CN106202271A - The read method of the product database of OTA - Google Patents
The read method of the product database of OTA Download PDFInfo
- Publication number
- CN106202271A CN106202271A CN201610505426.1A CN201610505426A CN106202271A CN 106202271 A CN106202271 A CN 106202271A CN 201610505426 A CN201610505426 A CN 201610505426A CN 106202271 A CN106202271 A CN 106202271A
- Authority
- CN
- China
- Prior art keywords
- read
- data
- time
- data segment
- application server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24552—Database cache management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2471—Distributed queries
Abstract
The invention discloses the read method of the product database of a kind of OTA, comprise the following steps: S1, to split data to be cached in described product database be some data segments；S2, select some application servers；S3, described some application servers read from described product database respectively described some data segments and will read data write memory database in cache；S4, described application server inquire about data time from described memory database read data inquire about.The read method of the product database of the OTA that the present invention provides, when carrying out product inquiry, the foundation of the caching of product database and renewal process introduce the concept of distributed reading data, memory database is utilized to reduce the competition to product data base resource, thus improve the speed that caching is set up and updated, improve the overall performance of product inquiry, it is possible to the response speed when competition reducing the resource to product database and then the inquiry request that magnanimity can be improved.
Description
Technical field
The present invention relates to the reading of the product database of a kind of OTA (Online Travel Agency, online tourism society)
Method.
Background technology
The product inquiry of large-scale OTA needs the querying condition given for user to export corresponding Query Result rapidly,
Such as the real time price in hotel, house type and residue room quantity etc. are inquired about.In early days simple and compare solution intuitively and be
Allowing the direct access product database of application server to respond inquiry request, the advantage of this way is that handling process is simple, nothing
Redundant data, ensure that simultaneously and obtains Query Result more in real time.But, because to the reading of product database after all
Being the read-write to disk, and the read or write speed of disk has the limit, the performance of the product inquiry that therefore which results in OTA is difficult
To be protected, especially when the visit capacity of OTA increases, product database can be by bigger pressure, and this can become impact
The bottleneck of access performance.
In order to solve above-mentioned bottleneck, it is common practice to increase caching.Realization is cached with two kinds of ways, and the first is passive
Caching, i.e. only has its result after certain querying condition is used to just can be buffered；Another kind is active cache, i.e. should
The most actively read data from product database with server to be put into caching for subsequent query.Be no matter active cache also
It is passive caching, data cached can have that application server is local or (Redis is a use increased income such as Redis
ANSI C language writes, support network, can also can the data base of log type of persistence based on internal memory) or Memcached
In the memory database of (Memcached is a high performance distributed memory target cache system).Where no matter it is buffered in,
There is a task that two needs complete: one grows out of nothing when being initial and sets up data cached, and another is according to data cached
Freshness requirements upgrade in time caching.It is desirable that from product database reading real-time it can be appreciated that complete the two task
Data.Along with hotel and the quick growth of house type number, add hotel's room rate, room availability, the Rapid Variable Design of room amount and inquiry about the hotels
Requirement of real-time and foreground queries amount increase need increasing application server, complete the two task and mean often
One application server will ceaselessly read the data of magnanimity, common competing product database resource from product database.This
Will cause two performance issues: one caches foundation slowly when being initial, two is that the incremental data cached updates not in time, and these are asked
Topic is all likely to result in inquiry about the hotels information and is forbidden, and causes order failure and traffic lost.
Summary of the invention
The technical problem to be solved in the present invention is to overcome in prior art the product database of OTA in the face of magnanimity
During inquiry request, response speed is slow and may result in the inaccurate defect of Query Result, it is provided that one can reduce product number
The reading of the product database of the OTA of the response speed according to the competition of resource in storehouse and then when can improve the inquiry request of magnanimity
Method.
The present invention is to solve above-mentioned technical problem by following technical proposals:
The read method of the product database of a kind of OTA, its feature is, comprises the following steps: S1, split described product
Data to be cached in data base are some data segments；S2, select some application servers；S3, described some application servers divide
From described product database, do not read described some data segments and will the data write memory database of reading cache；S4、
Described application server reads data when inquiring about data from described memory database and inquires about.
In this programme, during inquiry product database, introduce the concept of distributed reading data, poke in i.e. utilizing
Build caching according to storehouse, first inquire about in memory database when application server needs to inquire about product database, thus reduce
Competition to product data base resource, thus response speed when improving the inquiry request of magnanimity；Additionally including application server
When caching is built in deposit data storehouse, split by the data treating caching, some application servers read parallel and wait to cache
Data after write memory database such that it is able to improve the speed that caching is set up and updated in memory database, thus improve
Overall performance.
It is preferred that step S4Further comprising the steps of: the number that described application server will read from described memory database
Local cache is carried out according to being saved in described application server.
In this programme, application server inquire about data time, first inquire about local cache, if local cache inquiry less than,
Reading data from memory database again to inquire about, the data of reading are saved in this locality simultaneously and cache, in case subsequent query
Use during data.
It is preferred that step S3Some application servers described in if read described respectively from described product database at random
Dry data segment.
In this programme, each application server uses when reading from product database wait the some data segments cached
Random sequence reads, thus reduces multiple stage application server and read the probability of same data segment.
It is preferred that described memory database is Redis or Memcached.
It is preferred that use distributed lock in described memory database, described distributed lock is that every segment data section arranges lock shape
State parameter, described lock status parameter includes TRUE and FALSE two states, step S3In each application server read one
Whether the described lock status parameter judging described data segment to be read before data segment to be read is FALSE state, if it is not,
The most do not read described data segment to be read, the most then read described data segment to be read, arrange described to be read simultaneously
The described lock status parameter of data segment is TRUE state, and arranges described in described data segment to be read after reading again
Lock status parameter is FALSE state.
In this programme, TRUE state representation current time has application server that this segment data section is being built caching behaviour
Making, it is exactly to read data segment from product database and be written to memory database the behaviour carrying out caching that what is called builds caching
Making, FALSE state is original state, represents that this segment data section current time does not has application server that it is built caching behaviour
Make.Take building of this segment data section when application server to cache temporary, before formally starting to build, lock status parameter can be set to
TRUE state, when the caching of this segment data section built up by application server when, can be set to FALSE state.Each application takes
Whether business device reading the described lock status parameter judging described data segment to be read before a data segment to be read is
FALSE state, if it is not, the most do not read described data segment to be read, the most then reads described data segment to be read, simultaneously
The described lock status parameter arranging described data segment to be read is TRUE state, and read after arrange again described in continue
The described lock status parameter of the data segment taken is FALSE state.Can be very big by employing distributed lock in memory database
Ground reduces by two use above servers and builds the probability of caching simultaneously for same data segment.
It is preferred that described distributed lock also arranges parameter locking time, step S for every segment data section3In arrange described in treat
The described lock status parameter of the data segment read arranges the locking of described data segment to be read the most simultaneously when being TRUE state time
Between parameter be that to arrange the described lock status parameter of described data segment to be read be time value during TRUE state；Step S3In should
With server when judging that the described lock status parameter of described data segment to be read is TRUE state, then judge current time and
Described data segment to be read described locking time parameter difference whether more than very first time threshold value, the most described should
Reading described data segment to be read with server, parameter locking time simultaneously arranging described data segment to be read is current
Time, if it is not, the most do not read described data segment to be read.
In this programme, by relatively data segment to be read locking time parameter value and the difference of current time with in advance
The very first time threshold value preset, if difference exceedes very first time threshold value, then illustrate that data segment to be read is different between being read out
Often time-out, depends on even if the lock status parameter of data segment to be read when other application server finds this situation is TRUE state
So can read this data segment and write memory database.Locking time, the effect of parameter was to prevent application server from building
Occur the when of caching that abnormal conditions exit, this time application server will not actively lock status parameter from TRUE state
Being set to FALSE state, other application servers can be according to lock status parameter and parameter locking time, it is judged that lock is the most overtime.
If the difference of parameter locking time and current time sets up time-out time i.e. very first time threshold more than or equal to the largest buffered arranged
Value, then will be considered that problematic, and other application servers can be current time this time modification, continues to build this segment data section
Caching.
It is preferred that described distributed lock also arranges time cost parameter for every segment data section, described time cost parameter is used
In characterizing step S3Middle application server reads described data segment to be read the number that will read from described product database
The time spent according to caching in write memory database.
In this programme, time cost parameter builds, for monitoring each segment data section, the time that caching spends, and assesses the big of segmentation
Little the most reasonable.In general, the every period building caching can not be oversize, if oversize, then explanation data segment should be got again
Smaller.
It is preferred that described distributed lock also arranges renewal time parameter, step S for every segment data section3In each application clothes
When business device arranges described renewal time parameter for caching complete after by caching in the data described memory database of write of reading
Time；Step S3In each application server judge current time before described data segment to be read and described treat reading
The difference of the described renewal time parameter of the data segment read whether more than the second time threshold preset, the most described should
Described data segment to be read is read, if it is not, the most do not read described data segment to be read with server.
In this programme, update time parameter and build the time of caching for recording this segment data section the last time, for application clothes
Business device is used for judging whether to need again this segment data section to be built caching.If within the effect duration that data allow, the most not
More than the second time threshold, then need not again build caching, only when the last update time exceeded allow the longest time
Between, then can trigger and build caching.
It is preferred that step S3Described in some application servers from described product database, read described some numbers respectively
During according to section, the quantity of the application server simultaneously reading described product database is four to the maximum.
In this programme, when the situation occurring application server cluster to restart, if application server is a lot of in cluster, institute
Some application servers can be exerted heavy pressures on to product database when product database reads data simultaneously.Therefore in this programme
Limit one to allow to read data cached maximum number of concurrent from product database simultaneously.Every application server is restarted and is caused this locality
The when of data cached whole loss, all can be toward memory database write the reboot time of application server, when any one
Application server has had four application servers reading product simultaneously in seeing the set time from memory database when starting
Product data base toward when writing data in memory database, it the most only from memory database reading data not from product database reading
According to writing in memory database.
The most progressive effect of the present invention is: the read method of the product database of the OTA that the present invention provides, and is carrying out
During product inquiry, the foundation of the caching of product database and renewal process introduce the concept of distributed reading data, in utilizing
The competition to product data base resource is reduced in deposit data storehouse, thus improves the speed that caching is set up and updated, and improves product inquiry
Overall performance.By product database data efficient is read memory database cache or application server this locality is delayed
Deposit, until data cached by being read the data in product database by the division of labor of multiple stage application server after splitting in product database
Being then written in memory database share for other application servers, other application servers are within sharing the most as required
Deposit data banked cache reads application server local cache.The present invention can reduce the resource to product database competition and then
The response speed during inquiry request of magnanimity can be improved.
Accompanying drawing explanation
Fig. 1 is the flow chart of the read method of the product database of the OTA of a preferred embodiment of the present invention.
Fig. 2 is the schematic diagram that in a preferred embodiment of the present invention, step S103 random order reads data segment.
Detailed description of the invention
Further illustrate the present invention below by the mode of embodiment, but the most therefore limit the present invention to described reality
Execute among example scope.
As it is shown in figure 1, the read method of the product database of a kind of OTA, comprise the following steps:
Data to be cached in S101, partitioning products data base are some data segments, and in the present embodiment, product is hotel,
I.e. in segmentation hotel data base, data to be cached are that some data segments are for subsequent read.
S102, select some application servers, in memory database Redis, build caching, in the present embodiment for follow-up
Select four application servers, if the quantity reading the application server of hotel data base is too much simultaneously, easily cause hotel
Database resource dog-eat-dog, causes reading data performance low.
S103, four application servers choose data segment to be read with random order respectively from hotel data base,
Random order advantageously reduces four application servers and reads the probability of same data segment simultaneously.
Using distributed lock in S104, memory database Redis, distributed lock is that every segment data section arranges lock status ginseng
Number, lock status parameter includes TRUE and FALSE two states, and application server judges the lock status parameter of data segment to be read
Whether it is FALSE, if it is not, then perform step S106, the most then performs step S105.
S105, distributed lock also arrange renewal time parameter for every segment data section, and application server judges number to be read
According to the requirement updating the most satisfied more new data of time parameter of section, i.e. judge current time and the renewal of data segment to be read
The difference of time parameter whether more than the second time threshold preset, the most then performs step S108, if it is not, then perform step
S107。
S106, distributed lock also arrange parameter locking time for every segment data section, and this parameter is used for preventing application server
Read during a data segment due to abnormal cause cause reading time-out, specially application server judge to be read
The lock status parameter of data segment when being TRUE state, then judge parameter locking time of current time and data segment to be read
Difference whether more than very first time threshold value, the most then perform step S108, if it is not, then perform step S107.
S107, application server are abandoned reading data segment the most to be read, if also there is a need to the data segment read,
Returning S103, if all data segments have all read, then present application server is built caching task in Redis and is terminated.
It is TRUE that S108, application server arrange the lock status parameter of current data segment to be read, arranges ginseng locking time
Number is current real-time time.
S109, application server read data segment to be read from hotel data base, and write in Redis.
It is current real-time time that S110, application server arrange the time parameter that updates of current data section, it addition, in order to
The size segmentation of statistical data section is the most suitable, and distributed lock also arranges time cost parameter for every segment data section, the most also needs
The time cost parameter that notebook data section to be arranged is corresponding, so-called time cost parameter is used to characterize application server from hotel's number
According to storehouse being read data segment to be read and the data of reading being write the time that in Redis, caching is spent, if spent
Between long, then need the size of data segment, again partition data in set-up procedure S101.
S111, participate in building the application server of caching or other application servers when needs inquiry data from Redis
Read data to inquire about, the data of reading are carried out local cache in application server this locality simultaneously, if local built slow
Depositing, application server, when performing query task, can being inquired about by local cache now, if do not inquired, going the most again
Inquiring about in Redis, if also do not inquired, Ze Zaiqu hotel data base inquires about, can basis during inquiry
Need to repeat abovementioned steps.
In the present embodiment, step S101 uses every 200000 data needing caching to constitute one when partition data section
Data segment.In Fig. 2, server 1 to server 4 is four application servers chosen in step S102, and data segment to be cached is
First order is 1 to 9, and in four application servers, each application server is according to random suitable of as shown in Figure 2
Xu Cong hotel data base is sequentially read out and writes into memory database based on Redis each data segment.Can by Fig. 2
Know, use random order reading data segment can reduce by four application servers and read the probability of same data segment simultaneously.
Before Mei Cong hotel data base reads a data segment, application server first consults this data segment the most Already in
On Redis, if Already in Redis, and this data segment does not has expired, then need not again read or write Redis.As
Really application server also needs to set up local cache, then at this moment this data segment also should be loaded into local cache.
In the present embodiment, it is contemplated that when there is the situation that application server cluster is restarted, if application server in cluster
Quantity is a lot, and all of application server is read data Shi Huigei hotel data base from hotel data base simultaneously and exerted heavy pressures on.
Limiting one to allow to read data cached maximum number of concurrent from hotel data base, the present embodiment is limited to 4 simultaneously.Every application
The when that Server Restart causing local cache data all to lose, all can write restarting of present application server in Redis
Time, in seeing from Redis nearest one minute when any application server starts, there are 4 or more than 4
Other application server is simultaneously reading hotel data base and toward when writing data in Redis, and it is the most only from Redis reading data the most not
Read data from hotel data base to write Redis.
In the present embodiment, the foundation and renewal process of inquiry about the hotels caching introduce the concept of distributed reading data,
Utilize memory database Redis to reduce the competition to hotel's database resource, thus improve the speed that caching is set up and updated, carry
High overall performance.Specifically, full dose or increment treat data cached divided by multiple stage application server by certain allocation rule
Part work and part study takes hotel data base, and the data of reading are then written in memory database Redis share, then for other application servers
Other application servers read application server local cache from the caching of the memory database Redis shared as required.This
The method of embodiment mainly has several feature:
First, have the most data cached splicing mechanism.For substantial amounts of data cached, can be according to data record
Bar number or data volume are divided into multiple data segment.The caching of full dose and the caching of increment can use identical or different segmentation
Mechanism.
Second, have rational data pull distribution mechanism.According to certain rule, all of data segment is distributed to many
Individual application server.Rule for distribution can be the static rule reserved in advance, it is also possible to be to build caching progress according to actual and
The fixed DP with certain randomness.
3rd, application server pulls data cached to memory database Redis, pulling data parallel according to allocation rule
Certain synchronization mechanism can be used to avoid multiple stage application server to repeat to pull same data segment during section.
4th, when application server needs set up and update local cache, can read in memory database Redis
Data are to local internal memory.
5th, it is also possible to a caching is set and sets up the monitoring instrument of progress.
6th, there is suitable protection mechanism and prevent too much application server from pulling caching number from data base concurrency simultaneously
According to.
7th, there is the unexpected off-line of application server and carry out fault-tolerant mechanism, it is ensured that the reliable of caching is set up and update.
In the present embodiment, monitoring instrument monitoring data show: other application servers from Redis for data
The elapsed time of local cache is built less than or equal to 39 milliseconds in storehouse, and the time directly built local cache from hotel data base and need is
3236 milliseconds, the read method of the product database of the OTA provided by the present invention thus can be provided, be effectively improved slow
Deposit the speed set up and update, alleviate the pressure of product database server, and promote the overall performance of whole system and stablize
Property.
Although the foregoing describing the detailed description of the invention of the present invention, it will be appreciated by those of skill in the art that this is only
Being to illustrate, protection scope of the present invention is defined by the appended claims.Those skilled in the art without departing substantially from
On the premise of the principle of the present invention and essence, these embodiments can be made various changes or modifications, but these changes and
Amendment each falls within protection scope of the present invention.
Claims (9)
1. the read method of the product database of an OTA, it is characterised in that comprise the following steps:
S1, to split data to be cached in described product database be some data segments；
S2, select some application servers；
S3, described some application servers read from described product database respectively described some data segments and will read number
Cache according in write memory database；
S4, described application server inquire about data time from described memory database read data inquire about.
2. the read method of the product database of OTA as claimed in claim 1, it is characterised in that step S4Also include following step
Rapid: the data read from described memory database are saved in described application server and carry out local slow by described application server
Deposit.
3. the read method of the product database of OTA as claimed in claim 1, it is characterised in that step S3Described in some should
From described product database, described some data segments are read respectively at random with server.
4. the read method of the product database of OTA as claimed in claim 1, it is characterised in that described memory database is
Redis or Memcached.
5. the read method of the product database of OTA as claimed in claim 4, it is characterised in that in described memory database
Use distributed lock, described distributed lock is that every segment data section arranges lock status parameter, described lock status parameter include TRUE and
FALSE two states, step S3In each application server read judge before a data segment to be read described in continue
Whether the described lock status parameter of the data segment taken is FALSE state, if it is not, the most do not read described data segment to be read, if
Being then to read described data segment to be read, the described lock status parameter simultaneously arranging described data segment to be read is TRUE
State, and to arrange the described lock status parameter of described data segment to be read again after reading be FALSE state.
6. the read method of the product database of OTA as claimed in claim 5, it is characterised in that described distributed lock is also
Every segment data section arranges parameter locking time,
Step S3The middle described lock status parameter arranging described data segment to be read is treated described in arranging when being TRUE state the most simultaneously
Parameter locking time of data segment read is that to arrange the described lock status parameter of described data segment to be read be TRUE state
Time time value；
Step S3Middle application server when judging that the described lock status parameter of described data segment to be read is TRUE state, then
Judge current time and described data segment to be read described locking time parameter difference whether more than very first time threshold value,
The most described application server reads described data segment to be read, arranges the locking of described data segment to be read simultaneously
Time parameter is current time, if it is not, the most do not read described data segment to be read.
7. the read method of the product database of OTA as claimed in claim 5, it is characterised in that described distributed lock is also
Every segment data section arranges time cost parameter, and described time cost parameter is used for characterizing step S3Middle application server is from described product
Product data base reads described data segment to be read and the data of reading are write that caching in memory database spent time
Between.
8. the read method of the product database of OTA as claimed in claim 5, it is characterised in that described distributed lock is also
Every segment data section arranges renewal time parameter,
Step S3In each application server the data of reading are being write in described memory database and are arranging described renewal after caching
Time when time parameter is to cache complete；
Step S3In each application server judge current time and described to be read before described data segment to be read reading
The difference of described renewal time parameter of data segment whether more than the second time threshold preset, the most described application takes
Business device reads described data segment to be read, if it is not, the most do not read described data segment to be read.
9. the read method of the product database of the OTA as according to any one of claim 1 to 8, it is characterised in that step S3
Described in some application servers when reading described some data segments from described product database respectively, read described product simultaneously
The quantity of the application server of product data base is four to the maximum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610505426.1A CN106202271A (en) | 2016-06-30 | 2016-06-30 | The read method of the product database of OTA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610505426.1A CN106202271A (en) | 2016-06-30 | 2016-06-30 | The read method of the product database of OTA |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106202271A true CN106202271A (en) | 2016-12-07 |
Family
ID=57463922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610505426.1A Pending CN106202271A (en) | 2016-06-30 | 2016-06-30 | The read method of the product database of OTA |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106202271A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107911248A (en) * | 2017-11-27 | 2018-04-13 | 北京百度网讯科技有限公司 | Upgrade method and device |
| CN110673893A (en) * | 2019-09-24 | 2020-01-10 | 携程计算机技术（上海）有限公司 | Configuration method and system of application program, electronic device and storage medium |
| CN113918530A (en) * | 2021-12-14 | 2022-01-11 | 北京达佳互联信息技术有限公司 | Method and device for realizing distributed lock, electronic equipment and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130103655A1 (en) * | 2011-10-21 | 2013-04-25 | International Business Machines Corporation | Multi-level database compression |
| CN103561013A (en) * | 2013-10-29 | 2014-02-05 | 联想中望系统服务有限公司 | Streaming media data distributing system |
| CN103729239A (en) * | 2013-11-18 | 2014-04-16 | 芜湖大学科技园发展有限公司 | Distributed type lock algorithm of mirror-image metadata |
| CN104065636A (en) * | 2013-07-02 | 2014-09-24 | 腾讯科技（深圳）有限公司 | Data processing method and system |
| CN104932953A (en) * | 2015-06-04 | 2015-09-23 | 华为技术有限公司 | Data distribution method, data storage method, and relevant device and system |
| CN105224255A (en) * | 2015-10-14 | 2016-01-06 | 浪潮(北京)电子信息产业有限公司 | A kind of storage file management method and device |
-
2016
- 2016-06-30 CN CN201610505426.1A patent/CN106202271A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130103655A1 (en) * | 2011-10-21 | 2013-04-25 | International Business Machines Corporation | Multi-level database compression |
| CN104065636A (en) * | 2013-07-02 | 2014-09-24 | 腾讯科技（深圳）有限公司 | Data processing method and system |
| CN103561013A (en) * | 2013-10-29 | 2014-02-05 | 联想中望系统服务有限公司 | Streaming media data distributing system |
| CN103729239A (en) * | 2013-11-18 | 2014-04-16 | 芜湖大学科技园发展有限公司 | Distributed type lock algorithm of mirror-image metadata |
| CN104932953A (en) * | 2015-06-04 | 2015-09-23 | 华为技术有限公司 | Data distribution method, data storage method, and relevant device and system |
| CN105224255A (en) * | 2015-10-14 | 2016-01-06 | 浪潮(北京)电子信息产业有限公司 | A kind of storage file management method and device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107911248A (en) * | 2017-11-27 | 2018-04-13 | 北京百度网讯科技有限公司 | Upgrade method and device |
| CN107911248B (en) * | 2017-11-27 | 2020-11-10 | 北京百度网讯科技有限公司 | Upgrading method and device |
| CN110673893A (en) * | 2019-09-24 | 2020-01-10 | 携程计算机技术（上海）有限公司 | Configuration method and system of application program, electronic device and storage medium |
| CN110673893B (en) * | 2019-09-24 | 2023-06-09 | 携程计算机技术（上海）有限公司 | Application program configuration method, system, electronic device and storage medium |
| CN113918530A (en) * | 2021-12-14 | 2022-01-11 | 北京达佳互联信息技术有限公司 | Method and device for realizing distributed lock, electronic equipment and medium |
| CN113918530B (en) * | 2021-12-14 | 2022-05-13 | 北京达佳互联信息技术有限公司 | Method and device for realizing distributed lock, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105549905B (en) | A kind of method that multi-dummy machine accesses distributed objects storage system | |
| CN102831156B (en) | Distributed transaction processing method on cloud computing platform | |
| US7783607B2 (en) | Decentralized record expiry | |
| CN101013381B (en) | Distributed lock based on object memory system | |
| CN106599199A (en) | Data caching and synchronization method | |
| CN108509462B (en) | Method and device for synchronizing activity transaction table | |
| CN111338766A (en) | Transaction processing method and device, computer equipment and storage medium | |
| WO2021027956A1 (en) | Blockchain system-based transaction processing method and device | |
| CN103106286B (en) | Method and device for managing metadata | |
| US20070288587A1 (en) | Transactional shared memory system and method of control | |
| US20090063807A1 (en) | Data redistribution in shared nothing architecture | |
| US9922086B1 (en) | Consistent query of local indexes | |
| CN111767327B (en) | Data warehouse construction method and system with dependency relationship among data streams | |
| CN103226598A (en) | Method and device for accessing database and data base management system | |
| CN106777085A (en) | A kind of data processing method, device and data query system | |
| CN106202271A (en) | The read method of the product database of OTA | |
| CN109669975A (en) | A kind of industry big data processing system and method | |
| CN115587118A (en) | Task data dimension table association processing method and device and electronic equipment | |
| US11741081B2 (en) | Method and system for data handling | |
| CN111159140A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN102724301B (en) | Cloud database system and method and equipment for reading and writing cloud data | |
| EP3377970B1 (en) | Multi-version removal manager | |
| US11138231B2 (en) | Method and system for data handling | |
| CN112131305A (en) | Account processing system | |
| CN115964444A (en) | Cloud native distributed multi-tenant database implementation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |