CN106339331B - A kind of data buffer storage stratification scaling method based on user activity - Google Patents
A kind of data buffer storage stratification scaling method based on user activity Download PDFInfo
- Publication number
- CN106339331B CN106339331B CN201610792010.2A CN201610792010A CN106339331B CN 106339331 B CN106339331 B CN 106339331B CN 201610792010 A CN201610792010 A CN 201610792010A CN 106339331 B CN106339331 B CN 106339331B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- early warning
- activity
- liveness
- 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
- 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
- G06F12/12—Replacement control
- G06F12/121—Replacement control using replacement algorithms
Abstract
The invention discloses a kind of data buffer storage stratification scaling method based on user activity, comprising the following steps: S1, initialization caching any active ues；S2, the data between data buffer storage layer and database are synchronized；S3, the activity of the user is recalculated, third level early warning value is set；S4, setting first order early warning value and second level early warning value；S5, the activity of the user situation of change in cache layer is judged according to first order early warning value, second level early warning value, third level early warning value；S6, any active ues caching replacement is carried out；S7, the operation that step S2~S6 is repeated according to the period of system manager's setting, until system is out of service.The present invention is by calculating the activity of the user, the data of the high user of liveness are stored in data buffer storage layer, response can be directly directly obtained from cache layer when the high user's access of the liveness and need not access server, the time of significantly less response user request, improve the experience of user.
Description
Technical field
The invention belongs to Data cache technology field, in particular to a kind of data buffer storage stratification based on user activity changes
Algorithm.
Background technique
In database and server architecture, in order to improve the read or write speed of data, often in Web server and database
Increase a data cache layer between server, function is in a copy for establishing frequently accessed data, in this way original
Carry out the operation that a queried access request needs to request a database originally, can first have been searched whether now in cache layer
The copy of searched data, if having preserved copy, there is no need to requested databases, directly read in cache layer, in this way
The access speed of data is just greatly improved.However this mode is delayed using data of the lru algorithm to recent visit
It deposits, seems why unobvious for having the performance boost of the social network-i i-platform of particular traffic requirements just.
It is exactly the activity of the user in social network sites, there is critically important user's evaluation index.User is in social activity
It is all the more active in website, illustrate that the user is bigger for the contribution of this social platform, payes attention to this social networks further, recognizes
Can, be interested in continuing with using the platform, according to the active degree of user, when platform is released it is new movable or preferential when also can be excellent
The high user of these liveness is first considered, for low liveness, even almost without the user for using account, it is contemplated that will
It is possible to, we can not directly remove it.Therefore the thought that we can use this cache layer will be used actively
The data copy at family is stored in data buffer storage layer, will get a promotion for the data access speed of any active ues in this way and for
Sluggish user will not directly remove, if they continue to use platform back, at most namely at the beginning when number
Can be slow according to access, compared to most of any active ues, user experience is had little effect.
It, can not be by the technology of progress cache layer displacement after assessing liveness, in this way in present technology
So that cache layer is the method using common LRU replacement, the property for not being directed to business makes improvement for meeting.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of preservations of the data of user that liveness is high
In data buffer storage layer, response can be directly directly obtained from cache layer when the high user's access of the liveness and need not be visited
Ask server, the significantly less data buffer storage stratification scaling method based on user activity of the time of response user's request.
The purpose of the present invention is achieved through the following technical solutions: a kind of data buffer storage layer based on user activity
Replacement algorithm, comprising the following steps:
S1, initialization caching any active ues: calculate the activity of the user, by liveness it is forward meet buffer memory capacity size
User data set be moved into cache layer, the caching any active ues as initialization；Simultaneously to the key-value pair of the user of immigration
Information adds a value of statistical indicant for identifying the active state of the user, is initially both configured to any active ues；
S2, the data between data buffer storage layer and database are synchronized；
S3, when reach set cycle time when, recalculate the activity of the user, while the average work of counting user
Jerk is as liveness threshold value；Compare the size of the capacity of liveness threshold value and cache layer, if liveness threshold value is greater than caching
The capacity of layer is then using liveness threshold value as third level early warning value；Otherwise using cache layer capacity as third level early warning value；
S4, first order early warning value and second level early warning value are arranged according to business demand；
S5, the first order early warning value counted according to step S3 and S4, second level early warning value, third level early warning value judge
The activity of the user situation of change in cache layer；
S6, carry out any active ues caching replacement: background server repeat step S5, to the user data in cache layer into
Row statistics obtains one and needs the user data list that removes, interim moves the user data from data buffer storage layer in this week
It removes；After removing user, the user data in cache layer is not present in the new liveness position p in the top and moves into caching
Layer, the key-value pair information for the user that replaces the user data of removal, and will move into add a value of statistical indicant for identifying the user
Active state；Wherein, p is the number of the user data removed；
S7, the operation that step S2~S6 is repeated according to the period of system manager's setting, until system is out of service.
Further, it includes following sub-step that step S1 user activity, which calculates:
S11, the factor for choosing n influence liveness, choose m sample, form the primary data matrix of evaluation system:
That is: X={ xijM × n, 0≤i≤m, 0≤j≤n；
Wherein, xijIndicate the index value of i-th of user's jth item liveness factor；
S12, data standard processing is carried out:
Wherein,For the average value of the index value of jth item liveness factor,SjIt is active for jth item
The standard deviation of the index value of degree factor,
S13, the specific gravity Y for calculating i-th of user's index value of jth item liveness factorij:
S14, the specific gravity matrix for establishing data:
Y={ Yij}m*2；
S15, the information entropy for calculating jth item liveness factor:
Wherein, K is constant,
S16, the information utility value for calculating jth item liveness factor:
dj=1-Ej；
The weight of jth item liveness factor are as follows:
S17, the activity of the user is calculated:
Further, the processing side in step S5 under the activity of the user situation of change and different liveness situation of change
Method includes following several:
If the activity of the user is greater than first order early warning value in S51, data buffer storage layer, check that user's enlivens shape
State；If it is any active ues, then any operation is not done, directly reservation user data is in cache layer；If user is not active
User, then the state for modifying user is any active ues；
If the activity of the user is less than first order early warning value and is greater than second level early warning value in S52, data buffer storage layer,
The active state of user is set to remind any active ues, while sending message, such as sends short message and the mail reminder user；
If the activity of the user is less than second level early warning value and is greater than third level early warning value in S53, data buffer storage layer,
Whether the active state for continuing to judge the user in cache layer is active state, if it is active state, by the work of the user
Jump state is set as reminding any active ues, while sending the prompting message user；If it is any active ues are reminded, by user's
Active state is set as warning any active ues, while sending message and alerting the user, alerts the user and returns as early as possible and puts down
Platform is active；
If the activity of the user is less than third level early warning value in S54, data buffer storage layer, the user in cache layer is judged
Whether it is active state, if it is active state, then sets the active state of the user to reminding any active ues, and send and disappear
Breath reminds user；If it is any active ues are reminded, then the active state of the user is set for warning any active ues, concurrently
Message is sent to alert the user；If the active state of the user is warning any active ues in cache layer, by the user
The user data removed as needs from data buffer storage layer.
The beneficial effects of the present invention are:
1, the data of the high user of liveness are stored in data buffer storage layer by calculating the activity of the user by the present invention
In, response can be directly directly obtained from cache layer when the high user's access of the liveness and need not access server, greatly
The big less time of response user's request, improve the experience of user；
2, present invention employs active state of the three-level early warning to user to be monitored, if the active state of user cannot
It is maintained at active state, then can send short message to it or mail is reminded；It can will be right if liveness is still in decline
It is alerted, user's long period no activity is informed by mail and short message；If promoted again without by liveness
Come up, then can be removed from data buffer storage layer using user as inactive user；And user is recalculated by loop cycle and is lived
Jerk, making the user in data buffer storage layer forever is more enlivened, using more frequent user, less to user demand with this
Response time, achieve the effect that increase users' satisfaction degree.
Detailed description of the invention
Fig. 1 is data buffer storage stratification scaling method flow chart of the invention.
Specific embodiment
Technical solution of the present invention is further illustrated with reference to the accompanying drawing.
As shown in Figure 1, a kind of data buffer storage stratification scaling method based on user activity, comprising the following steps:
S1, the activity of the user is calculated, initialization caching any active ues: background server is read after initialization system
User's logs in log and platform activity journal file, is calculated according to log recording all the activity of the user, will
The forward user data set for meeting buffer memory capacity size of liveness is moved into cache layer, and the caching as initialization is actively used
Family；Add a value of statistical indicant for identifying the active state of the user key-value pair information of the user of immigration simultaneously, initially all sets
It is set to any active ues；User activity calculating specifically includes following sub-step:
S11, the factor for choosing n influence liveness, choose m sample, form the primary data matrix of evaluation system:
That is: X={ xijM × n, 0≤i≤m, 0≤j≤n；
Wherein, xijIndicate the index value of i-th of user's jth item liveness factor；
S12, data standard processing is carried out:
Wherein,For the average value of the index value of jth item liveness factor,SjIt is active for jth item
The standard deviation of the index value of degree factor,
S13, the specific gravity Y for calculating i-th of user's index value of jth item liveness factorij:
S14, the specific gravity matrix for establishing data:
Y={ Yij}m*2；
S15, the information entropy for calculating jth item liveness factor:
Wherein, K is constant,
S16, the information utility value for calculating jth item liveness factor:
dj=1-Ej；
The weight of jth item liveness factor are as follows:
S17, weight shared by each factor, common liveness are calculated by a certain number of the activity of the user factors
Factor includes that the login times (C1) of user and each of user log in two kinds of duration (C2)；Then the activity of the user is calculated:
S2, the data between data buffer storage layer and database are synchronized；When system operates normally within the period, need
Accomplish that data between data buffer storage layer and database are synchronous, can the part smaller for requirement of real-time take tolerance
Higher synchronization timing can then relax requirement for general community function；
S3, the activity of the user is recalculated when reaching the cycle time set through operation after a period of time,
The average active degree of counting user is as liveness threshold value simultaneously；Compare the size of the capacity of liveness threshold value and cache layer, such as
Fruit liveness threshold value is greater than the capacity of cache layer then using liveness threshold value as third level early warning value；Otherwise cache layer capacity is made
For third level early warning value；
S4, first order early warning value and second level early warning value are arranged according to business demand；
S5, the first order early warning value counted according to step S3 and S4, second level early warning value, third level early warning value judge
The activity of the user situation of change in cache layer；Under the activity of the user situation of change and different liveness situations of change
Processing method include following several:
If the activity of the user is greater than first order early warning value in S51, data buffer storage layer, check that user's enlivens shape
State；If it is any active ues, then any operation is not done, directly reservation user data is in cache layer；If user is not active
User, then the state for modifying user is any active ues；
If the activity of the user is less than first order early warning value and is greater than second level early warning value in S52, data buffer storage layer,
The active state of user is set to remind any active ues, while sending message, such as sends short message and/or the mail reminder user；
If the activity of the user is less than second level early warning value and is greater than third level early warning value in S53, data buffer storage layer,
Whether the active state for continuing to judge the user in cache layer is active state, if it is active state, by the work of the user
Jump state is set as reminding any active ues, while sending the prompting message user；If it is any active ues are reminded, by user's
Active state is set as warning any active ues, while sending message and alerting the user, alerts the user and returns as early as possible and puts down
Platform is active；
If the activity of the user is less than third level early warning value in S54, data buffer storage layer, the user in cache layer is judged
Whether it is active state, if it is active state, then sets the active state of the user to reminding any active ues, and send and disappear
Breath reminds user；If it is any active ues are reminded, then the active state of the user is set for warning any active ues, concurrently
Message is sent to alert the user；If the active state of the user is warning any active ues in cache layer, by the user
The user data removed as needs from data buffer storage layer.
S6, carry out any active ues caching replacement: background server repeat step S5, to the user data in cache layer into
Row statistics obtains one and needs the user data list that removes, interim moves the user data from data buffer storage layer in this week
It removes；After removing user, the user data in cache layer is not present in the new liveness position p in the top and moves into caching
Layer, the key-value pair information for the user that replaces the user data of removal, and will move into add a value of statistical indicant for identifying the user
Active state；Wherein, p is the number of the user data removed；
S7, the operation that step S2~S6 is repeated according to the period of system manager's setting, are constantly updated in user cache layer
User.User experience is set to reach best；Until system is out of service, terminate the operation.
Operation is divided into two steps when user logs in, specific as follows: user first accesses data cache layer when logging in, and retrieves user
Whether in data buffer storage layer, if user is in data buffer storage layer, directly from data buffer storage layer obtain user's master data and
The associated documents of user；If user obtains related data from background server not in data buffer storage layer.
The active state of user is judged using multistage liveness early warning value:
It uses three-level early warning in the present invention to be monitored the active state of user, if the active state of user is not
It is able to maintain in active state, then can send short message to it or mail is reminded；It can be incited somebody to action if liveness is still in decline
To alerting, user's long period no activity is informed by mail and short message；If mentioned again without by liveness
It goes up and, then can be removed from data buffer storage layer using user as inactive user.The third level warning value of multistage warning value is
Average value based on user activity needs just to can determine that after the good overall liveness of statistics.And due to the appearance of caching
Measure it is limited, will appear if when the data volume of user is bigger user average active degree exceed buffer memory capacity limit
System, then can be using user activity that buffer memory capacity is truncated as liveness threshold value.As for the first second level warning value setting then
Need to be configured according to the needs of system, if be arranged it is too big if will lead to any active ues displacement it is more frequent, too
It is small, it will lead to any active ues list and almost seldom change, it is therefore desirable to which administrator provides ratio according to the specific implementation situation of business
Preferable scheme.
The present invention promotes the experience of user using data buffer storage layer.The data of any active ues are stored in data buffer storage
Layer, and the key-value pair information that will move into the user of data buffer storage layer adds a value of statistical indicant for identifying the user.Any active ues
Data directly are obtained from data buffer storage layer when logging in and need not access server, are reduced the response time, are promoted user satisfaction.
Meanwhile according to each cycle again to the calculated for rank of user activity, according to hash by the condition that is unsatisfactory for of data buffer storage layer
Any active ues remove data buffer storage layer, while moving into liveness and meeting the user for moving into data buffer storage layer to user cache layer.It is logical
Cross and be constantly movable into and out, making the information of the user saved in data buffer storage layer is more active user always, make be
System substantially reduces the response time of any active ues, and the overall experience of user is elevated.
Those of ordinary skill in the art will understand that the embodiments described herein, which is to help reader, understands this hair
Bright principle, it should be understood that protection scope of the present invention is not limited to such specific embodiments and embodiments.This field
Those of ordinary skill disclosed the technical disclosures can make according to the present invention and various not depart from the other each of essence of the invention
The specific variations and combinations of kind, these variations and combinations are still within the scope of the present invention.
Claims (3)
1. a kind of data buffer storage stratification scaling method based on user activity, which comprises the following steps:
S1, initialization caching any active ues: the activity of the user is calculated, by the forward use for meeting buffer memory capacity size of liveness
User data collection is moved into cache layer, the caching any active ues as initialization；The key-value pair information for moving into user is added simultaneously
One value of statistical indicant, the value of statistical indicant are used to identify the active state of the user, are initially both configured to any active ues；
S2, the data between data buffer storage layer and database are synchronized；
S3, when reach set cycle time when, recalculate the activity of the user, while the average active degree of counting user
As liveness threshold value；Compare the size of the capacity of liveness threshold value and cache layer, if liveness threshold value is greater than cache layer
Capacity is then using liveness threshold value as third level early warning value；Otherwise using cache layer capacity as third level early warning value；
S4, first order early warning value and second level early warning value are arranged according to business demand；
S5, the first order early warning value counted according to step S3 and S4, second level early warning value, third level early warning value judge slow
Deposit the activity of the user situation of change in layer；
S6, carry out any active ues caching replacement: background server repeats step S5, unites to the user data in cache layer
Meter obtains one and needs the user data list that removes, interim removes the user data from data buffer storage layer in this week；It moves
After user, the user data in cache layer is not present in the new liveness position p in the top and moves into cache layer, is replaced
The key-value pair information for the user that changes the user data of removal, and will move into adds a value of statistical indicant for identifying enlivening for the user
State；Wherein, p is the number of the user data removed；
S7, the operation that step S2~S6 is repeated according to the period of system manager's setting, until system is out of service.
2. a kind of data buffer storage stratification scaling method based on user activity according to claim 1, which is characterized in that institute
Stating step S1 user activity and calculating includes following sub-step:
S11, the factor for choosing n influence liveness, choose m sample, form the primary data matrix of evaluation system:
That is: X={ xijM × n, 0≤i≤m, 0≤j≤n；
Wherein, xijIndicate the index value of i-th of user's jth item liveness factor；
S12, data standard processing is carried out:
Wherein,For the average value of the index value of jth item liveness factor,SjFor jth item liveness because
The standard deviation of the index value of element,
S13, the specific gravity Y for calculating i-th of user's index value of jth item liveness factorij:
S14, the specific gravity matrix for establishing data:
Y={ Yij}m*2；
S15, the information entropy for calculating jth item liveness factor:
Wherein, K is constant,
S16, the information utility value for calculating jth item liveness factor:
dj=1-Ej；
The weight of jth item liveness factor are as follows:
S17, the activity of the user is calculated:
3. a kind of data buffer storage stratification scaling method based on user activity according to claim 2, which is characterized in that institute
It includes following several for stating the activity of the user situation of change in step S5 and the processing method under different liveness situations of change:
If the activity of the user is greater than first order early warning value in S51, data buffer storage layer, the active state of user is checked；Such as
Fruit is any active ues, then does not do any operation, and directly reservation user data is in cache layer；If user is not any active ues,
The state for then modifying user is any active ues；
If the activity of the user is less than first order early warning value and is greater than second level early warning value in S52, data buffer storage layer, it is arranged
The active state of user is to remind any active ues, while sending the prompting message user；
If the activity of the user is less than second level early warning value and is greater than third level early warning value in S53, data buffer storage layer, continue
Whether the active state for judging the user in cache layer is active state, and if it is active state, which is enlivened shape
State is set as reminding any active ues, while sending the prompting message user；If it is any active ues are reminded, by enlivening for user
State is set as warning any active ues, while sending message and alerting the user, alerts the user and returns platform work as early as possible
Jump；
If the activity of the user is less than third level early warning value in S54, data buffer storage layer, judge in cache layer whether is the user
It then sets the active state of the user to remind any active ues, and send message pair if it is active state for active state
User reminds；If it is any active ues are reminded, then the active state of the user is set for warning any active ues, and send and disappear
Breath alerts the user；If in cache layer the active state of the user be warning any active ues, using the user as
Need the user data removed from data buffer storage layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610792010.2A CN106339331B (en) | 2016-08-31 | 2016-08-31 | A kind of data buffer storage stratification scaling method based on user activity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610792010.2A CN106339331B (en) | 2016-08-31 | 2016-08-31 | A kind of data buffer storage stratification scaling method based on user activity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106339331A CN106339331A (en) | 2017-01-18 |
| CN106339331B true CN106339331B (en) | 2019-01-25 |
Family
ID=57823744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610792010.2A Active CN106339331B (en) | 2016-08-31 | 2016-08-31 | A kind of data buffer storage stratification scaling method based on user activity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106339331B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510298B (en) * | 2017-02-28 | 2021-09-21 | 创新先进技术有限公司 | Target user identification method and device |
| CN110704456A (en) * | 2018-06-22 | 2020-01-17 | 中兴通讯股份有限公司 | Method and device for updating business rules and computer readable storage medium |
| CN109379395B (en) * | 2018-08-22 | 2022-11-11 | 中国平安人寿保险股份有限公司 | Interface data cache setting method and terminal equipment |
| CN109285233A (en) * | 2018-09-28 | 2019-01-29 | 深圳壹账通智能科技有限公司 | Application based on big data is automatically signing in method, apparatus, medium and equipment |
| CN110061933B (en) * | 2019-04-03 | 2023-03-07 | 网宿科技股份有限公司 | Data processing method and device, equipment and storage medium |
| CN110738773A (en) * | 2019-10-30 | 2020-01-31 | 广州市果豆科技有限责任公司 | face recognition forbidden management method and system with local capacity expansion function |
| CN113688338A (en) * | 2020-05-19 | 2021-11-23 | 上海惠芽信息技术有限公司 | User data processing method, processing device and computer storage medium |
| CN113342780A (en) * | 2021-06-28 | 2021-09-03 | 深圳壹账通智能科技有限公司 | DSU data migration method and device and computer equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663096A (en) * | 2012-04-11 | 2012-09-12 | 北京像素软件科技股份有限公司 | Method for reading data based on data cache technology |
| CN105391654A (en) * | 2015-11-26 | 2016-03-09 | 中国建设银行股份有限公司 | Account activeness-based system resource allocation method and device |
| CN105656997A (en) * | 2015-12-25 | 2016-06-08 | 中国科学院信息工程研究所 | Hotness cache content active pushing method based on mobile user relationship |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120023226A1 (en) * | 2010-07-26 | 2012-01-26 | Steve Petersen | Prediction of activity session for mobile network use optimization and user experience enhancement |
-
2016
- 2016-08-31 CN CN201610792010.2A patent/CN106339331B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663096A (en) * | 2012-04-11 | 2012-09-12 | 北京像素软件科技股份有限公司 | Method for reading data based on data cache technology |
| CN105391654A (en) * | 2015-11-26 | 2016-03-09 | 中国建设银行股份有限公司 | Account activeness-based system resource allocation method and device |
| CN105656997A (en) * | 2015-12-25 | 2016-06-08 | 中国科学院信息工程研究所 | Hotness cache content active pushing method based on mobile user relationship |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106339331A (en) | 2017-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106339331B (en) | A kind of data buffer storage stratification scaling method based on user activity | |
| US8145623B1 (en) | Query ranking based on query clustering and categorization | |
| US10140472B2 (en) | Multi-level privacy evaluation | |
| CN106202569A (en) | A kind of cleaning method based on big data quantity | |
| CN106202581B (en) | A kind of data search method and device | |
| Park et al. | Characterizing datasets for data deduplication in backup applications | |
| CN106776967A (en) | Mass small documents real-time storage method and device based on sequential aggregating algorithm | |
| CN106649349A (en) | Method, device and system for data caching, applicable to game application | |
| Haas et al. | Suicide among older people: projections for the future | |
| US10318540B1 (en) | Providing an explanation of a missing fact estimate | |
| CN106528451B (en) | The cloud storage frame and construction method prefetched for the L2 cache of small documents | |
| CN104081364A (en) | Collaborative caching | |
| Ma et al. | An improved web cache replacement algorithm based on weighting and cost | |
| CN107908518A (en) | Database performance monitoring method, device, storage medium and computer equipment | |
| Chen et al. | Popularity-aware differentiated distributed stream processing on skewed streams | |
| Ren et al. | Using clustering for effective management of a semantic cache in mobile computing | |
| CN107133367A (en) | A kind of information processing method and system, server | |
| CN110309143A (en) | Data similarity determines method, apparatus and processing equipment | |
| US8874666B2 (en) | Publisher-assisted, broker-based caching in a publish-subscription environment | |
| CN114253917A (en) | Distributed self-adaptive storage method and system based on file access characteristics | |
| US9053155B2 (en) | Proxy cache aggregator | |
| CN106131118B (en) | A kind of methods, devices and systems of distributed caching | |
| Teymorian et al. | Rank-energy selective query forwarding for distributed search systems | |
| US20140229679A1 (en) | Management of cached data based on user engagement | |
| CN109189696B (en) | SSD (solid State disk) caching system and caching method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |