sr-1；Program P storage object set A (P) is that the storage object in storage object set A (P) is distributed on piece operationally
Programmable memory space：
If r >=n, storage object all in storage object set A (P) is all stored memory space is may be programmed onto piece
In；Being assigned for memory space is may be programmed on piece；
If r<The method of salary distribution of programmable memory space is on n, piece：
A) it will be may be programmed in preceding r storage object storage in Q (P) to piece in memory space；
B) a certain moment t before completing is performed in program P, there is 1 storage object MO u Used and complete by program P, no longer
Need to be stored in and memory space cell S is may be programmed on piecekIn, now since the r+1 storage object in Q (P), selection 1
Individual storage object, this storage object meets condition C 1：With n-r-1 since the r+1 storage object in Q (P)
The maximum dispatching priority initial time that last time is used by program P for the time being is more than t in storage object；Should if finding satisfaction
The storage object MO of conditionv, then the storage object MO of the condition will be metvStore storage object MO u It may be programmed on the piece at place
Memory space cell SkIn；
C) in program P runnings, repeat step b), until in the absence of the storage object for meeting condition C 1；It may be programmed on piece
Memory space is assigned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710132267.XA CN106940682B (en) | 2017-03-07 | 2017-03-07 | Embedded system optimization method based on-chip programmable memory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710132267.XA CN106940682B (en) | 2017-03-07 | 2017-03-07 | Embedded system optimization method based on-chip programmable memory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106940682A true CN106940682A (en) | 2017-07-11 |
| CN106940682B CN106940682B (en) | 2020-06-09 |
Family
ID=59468857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710132267.XA Expired - Fee Related CN106940682B (en) | 2017-03-07 | 2017-03-07 | Embedded system optimization method based on-chip programmable memory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106940682B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832085A (en) * | 2017-09-27 | 2018-03-23 | 武汉科技大学 | A kind of flowing water optimization method of reconfigurable task |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6021425A (en) * | 1992-04-03 | 2000-02-01 | International Business Machines Corporation | System and method for optimizing dispatch latency of tasks in a data processing system |
| CN101714161A (en) * | 2009-12-23 | 2010-05-26 | 上海全景数字技术有限公司 | Cache device and method for embedded browser |
| CN201707662U (en) * | 2009-04-09 | 2011-01-12 | 青岛海信移动通信技术股份有限公司 | Embedded type equipment supporting JAVA technology |
| US20110235706A1 (en) * | 2010-03-25 | 2011-09-29 | Texas Instruments Incorporated | Region of interest (roi) video encoding |
| CN103218304A (en) * | 2013-04-03 | 2013-07-24 | 杭州电子科技大学 | On-chip and off-chip distribution method for embedded memory data |
| CN103559148A (en) * | 2013-11-15 | 2014-02-05 | 山东大学 | On-chip scratch-pad memory (SPM) management method facing multitasking embedded system |
| CN103647807A (en) * | 2013-11-27 | 2014-03-19 | 华为技术有限公司 | Information caching method, device and communication apparatus |
-
2017
- 2017-03-07 CN CN201710132267.XA patent/CN106940682B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6021425A (en) * | 1992-04-03 | 2000-02-01 | International Business Machines Corporation | System and method for optimizing dispatch latency of tasks in a data processing system |
| CN201707662U (en) * | 2009-04-09 | 2011-01-12 | 青岛海信移动通信技术股份有限公司 | Embedded type equipment supporting JAVA technology |
| CN101714161A (en) * | 2009-12-23 | 2010-05-26 | 上海全景数字技术有限公司 | Cache device and method for embedded browser |
| US20110235706A1 (en) * | 2010-03-25 | 2011-09-29 | Texas Instruments Incorporated | Region of interest (roi) video encoding |
| CN103218304A (en) * | 2013-04-03 | 2013-07-24 | 杭州电子科技大学 | On-chip and off-chip distribution method for embedded memory data |
| CN103559148A (en) * | 2013-11-15 | 2014-02-05 | 山东大学 | On-chip scratch-pad memory (SPM) management method facing multitasking embedded system |
| CN103647807A (en) * | 2013-11-27 | 2014-03-19 | 华为技术有限公司 | Information caching method, device and communication apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 毛佳: "嵌入式实时系统中关键技术的研究", 《中国优秀博硕士学位论文全文数据库 (博士) 信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832085A (en) * | 2017-09-27 | 2018-03-23 | 武汉科技大学 | A kind of flowing water optimization method of reconfigurable task |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106940682B (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103853618B (en) | Resource allocation method with minimized cloud system cost based on expiration date drive | |
| US9031826B2 (en) | Method and apparatus for simulating operation in a data processing system | |
| EP2400359A1 (en) | System, method, and program product for targeting and optimal driving force distribution in energy recovery systems | |
| CN104361415B (en) | A kind of choosing method and device for showing information | |
| Zapata et al. | The multimode resource constrained multiproject scheduling problem: Alternative formulations | |
| CN110032396A (en) | Storage system and its operating method | |
| CN104035747B (en) | Method and apparatus for parallel computation | |
| Khouja | The economic lot and delivery scheduling problem: common cycle, rework, and variable production rate | |
| CN102708009A (en) | Method for sharing GPU (graphics processing unit) by multiple tasks based on CUDA (compute unified device architecture) | |
| CN103473120A (en) | Acceleration-factor-based multi-core real-time system task partitioning method | |
| CN104281495A (en) | Method for task scheduling of shared cache of multi-core processor | |
| CN105095646A (en) | Data prediction method and electronic device | |
| CN110188086A (en) | Database automated tuning method and device based on load automatic Prediction | |
| CN104915557A (en) | Cloud task allocation method based on double-objective ant colony algorithm | |
| CN106940682A (en) | A kind of embedded system optimization method based on programmable storage on piece | |
| CN114862231A (en) | Production scheduling method and device and electronic equipment | |
| CN103389964A (en) | Apparatus and method for solving mathematical programming problem | |
| US20210005311A1 (en) | Normalizing data sets for predicting an attribute of the data sets | |
| US10445399B2 (en) | Forecast-model-aware data storage for time series data | |
| Brüggemann et al. | DLSP for two-stage multi-item batch production | |
| CN117057550A (en) | Material demand plan calculation method, device, network equipment and medium | |
| US11443262B2 (en) | Energy efficient scheduling of industrial process by reducing idle time and tardiness of jobs | |
| JP2009098924A (en) | Member assignment system | |
| CN115936875A (en) | Financial product form hanging processing method and device | |
| Burke et al. | A multi-stage approach for the thermal generator maintenance scheduling problem |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200609 Termination date: 20210307 |