CN106469121A - A kind of method and device of Memory Allocation - Google Patents
A kind of method and device of Memory Allocation Download PDFInfo
- Publication number
- CN106469121A CN106469121A CN201610813881.8A CN201610813881A CN106469121A CN 106469121 A CN106469121 A CN 106469121A CN 201610813881 A CN201610813881 A CN 201610813881A CN 106469121 A CN106469121 A CN 106469121A
- Authority
- CN
- China
- Prior art keywords
- slab
- class
- influence
- source
- pointer
- 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
- G06F12/0253—Garbage collection, i.e. reclamation of unreferenced memory
-
- 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
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0893—Caches characterised by their organisation or structure
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Memory System (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of method and device of Memory Allocation, by obtaining the information of each slab class;Calculate the factor of influence of each slab class according to the minimum influence selecting predictors algorithm pre-setting and described information;The minimum slab class of factor of influence is chosen for source slab class, the most slab of quantity is arranged as purpose slab by swapping out in a period of time;One of described source slab class slab internal memory is emptied and after recovery, is re-assigned to described purpose slab, minimum influence selecting predictors algorithm is proposed, consider the feature of the data storage in the case of compression, the appropriate source slab class of reasonable selection and purpose slab are reclaimed and reassign;The Memory Allocation mechanism of adjustment Memcached, has divided region effectively according to slab storage state is empty, full etc., the slab idle from end starts to reclaim when needing internal memory.
Description
Technical field
The present embodiments relate to the technical field of Memory Allocation, more particularly, to a kind of method and device of Memory Allocation.
Background technology
Memcached, according to the size of request object, Memory Allocation is given corresponding slab class.Once in Memcached
All free memories all dispense, and when leading to internal memory to reassign because request distribution changes, problem just produces
, referred to as calcification problem (Calcification), that is, in Memcached, the number ratio of all slab classes has been fixed, no
Method dynamically adjusts according to the external request of change.If Memcached need storage object size distribution situation not with
Time changes, then calcification problem does not interfere with the performance of Memcached, because the base having been carried out in Memcached
Originally the strategy that swaps out can ensure to store heat target well.If however, the size distribution situation of storage object is at any time in change,
Calcification problem is made to have considerable influence to the hit rate of Memcached.
Solve the problems, such as Memcached calcification direct method allow Memcached will dispense in deposit into
Row reclaims and is reassigned to the slab class of needs.Data compression exacerbates the calcification problem of Memcached.In Memcached
When log-on data compression function, the space that not only can stay in the slab apoplexy due to endogenous wind being compressed cannot return to Memcached,
And larger pressure can be caused to the slab class being stored in after compression.In the Memcached that with the addition of data compression function, to the greatest extent
Pipe data compression brings different performance boost effects to Memcached, but the calcification also exacerbating Memcached is asked
Topic.
Content of the invention
The purpose of the embodiment of the present invention is to propose a kind of method and device of Memory Allocation it is intended to solve to the addition of data
Calcification problem in the Memcached of compression function.
For reaching this purpose, the embodiment of the present invention employs the following technical solutions:
In a first aspect, a kind of method of Memory Allocation, methods described includes:
Obtain the information of each slab class, described information includes using number of objects, number of requests, object sum and changes
Go out quantity;
Calculate the factor of influence of each slab class according to the minimum influence selecting predictors algorithm pre-setting and described information;
The minimum slab class of factor of influence is chosen for source slab class, the most slab of quantity sets by swapping out in a period of time
It is set to purpose slab;
One of described source slab class slab internal memory is emptied and after recovery, is re-assigned to described purpose slab.
Preferably, the minimum influence selecting predictors algorithm that described basis pre-sets and described information calculate each slab class
Factor of influence, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, described request table
Show described specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents described specific slab apoplexy due to endogenous wind
Object sum.
Preferably, described empty one of described source slab class slab internal memory is re-assigned to described purpose after recovery
Slab, including:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and will be heavy after the Memory recycle after emptying
Newly it is assigned to described purpose slab.
Preferably, methods described also includes:
The space of use and untapped space in Memcached is treated with a certain discrimination, by the regional centralized of use in
Slab front end, by untapped regional centralized in slab rear end.
Preferably, methods described also includes:
It is provided with three pointers of the scope for specifying this slab, institute in the front end of each slab of each slab apoplexy due to endogenous wind
State three pointers include for identify first storage object in slab pointer slab_start, be used for identifying in slab last
The pointer slab_end of one storage object, the pointer cur_free for identifying currently available storage object in slab.
Second aspect, a kind of device of Memory Allocation is it is characterised in that described device includes:
Acquisition module, for obtaining the information of each slab class, described information is included using number of objects, number of request
Amount, object sum and the quantity that swaps out;
Selecting module, for calculating each slab according to the minimum influence selecting predictors algorithm pre-setting and described information
The factor of influence of class;
Choose module, be chosen for source slab class for the slab class that factor of influence is minimum, by the number that swaps out in a period of time
Measure most slab to arrange as purpose slab;
Distribute module, is re-assigned to described purpose for emptying one of described source slab class slab internal memory after recovery
slab.
Preferably, described selecting module, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, described request table
Show described specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents described specific slab apoplexy due to endogenous wind
Object sum.
Preferably, described distribute module, is used for:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and will be heavy after the Memory recycle after emptying
Newly it is assigned to described purpose slab.
Preferably, described device also includes:
Adjusting module, for treating the space of use and untapped space in Memcached with a certain discrimination, has been used
Regional centralized in slab front end, by untapped regional centralized in slab rear end.
Preferably, described device also includes:
Setup module, is provided for specifying three of the scope of this slab in the front end of each slab of each slab apoplexy due to endogenous wind
Pointer, described three pointers include for identify first storage object in slab pointer slab_start, be used for identifying
The pointer slab_end of last storage object, the pointer cur_ for identifying currently available storage object in slab in slab
free.
The embodiment of the present invention provides a kind of method and device of Memory Allocation, and methods described is passed through to obtain each slab class
Information, described information is included using number of objects, number of requests, object sum and the quantity that swaps out;According to pre-setting
Little factor of influence selection algorithm and described information calculate the factor of influence of each slab class;By slab class choosing minimum for factor of influence
It is taken as source slab class, the most slab of quantity is arranged as purpose slab by swapping out in described a period of time;By described source slab class
One of slab internal memory empty recovery after be re-assigned to described purpose slab, minimum influence selecting predictors algorithm is proposed, comprehensive
The feature of the data storage in the case of considering to compress, the appropriate source slab class of reasonable selection and purpose slab are reclaimed and are divided again
Join;The Memory Allocation mechanism of adjustment Memcached, has effectively divided region according to slab storage state is empty, full etc., has worked as needs
When internal memory, the slab idle from end starts to reclaim.
Brief description
Fig. 1 is a kind of schematic flow sheet of the method for Memory Allocation provided in an embodiment of the present invention;
Fig. 2 is a kind of another schematic flow sheet of the method for Memory Allocation provided in an embodiment of the present invention;
Fig. 3 is a kind of method schematic diagram of space distribution provided in an embodiment of the present invention;
Fig. 4 is the method schematic diagram of another kind space distribution provided in an embodiment of the present invention;
Fig. 5 is a kind of pointer structure schematic diagram of Memory Allocation provided in an embodiment of the present invention;
Fig. 6 be a kind of Memory Allocation provided in an embodiment of the present invention before and after calcification problem contrast schematic diagram;
Fig. 7 is a kind of high-level schematic functional block diagram of the device of Memory Allocation provided in an embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples the embodiment of the present invention is described in further detail.It is understood that this
The described specific embodiment in place is used only for explaining the embodiment of the present invention, rather than the restriction to the embodiment of the present invention.In addition also
It should be noted that for the ease of description, illustrate only the part related to the embodiment of the present invention rather than entire infrastructure in accompanying drawing.
With reference to Fig. 1, Fig. 1 is a kind of schematic flow sheet of the method for Memory Allocation provided in an embodiment of the present invention.
In FIG, the method for described Memory Allocation includes:
Step 101, obtains the information of each slab class, and described information is included using number of objects, number of requests, object
Sum and the quantity that swaps out;
Specifically, due to having increased data compression function in Memcached newly, the data of storage object in Memcached
Feature changes.From the angle compressing quantity, number of requests and object sum it is proposed that minimum influence selecting predictors are calculated
Method.Factor of influence (Impact factor, IF) is that comprehensive compression quantity, number of requests and object sum reclaimed to slab
The index that the impact of journey obtains.
Step 102, calculates each slab class according to the minimum influence selecting predictors algorithm pre-setting and described information
Factor of influence;
Preferably, the minimum influence selecting predictors algorithm that described basis pre-sets and described information calculate each slab class
Factor of influence, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, described request table
Show described specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents described specific slab apoplexy due to endogenous wind
Object sum.
Specifically, factor of influence is less, illustrates that this slab is more suitable as source slab class and provides internal memory.
Step 103, the minimum slab class of factor of influence is chosen for source slab class, quantity is most by swapping out in a period of time
Slab arrange as purpose slab;
Specifically, collect each slab class using number of objects, number of requests, object sum and the number that swaps out
Amount.Factor of influence is calculated to each slab class, therefrom selects the minimum slab class of a factor of influence as source slab class.From
Slab apoplexy due to endogenous wind chooses the slab class of the quantity at most (frequent) that swaps out as purpose slab.
Step 104, one of described source slab class slab internal memory is emptied and is re-assigned to described purpose slab after recovery.
Preferably, described empty one of described source slab class slab internal memory is re-assigned to described purpose after recovery
Slab, including:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and will be heavy after the Memory recycle after emptying
Newly it is assigned to described purpose slab.
Wherein, memory headroom is the internal memory of 1M size for fixed size.
Specifically, with reference to Fig. 2, Fig. 2 is that a kind of another flow process of the method for Memory Allocation provided in an embodiment of the present invention is shown
It is intended to.
Step 201, if judging, slab needs to reclaim and reassigns, and selects source slab class;
Step 202, selects purpose slab;
Step 203, judges whether to slab and reclaims and reassign;
Step 204, if judging, slab does not need to reclaim and reassigns, and judges whether the slab class selected is effective;
Step 205, if the slab class selected effectively, carries out slab recovery and reassigns work.
Preferably, methods described also includes:
The space of use and untapped space in Memcached is treated with a certain discrimination, by the regional centralized of use in
Slab front end, by untapped regional centralized in slab rear end.
Specifically, the space that Memcached in Fig. 3 has used is distinguished with idle space and treat, will used area
Domain concentrates on the front end of slab and clear area concentrates on the rear end of slab.When there is calcification problem, preferentially to rear
End regions carry out Memory recycle and reassignment.
Fig. 4 is the slab allotter schematic diagram of Memcached after modification, not only occupied area is concentrated on front end,
And pointer is more regular between each slab, be conducive to carrying out the chain rupture operation of chained list when reclaiming internal memory.?
When Memcached reclaims internal memory, the object of the slab choosing storage can be emptied, cutting again is the little of other sizes
Block.The operation emptying object can affect the hit rate of Memcached, to a certain extent using the slab allotter shown in Fig. 4
Structure can reduce this impact.
Preferably, methods described also includes:
It is provided with three pointers of the scope for specifying this slab, institute in the front end of each slab of each slab apoplexy due to endogenous wind
State three pointers include for identify first storage object in slab pointer slab_start, be used for identifying in slab last
The pointer slab_end of one storage object, the pointer cur_free for identifying currently available storage object in slab.
Specifically, with reference to shown in Fig. 5, Fig. 5 is that a kind of pointer structure of Memory Allocation provided in an embodiment of the present invention is illustrated
Figure.Fig. 5 is specific amending method, except adding 3 pointers as depicted in slab structure in addition it is also necessary to divide in internal memory
Join and in function, add corresponding process code.
With reference to Fig. 6, Fig. 6 be a kind of Memory Allocation provided in an embodiment of the present invention before and after calcification problem contrast schematic diagram.
Dash-dotted gray line in Fig. 6 is experiment effect, therefrom can see and be transitioned into request type B from request type A, under calcification problem
The improvement of hit rate has clear improvement, and this scheme effectively improves what Memcached was led to by calcification under compression
The problem that hit rate declines.Grey curves are the expectancy curves of Memcached being affected by calcification in the case of compression.
The embodiment of the present invention provides a kind of method of Memory Allocation, and methods described is passed through to obtain the information of each slab class,
Described information is included using number of objects, number of requests, object sum and the quantity that swaps out;According to the minimum influence pre-setting
Selecting predictors algorithm and described information calculate the factor of influence of each slab class;The minimum slab class of factor of influence is chosen for source
Slab class, by swapping out in described a period of time, the most slab of quantity is arranged as purpose slab;By described source slab apoplexy due to endogenous wind one
Individual slab internal memory is re-assigned to described purpose slab after emptying recovery, propose minimum influence selecting predictors algorithm, consider pressure
The feature of the data storage in the case of contracting, the appropriate source slab class of reasonable selection and purpose slab are reclaimed and are reassigned;Adjust
The Memory Allocation mechanism of whole Memcached, has divided region effectively according to slab storage state is empty, full etc., when needing internal memory
When the slab idle from end start to reclaim.
With reference to Fig. 7, Fig. 7 is a kind of high-level schematic functional block diagram of the device of Memory Allocation provided in an embodiment of the present invention.
In Fig. 7, the device of described Memory Allocation includes:
Acquisition module 701, for obtaining the information of each slab class, described information is included using number of objects, request
Quantity, object sum and the quantity that swaps out;
Selecting module 702, for calculating each according to the minimum influence selecting predictors algorithm pre-setting and described information
The factor of influence of slab class;
Preferably, described selecting module 702, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, described request table
Show described specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents described specific slab apoplexy due to endogenous wind
Object sum.
Choose module 703, be chosen for source slab class for the slab class that factor of influence is minimum, swap out in a period of time
The most slab of quantity is arranged as purpose slab;
Distribute module 704, for one of described source slab class slab internal memory is emptied be re-assigned to after recovery described
Purpose slab.
Preferably, described distribute module 704, is used for:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and will be heavy after the Memory recycle after emptying
Newly it is assigned to described purpose slab.
Preferably, described device also includes:
Adjusting module, for treating the space of use and untapped space in Memcached with a certain discrimination, has been used
Regional centralized in slab front end, by untapped regional centralized in slab rear end.
Preferably, described device also includes:
Setup module, the front end for each slab in each slab apoplexy due to endogenous wind is provided for specifying the scope of this slab
Three pointers, described three pointers include for identify first storage object in slab pointer slab_start, be used for marking
Know the pointer slab_end of last storage object in slab, the pointer for identifying currently available storage object in slab
cur_free.
The embodiment of the present invention provides a kind of device of Memory Allocation, by obtaining the information of each slab class, described information
Including using number of objects, number of requests, object sum and the quantity that swaps out;According to the minimum influence selecting predictors pre-setting
Algorithm and described information calculate the factor of influence of each slab class;The minimum slab class of factor of influence is chosen for source slab class,
By swapping out in described a period of time, the most slab of quantity is arranged as purpose slab;By in one of described source slab class slab
Deposit and after emptying recovery, be re-assigned to described purpose slab, propose minimum influence selecting predictors algorithm, in the case of considering compression
Data storage feature, the appropriate source slab class of reasonable selection and purpose slab are reclaimed and are reassigned;Adjustment
The Memory Allocation mechanism of Memcached, has divided region effectively according to slab storage state is empty, full etc., when needing internal memory
When the slab idle from end start to reclaim.
Describe the know-why of the embodiment of the present invention above in association with specific embodiment.These descriptions are intended merely to explain this
The principle of inventive embodiments, and the restriction to embodiment of the present invention protection domain can not be construed to by any way.Based on herein
Explanation, those skilled in the art do not need to pay performing creative labour and can associate the other concrete of the embodiment of the present invention
Embodiment, these modes fall within the protection domain of the embodiment of the present invention.
Claims (10)
1. a kind of method of Memory Allocation is it is characterised in that methods described includes:
Obtain the information of each slab class, described information is included using number of objects, number of requests, object sum and the number that swaps out
Amount;
Calculate the factor of influence of each slab class according to the minimum influence selecting predictors algorithm pre-setting and described information;
The minimum slab class of factor of influence is chosen for source slab class, the most slab of quantity is set to by swapping out in a period of time
Purpose slab;
One of described source slab class slab internal memory is emptied and after recovery, is re-assigned to described purpose slab.
2. method according to claim 1 is it is characterised in that the minimum influence selecting predictors that described basis pre-sets are calculated
Method and described information calculate the factor of influence of each slab class, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, and described request represents institute
State specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents the object of described specific slab apoplexy due to endogenous wind
Sum.
3. method according to claim 1 is it is characterised in that described will be clear for one of described source slab class slab internal memory
Backlash is re-assigned to described purpose slab after receiving, including:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and will reassign after the Memory recycle after emptying
To described purpose slab.
4. the method according to claims 1 to 3 any one is it is characterised in that methods described also includes:
The space of use and untapped space in Memcached is treated with a certain discrimination, by the regional centralized of use in slab
Front end, by untapped regional centralized in slab rear end.
5. the method according to claims 1 to 3 any one is it is characterised in that methods described also includes:
It is provided with three pointers of the scope for specifying this slab in the front end of each slab of each slab apoplexy due to endogenous wind, described three
Individual pointer include for identify first storage object in slab pointer slab_start, be used for identifying last in slab
The pointer slab_end of storage object, the pointer cur_free for identifying currently available storage object in slab.
6. a kind of device of Memory Allocation is it is characterised in that described device includes:
Acquisition module, for obtaining the information of each slab class, described information is included using number of objects, number of requests, right
As sum and the quantity that swaps out;
Selecting module, for calculating each slab class according to the minimum influence selecting predictors algorithm pre-setting and described information
Factor of influence;
Choose module, be chosen for source slab class for the slab class that factor of influence is minimum, quantity will be swapped out in a period of time
Many slab are arranged as purpose slab;
Distribute module, is re-assigned to described purpose for emptying one of described source slab class slab internal memory after recovery
slab.
7. device according to claim 6 is it is characterised in that described selecting module, including:
Wherein, described used_chunks represents described specific slab apoplexy due to endogenous wind using number of objects, and described request represents institute
State specific slab apoplexy due to endogenous wind number of requests interior for a period of time, described total_chunks represents the object of described specific slab apoplexy due to endogenous wind
Sum.
8. device according to claim 6, it is characterised in that described distribute module, is used for:
Obtain the maximum slab of described source slab class free space quantity;
Empty the maximum slab internal memory of described source slab class free space quantity, and again divide after the Memory recycle after emptying
It is fitted on described purpose slab.
9. the device according to claim 6 to 8 any one is it is characterised in that described device also includes:
Adjusting module, for treating the space of use and untapped space in Memcached with a certain discrimination, by the area of use
Domain concentrates on slab front end, by untapped regional centralized in slab rear end.
10. the device according to claim 6 to 8 any one is it is characterised in that described device also includes:
Setup module, the front end for each slab in each slab apoplexy due to endogenous wind is provided for specifying three of the scope of this slab
Pointer, described three pointers include for identify first storage object in slab pointer slab_start, be used for identifying
The pointer slab_end of last storage object, the pointer cur_ for identifying currently available storage object in slab in slab
free.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610813881.8A CN106469121B (en) | 2016-09-09 | 2016-09-09 | Memory allocation method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610813881.8A CN106469121B (en) | 2016-09-09 | 2016-09-09 | Memory allocation method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106469121A true CN106469121A (en) | 2017-03-01 |
CN106469121B CN106469121B (en) | 2019-12-13 |
Family
ID=58230412
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610813881.8A Active CN106469121B (en) | 2016-09-09 | 2016-09-09 | Memory allocation method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106469121B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110209594A (en) * | 2018-02-28 | 2019-09-06 | 贵州白山云科技股份有限公司 | A kind of EMS memory management process and device suitable for slab structure |
CN113434285A (en) * | 2020-03-23 | 2021-09-24 | 阿里巴巴集团控股有限公司 | Key value cache system-based memory management method and device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101676883A (en) * | 2008-09-17 | 2010-03-24 | 英业达股份有限公司 | Memory allocation and management method |
US20160077980A1 (en) * | 2014-09-13 | 2016-03-17 | Agoraic, Inc. | Methods, devices and systems for caching data items |
-
2016
- 2016-09-09 CN CN201610813881.8A patent/CN106469121B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101676883A (en) * | 2008-09-17 | 2010-03-24 | 英业达股份有限公司 | Memory allocation and management method |
US20160077980A1 (en) * | 2014-09-13 | 2016-03-17 | Agoraic, Inc. | Methods, devices and systems for caching data items |
Non-Patent Citations (2)
Title |
---|
DAMIANO CARRA ET AL.: "Memory Partitioning in Memcached: An Experimental Performance Ayalysis", 《IEEE ICC 2014-COMMUNICATION QOS,RELIABILITY AND MODELING SYSMPOSIUM》 * |
LINUX系列教材编写组: "《Linux操作系统分析与实践》", 30 November 2008 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110209594A (en) * | 2018-02-28 | 2019-09-06 | 贵州白山云科技股份有限公司 | A kind of EMS memory management process and device suitable for slab structure |
CN110209594B (en) * | 2018-02-28 | 2020-01-24 | 贵州白山云科技股份有限公司 | Memory management method and device suitable for slab structure |
CN113434285A (en) * | 2020-03-23 | 2021-09-24 | 阿里巴巴集团控股有限公司 | Key value cache system-based memory management method and device |
WO2021190468A1 (en) * | 2020-03-23 | 2021-09-30 | 阿里巴巴集团控股有限公司 | Memory management method and apparatus based on key-value cache system |
CN113434285B (en) * | 2020-03-23 | 2024-03-26 | 阿里巴巴集团控股有限公司 | Memory management method and device based on key value cache system |
Also Published As
Publication number | Publication date |
---|---|
CN106469121B (en) | 2019-12-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7908454B2 (en) | Application-specific heap management | |
CN109815162A (en) | EMS memory management process, device, mobile terminal and storage medium | |
CN102968507B (en) | data query method based on cache table | |
CN1561486A (en) | Dynamic and automatic memory management | |
CN106502918B (en) | A kind of scheduling memory method and device | |
CN104426953A (en) | Method and apparatus for distributing calculation resources | |
CN105868028A (en) | Method and device for sharing data between processes, and terminal | |
CN101373445B (en) | Method and apparatus for scheduling memory | |
CN105917319A (en) | Memory unit and method | |
CN107894922B (en) | RAM resource allocation method | |
CN105389210A (en) | Storage space management method and device | |
CN106469121A (en) | A kind of method and device of Memory Allocation | |
CN110502321A (en) | A kind of resource regulating method and system | |
CN106445835A (en) | Memory allocation method and apparatus | |
CN104966352A (en) | Method and device for depositing and withdrawing money by reasonably utilizing bank note boxes | |
CN107515788A (en) | A kind of method and device of Memory Allocation | |
CN106980680A (en) | Date storage method and storage device | |
CN104111896A (en) | Virtual memory management method and virtual memory management device for mass data processing | |
CN102855137B (en) | Methods and procedures design system for the programming of automation component | |
CN109376003A (en) | A kind of GPU video memory management method of chain structure | |
CN106294192A (en) | Memory allocation method, Memory Allocation device and server | |
CN106356088A (en) | Data processing method and device | |
CN103488577A (en) | Method and device of memory allocation and batch recovery for user applications based on use numbering | |
CN105023353A (en) | Method and apparatus for depositing and withdrawing money with reasonable use of money cassettes | |
CN111258924B (en) | Mapping method based on satellite-borne solid-state storage system self-adaptive flash translation layer |
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 |