CN106683709B - NOR and nand flash memory performance measuring and evaluating method - Google Patents
NOR and nand flash memory performance measuring and evaluating method Download PDFInfo
- Publication number
- CN106683709B CN106683709B CN201611230307.6A CN201611230307A CN106683709B CN 106683709 B CN106683709 B CN 106683709B CN 201611230307 A CN201611230307 A CN 201611230307A CN 106683709 B CN106683709 B CN 106683709B
- Authority
- CN
- China
- Prior art keywords
- flash memory
- energy consumption
- read
- erasing
- test
- 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
Links
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/56—External testing equipment for static stores, e.g. automatic test equipment [ATE]; Interfaces therefor
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/56—External testing equipment for static stores, e.g. automatic test equipment [ATE]; Interfaces therefor
- G11C29/56016—Apparatus features
Abstract
The invention discloses a kind of NOR and nand flash memory performance measuring and evaluating method, for solving the technical problem of existing flash memory performance assessment method integrated operation ability difference.Technical solution is test flash memory read or write speed U first1, then successively test flash memory mounting time U2, flash memory average loss U3, flash memory consume U4, flash hardware read/write speed U5, flash memory erasing speed U6, flash memory wipe energy consumption U7, flash hardware read and write energy consumption U8With flash memory file system energy consumption U9, finally using order relation analytic approach to U1To U9Index carries out Comprehensive Assessment.Due to using the read-write for flash memory, load, erasing and the comprehensive evaluating of relevant 9 indexs of energy consumption, to read or write speed, file system service ability and the Estimation of energy consumption being related in flash memory performance assessment, from three file system, hardware driving and energy consumption levels, it is proposed corresponding index system, comprehensive evaluating calculating is carried out to it by order relation method, so that test result is able to reflect the integration capability in terms of flash reading and writing, load, erasing and energy consumption.
Description
Technical field
The present invention relates to a kind of flash memory performance assessment methods, more particularly to a kind of NOR and nand flash memory performance measuring and evaluating side
Method.
Background technique
" NOR type FLASH memory measuring technology, electronics and encapsulation, in March, 2016, Vol16 (3), p15-19 " are public for document
The method that Flash test is carried out using the DSIO module dynamic generation test vector of J750EX system is opened, this method is using hard
Part circuit tests the hardware layer of flash memory, completes first, in accordance with J750EX standard method to flash memory system channel and pin
Setting then writes test vector and instructs setting data transmission trigger point to send data to special pin by microcontroller, finally
By determining that the data of special pin evaluate and test flash hardware situation.The method is the test carried out for the hardware layer of Flash, nothing
Method comprehensively evaluates the performance of flash memory.For such as erasing, mounting time length, the energy consumption characteristics of flash memory, which can not reflect,
And the research in terms of Non-energy-consumption performance measuring and evaluating.In view of the above-mentioned problems, the present invention passes through hard to flash memory file system index, flash memory
Part driving layer index and flash memory energy consumption index are established and are comprehensively considered on the basis of being tested, and comprehensive flash memory performance is established
Measurement system.
Summary of the invention
In order to overcome the shortcomings of that existing flash memory performance assessment method integrated operation ability is poor, the present invention provide a kind of NOR and
Nand flash memory performance measuring and evaluating method.This method test flash memory read or write speed U first1, then successively test flash memory mounting time U2, dodge
Deposit average loss U3, flash memory consume U4, flash hardware read/write speed U5, flash memory erasing speed U6, flash memory wipe energy consumption U7、
Flash hardware reads and writes energy consumption U8With flash memory file system energy consumption U9, finally using order relation analytic approach to U1To U9Index carries out comprehensive
Close evaluation.Due to using the read-write for flash memory, load, erasing and the comprehensive evaluating of relevant 9 indexs of energy consumption, systematically
Read or write speed, file system service ability and the Estimation of energy consumption being related in flash memory performance assessment are analyzed, from file system, firmly
Part driving and three levels of energy consumption, propose corresponding index system, have carried out comprehensive evaluating meter to it finally by order relation method
It calculates, so that test result is able to reflect the integration capability in terms of flash reading and writing, load, erasing and energy consumption.
The technical solution adopted by the present invention to solve the technical problems is: a kind of NOR and nand flash memory performance measuring and evaluating method,
Its main feature is that the following steps are included:
Step 1: test flash memory read or write speed U1, shown in calculation expression such as formula (1), wherein Speed refers to read/write
Speed, TotalSize refer to actual transmissions file size, and TotalTime refers to the actual measurement time.
Speed=TotalSize/TotalTime (1)
Step 2: test flash memory mounting time U2, using multiple circulation carry and corresponding document system is unloaded, asks it average
Value.Shown in mounting time expression formula such as formula (2), wherein End_time is in file system mounted end point time, Start_
Time is the file system mounted starting point time, and M_time is the primary file system mounted time.
M_time=End_Time-Start_Time (2)
Step 3: test flash memory average loss U3, shown in calculation method such as formula (3), formula (4).Abrasion equilibrium difference ε
For defining the erasing times difference in flash memory between erasing times most muti-piece and the minimum block of erasing times.ε value is smaller, represents and dodges
Deposit system longer life expectancy.Standard deviation S evaluates average loss performance indicator as another, and wherein n represents the erasable of flash memory
Block number, EiEvery piece of erasing times are represented, E represents fifty-fifty erasing times, EmaxFor E maximum value EminFor E minimum value.
ε=Emax-Emin (3)
Step 4: test flash memory memory consumption U4, by read under linux kernel/proc/meminfo file dynamic obtains
Take the memory service condition of system at this time.When test module size, after calling system calls function mount carry file system,
Comparison front and back memory headroom Expenditure Levels.
Step 5: test flash memory hardware read/write speed U5, shown in calculation expression such as formula (5), wherein Speed refers to
Read/write speed, TotalSize refer to actual transmissions file size, and TotalTime refers to the actual measurement time.
Step 6: test flash memory erasing speed U6, shown in calculation expression such as formula (6), wherein e_speed refers to wiping
Except speed, Block_Size refers to that practical erasing block size, TotalTime refer to the actual measurement time.Erasing speed test exists
Hardware layer circulation reads the every block message of flash memory, and API is called to carry out erasing operation to flash block.
Step 7: test flash memory wipes energy consumption U7, calculated with formula (7), wherein E indicates energy consumption, and P (t) indicates transient state function
Rate;P indicates average power consumption;I is average current;V is operating voltage, and T is runing time.
Step 8: test flash memory hardware reads and writes energy consumption U8, obtained and dodged by API provided by MTD layers in linux kernel
Deposit information.Since flash reading and writing energy consumption belongs to hardware energy consumption, the true energy of hardware is tested by fluke scopmeter table
Consumption.
Step 9: test flash memory file system energy consumption U9, it is the interface function that provides of user according to file system, to needing to survey
Fixed energy consumption index is quantified.It calls the energy consumption Eroutine of function by node f0, f1 ... .fn composition.Function fiSon
Function is fj1, and fj2 ... .fjm, the single clock cycle energy consumption of instruction is ei, then it calls functionEnergy consumption formulas such as
(8) shown in.
For each function, circulation measurement energy consumption is taken repeatedly to average.
Step 10: using order relation analytic approach to U1To U9Index carries out Comprehensive Assessment, and comprehensive evaluation value calculation expression is
(9) shown in:
Z=∑I=1xiwi (9)
Wherein xiFor calculative each index, weight coefficient wkCalculation method such as formula (11), shown in (12), comment
K=m in relative importance formula, m-1 ..., 3,2 take r by order relation when m is larger between valence indexk=1.rkValue
It is 1.0,1.2,1.4,1.6,1.8.
wk-1=wkrk (12)。
The beneficial effects of the present invention are: this method test flash memory read or write speed U first1, then successively test flash memory carry when
Between U2, flash memory average loss U3, flash memory consume U4, flash hardware read/write speed U5, flash memory erasing speed U6, flash memory erasing
Energy consumption U7, flash hardware read and write energy consumption U8With flash memory file system energy consumption U9, finally using order relation analytic approach to U1To U9Index
Carry out Comprehensive Assessment.Due to using the read-write for flash memory, load, erasing and the comprehensive evaluating of relevant 9 indexs of energy consumption,
Read or write speed, file system service ability and the Estimation of energy consumption being related in flash memory performance assessment are systematically analyzed, from file
Three system, hardware driving and energy consumption levels, propose corresponding index system, are integrated finally by order relation method to it
Evaluation and test calculates, so that test result is able to reflect the integration capability in terms of flash reading and writing, load, erasing and energy consumption.
It elaborates With reference to embodiment to the present invention.
Specific embodiment
Specific step is as follows for NOR and nand flash memory performance measuring and evaluating method of the present invention:
1, read or write speed U1Test.
Shown in read/write speed expression formula such as formula (1):
Speed=TotalSize/TotalTime (1)
Readwrite tests method and step is as follows:
1) initialization time variable and timer, and original document size and block size are set.
2) corresponding document is created, and sets uncached mode for file WriteMode.
3) logging timestamp is written and read file as unit of block size is arranged, and reads and writes size of data as text
Part size.
4) logging timestamp, and delete file.
5) increase block size, 2-5 is walked and is recycled, until block size is setting maximum value.
6) read or write speed of flash memory is calculated according to formula (1).
2, mounting time U2Test.
It repeatedly recycles carry and unloads corresponding document system, be averaged.Mounting time expression formula such as formula (2) institute
Show.
M_time=End_Time-Start_Time (2)
Mounting time testing procedure is as follows:
1) initialization time variable and timer and relevant parameter.
2) corresponding document is created according to parameter.
3) unloading of file system, timing start.
4) carry file system, timing terminate.
5) by 2-4 step circulation 10 times.
6) mounting time is calculated according to formula (2).
3, average loss U3Test.
Average loss test program is that write-in is written several times in flash memory space data, data by random into flash memory
File, creation file, random writing file operation composition are deleted, analog subscriber uses Flash-process.In program operation process, note
Record flash memory wipes the block number of block every time, finally exports the erasing times of the every erasing block of flash memory, for the quantitative determination performance, formulates
Two focus and index, abrasion equilibrium difference ε and erasing standard deviation S assess flash memory file system performance, such as public
Shown in formula (3) and (4).
ε=Emax-Emin (3)
Testing procedure is as follows:
1) relevant parameter is initialized, flash memory total capacity is obtained.
2) file is created at random, and data are written.
3) random erasure existing file.
4) existing file is randomly choosed, random data is written over.
5) step b-d is repeated, until write-in total amount of data is greater than ten times of flash memory total capacity.
6) erasing times of the every erasing block of output flash memory.
7) according to formula (3) and (4), average loss is calculated.
4, memory consumption U4Test.
The memory service condition that linux kernel passes through reading/proc/meminfo file dynamic acquisition system at this time.Test
When block size, after calling system calls function mount carry file system, comparison front and back memory headroom Expenditure Levels;
Memory consumption when test run, in data writing process, dynamic collection correspondence memory consumption information.
5, hardware read/write speed U5Test.
It is tested for hardware read or write speed, shown in hardware read/write speed expression formula such as formula (5):
It is tested for hardware layer read or write speed, using simplest sequence read-write mode.Flash memory is obtained by hardware layer API
After information, being taken respectively with page size, the sizes of page two and erasing block size is that basic read/write unit is tested.Circulation is to flash memory
Every erasing block is written and read velocity test, encounters bad block and skips.When hardware driving layer readwrite tests, pass through setting write-in block and reading
The size for entering block tests out data as input data and provides foundation for overall merit.
6, erasing speed U6Test.
It should be taken into account that flash memory is wiped by basic unit of block, therefore, the data volume size that when test wipes every time
For the size for wiping block.Shown in erasing speed expression formula such as formula (6):
Erasing speed test is recycled in hardware layer reads the every block message of flash memory, and API is called to carry out erasing operation to flash block,
Method and step is as follows:
1) timer and time variable are initialized.
2) logging timestamp.
3) it recycles erasing block every to flash memory and carries out erasing operation.
4) logging timestamp.
5) erasing speed is calculated according to formula (6).
7, energy consumption U is wiped7Test.
Erasing energy consumption refers to that flash memory will wipe the energy consumption generated when all memory cell datas are set to " 1 " on block.
After obtaining flash information by hardware layer API, erasing block every to flash memory is wiped respectively, and circulation erasing flash memory wipes every block number
According to the energy consumption of generation, average.The universal model of hardware energy consumption is as shown in formula (7):
Can be time-consuming by hardware testing, stablize since electric current inputs, therefore measuring voltage when hardware effort can be obtained by formula
It takes.Test erasing can be time-consuming, in linux kernel, passes through API acquisition flash information provided by MTD layer.Since flash memory is wiped
Energy consumption is tested by fluke scopmeter table hard due to needing to apply the energy consumption that voltage generates to flash memory source electrode
The true energy consumption of part.
8, hardware reads and writes energy consumption U8Test.
Read/write operation, the unit of setting read-write energy consumption are as follows: uj/page are carried out by basic unit of page for nand flash memory.
After obtaining flash information by hardware layer API, page size is obtained, recycles toward every page of write-in random data and records corresponding energy consumption
Value, until terminating to average.Similarly, to the flash memory of write-in random data, circulation reads every page data and records corresponding energy consumption
Value, averages.Test read-write can be time-consuming in linux kernel, passes through API acquisition flash information provided by MTD layer.Due to
Flash reading and writing energy consumption belongs to hardware energy consumption, therefore the present invention tests the true energy consumption of hardware by fluke scopmeter table.
9, file system energy consumption U9Test.
The basic unit of Energy Consumption Evaluation are as follows: uj.It is the interface function that user provides according to file system, to the energy that need to be measured
Consumption index is quantified.According to the VFS layer interface function that need to be realized, selection stat64, open, write, read, close,
Delete function is tested.Software energy consumption is to instruct the energy that system unit consumes in implementation procedure on a processor.Function
Energy consumption is usually especially all to call directly the instruction energy consumption summation called with son.It calls functionEnergy consumption formulas such as
(8) shown in.
For each function, circulation measurement energy consumption is taken repeatedly to average.For write and read function, when executing
Write-in and reading data are 0, in this way the energy consumption convenient for measurement file system instruction energy consumption rather than when flash hardware is read.It is logical
Cross energy consumption caused by formula (8) estimation flash memory file system.
10, flash memory overall merit.
It sorts according to flash memory characteristics to index significance level as follows:
U1> U3> U2> U4> U5> U5> U6> U7> U8> U9
U1To U9Weight coefficient wkIt is calculated by shown in calculation expression (9), (10)
wk-1=wkrk (10)
Wherein riCalculation formula such as expression formula (11) shown in, k=m, m-1 ..., 3,2, when m is larger, by order relation
Take rk=1.rkCan value be 1.0,1.2,1.4,1.6,1.8.
Comprehensive evaluation value calculation expression such as formula (12):
Z=ΣI=1xiwi (12)
Analyze to obtain following result:
According to formula (9) and formula (10), weight is calculated separately out:
U1=0.20758, U2=0.14828, U3=0.12356, U4=0.10297, U5=0.10297, U=0.08581,
U7=0.08581, U8=0.07151, U9=0.07151
Then, after quantifying to each index, flash memory overall merit counting system is obtained according to formula (13):
Claims (1)
1. a kind of NOR and nand flash memory performance measuring and evaluating method, it is characterised in that the following steps are included:
Step 1: test flash memory read or write speed U1, shown in calculation expression such as formula (1), wherein and Speed refers to read/write speed,
TotalSize refers to actual transmissions file size, and TotalTime refers to the actual measurement time;
Speed=TotalSize/TotalTime (1)
Initialization time variable and timer, and original document size and block size are set;Corresponding document is created, and file is write
Mode setting is uncached mode;Logging timestamp is written and read file as unit of block size is arranged, reads and writes number
It is file size according to size;Logging timestamp, and delete file;Increase block size, until block size is setting maximum value;
Step 2: test flash memory mounting time U2, using multiple circulation carry and corresponding document system is unloaded, is averaged;It hangs
It carries shown in temporal expression such as formula (2), wherein End_time is in file system mounted end point time, Start_Time
It is the file system mounted starting point time, M_time is the primary file system mounted time;
M_time=End_Time-Start_Time (2)
Step 3: test flash memory average loss U3, shown in calculation method such as formula (3), formula (4);Abrasion equilibrium difference ε is used for
Define the erasing times difference in flash memory between erasing times most muti-piece and the minimum block of erasing times;ε value is smaller, represents flash memory system
System longer life expectancy;Standard deviation S evaluates average loss performance indicator as another, and wherein n represents the erasable block number of flash memory,
EiEvery piece of erasing times are represented, E represents fifty-fifty erasing times, EmaxFor E maximum value EminFor E minimum value;
ε=Emax-Emin (3)
Step 4: test flash memory memory consumption U4, by read linux kernel under/proc/meminfo file dynamic acquisition at this time
The memory service condition of system;When test module size, after calling system calls function mount carry file system, before comparison
Memory headroom Expenditure Levels afterwards;
Step 5: test flash memory hardware read/write speed U5, shown in calculation expression such as formula (5), wherein Speed refers to read/write
Speed, TotalSize refer to actual transmissions file size, and TotalTime refers to the actual measurement time;
After obtaining flash information by hardware layer API, being taken respectively with page size, the sizes of page two and erasing block size is basic read
Unit is write to be tested;It recycles erasing block every to flash memory and is written and read velocity test, encounter bad block and skip;The read-write of hardware driving layer
When test, by the size of setting write-in block and reading block, data are tested out as input data and provide foundation for overall merit;
Step 6: test flash memory erasing speed U6, shown in calculation expression such as formula (6), wherein e_speed refers to erasing speed
Degree, Block_Size refer to that practical erasing block size, TotalTime refer to the actual measurement time;Erasing speed is tested in hardware
Layer circulation reads the every block message of flash memory, and API is called to carry out erasing operation to flash block;
Step 7: test flash memory wipes energy consumption U7, calculated with formula (7), wherein E indicates energy consumption, and P (t) indicates transient power;P table
Show average power consumption;I is average current;V is operating voltage, and T is runing time;
Step 8: test flash memory hardware reads and writes energy consumption U8, flash memory letter is obtained by API provided by MTD layers in linux kernel
Breath;Since flash reading and writing energy consumption belongs to hardware energy consumption, the true energy consumption of hardware is tested by fluke scopmeter table;
Step 9: test flash memory file system energy consumption U9, it is the interface function that user provides according to file system, to what need to be measured
Energy consumption index is quantified;It calls the energy consumption Eroutine of function by node f0, f1 ... .fn composition;Function fiSubfunction
For fj1, fj2 ... .fjm, the single clock cycle energy consumption of instruction is ei, then it calls functionEnergy consumption formulas such as (8) institute
Show;
For each function, circulation measurement energy consumption is taken repeatedly to average;
Step 10: using order relation analytic approach to U1To U9Index carries out Comprehensive Assessment, and comprehensive evaluation value calculation expression is (9)
It is shown:
Z=∑I=1xiwi (9)
Wherein xiFor calculative each index, weight coefficient wkCalculation method such as formula (11), shown in (12), evaluation refers to
K=m in relative importance formula, m-1 ..., 3,2 take r by order relation when m is larger between markk=1;rkValue be
1.0,1.2,1.4,1.6,1.8;
wk-1=wkrk (12)。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611230307.6A CN106683709B (en) | 2016-12-28 | 2016-12-28 | NOR and nand flash memory performance measuring and evaluating method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611230307.6A CN106683709B (en) | 2016-12-28 | 2016-12-28 | NOR and nand flash memory performance measuring and evaluating method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106683709A CN106683709A (en) | 2017-05-17 |
CN106683709B true CN106683709B (en) | 2019-08-09 |
Family
ID=58871747
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611230307.6A Active CN106683709B (en) | 2016-12-28 | 2016-12-28 | NOR and nand flash memory performance measuring and evaluating method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106683709B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108269606B (en) * | 2018-01-10 | 2020-07-14 | 重庆金山医疗器械有限公司 | Method and system for testing storage parameters of eMMC chip |
CN108491740A (en) * | 2018-02-01 | 2018-09-04 | 珠海全志科技股份有限公司 | A kind of TF card performance method of real-time |
CN108345752B (en) * | 2018-02-24 | 2022-02-11 | 北京芯可鉴科技有限公司 | Method for evaluating life characteristic of wafer-level nonvolatile memory |
CN108564985B (en) * | 2018-03-16 | 2021-06-25 | 广州视源电子科技股份有限公司 | EMMC (embedded multi-media card) testing method and device, mobile terminal and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101067968A (en) * | 2007-04-29 | 2007-11-07 | 北京中星微电子有限公司 | Apparatus and method for adaptive controlling flash storage interface reading and writing speed |
CN103324560A (en) * | 2013-06-28 | 2013-09-25 | 深圳市江波龙电子有限公司 | Testing method and device of flash memory |
CN104810062A (en) * | 2015-05-12 | 2015-07-29 | 东南大学 | PUF (polyurethane foam) feature test method and device of SRAM (static random access memory) chip |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150255176A1 (en) * | 2014-03-10 | 2015-09-10 | Advantest Corporation | Memory test ecc auto-correction of failing data |
US9558064B2 (en) * | 2015-01-28 | 2017-01-31 | Micron Technology, Inc. | Estimating an error rate associated with memory |
-
2016
- 2016-12-28 CN CN201611230307.6A patent/CN106683709B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101067968A (en) * | 2007-04-29 | 2007-11-07 | 北京中星微电子有限公司 | Apparatus and method for adaptive controlling flash storage interface reading and writing speed |
CN100458977C (en) * | 2007-04-29 | 2009-02-04 | 北京中星微电子有限公司 | Apparatus and method for adaptive controlling flash storage interface reading and writing speed |
CN103324560A (en) * | 2013-06-28 | 2013-09-25 | 深圳市江波龙电子有限公司 | Testing method and device of flash memory |
CN103324560B (en) * | 2013-06-28 | 2016-05-25 | 深圳市江波龙电子有限公司 | A kind of flash memories method of testing and device |
CN104810062A (en) * | 2015-05-12 | 2015-07-29 | 东南大学 | PUF (polyurethane foam) feature test method and device of SRAM (static random access memory) chip |
Non-Patent Citations (1)
Title |
---|
NOR型FLASH存储器测试;王征宇;《电子与封装》;20160331;15-17,30 * |
Also Published As
Publication number | Publication date |
---|---|
CN106683709A (en) | 2017-05-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106683709B (en) | NOR and nand flash memory performance measuring and evaluating method | |
US8806106B2 (en) | Estimating wear of non-volatile, solid state memory | |
Meza et al. | A large-scale study of flash memory failures in the field | |
CN104092835B (en) | A kind of mobile terminal power consumption method for detecting abnormality and device | |
CN103559115A (en) | SSD intelligent monitoring system based on SMART | |
US20120084028A1 (en) | Framework and Methodology for a Real-Time Fine-Grained Power Profiling with Integrated Modeling | |
Hu et al. | Lightweight energy consumption analysis and prediction for android applications | |
CN107068199B (en) | Performance test method of SSD (solid State disk) | |
CN114355094B (en) | Product reliability weak link comprehensive evaluation method and device based on multi-source information | |
CN115171771A (en) | Solid state disk testing method, device, equipment and storage medium | |
CN117151329A (en) | Carbon emission strength prediction method, device, equipment and storage medium | |
CN110837477B (en) | Storage system loss balancing method and device based on life prediction | |
US20110296249A1 (en) | Selecting a configuration for an application | |
Ghaleb | Software energy measurement at different levels of granularity | |
US20090271643A1 (en) | Real-time inference of power efficiency metrics for a computer system | |
Procaccianti et al. | Energy efficiency in the ict-profiling power consumption in desktop computer systems | |
CN109582527A (en) | Storage server and its solid-state disk service life monitoring method | |
CN116682479A (en) | Method and system for testing enterprise-level solid state disk time delay index | |
CN106610873A (en) | Method for predicting application energy consumption on Android equipment | |
Wu et al. | A large-scale study of I/O workload’s impact on disk failure | |
Yan et al. | Accurate and low-overhead process-level energy estimation for modern hard disk drives | |
CN109308162B (en) | Flash memory optimization device, optimization method and equipment | |
CN110716833B (en) | Method for measuring NAND FLASH write quantity caused by single entry of SSD into PS4 state | |
CN112446490A (en) | Network training data set caching method, device, equipment and storage medium | |
Seegert et al. | Energy and resource comparison of current applications with a focus on statistical analyses and evaluations using the example of MATLAB and R |
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 |