CN105677650A - Cache data storage method and device - Google Patents

Abstract

The invention provides a cache data storage method and device; the method comprises the following steps: obtaining a first update verify clock and a second update verify clock of present cache data with no download life cycle, wherein the first and second update verify clocks are adjacent two update verify clocks; obtaining the lift cycle of the present cache data according to the first and second update verify clocks; carrying out leveled storage for the present cache data according to the life cycle. The novel method and device can effectively alleviate server load, reduce network jamming, reduce user access delays, thus improving processing efficiency for different resource data.

Description

A kind of data cached storage means and device

Technical field

The present invention relates to technical field of data storage, particularly relate to a kind of data cached storage means and device.

Background technology

The develop rapidly of Internet technology and the continuous increase of user's visit capacity result in that network is congested and the overload of server, and buffer memory technology can effectively alleviate the load of server, to reduce network congested and reduces the access delay of user. The content that user can be accessed by buffer memory technology retains one part of copy in storing device, in order to user's access next time. Web content is carried out classification stores processor by the life cycle that existing technical scheme is mostly the access temperature according to content or content, and most caching system all have employed classification memory technology, is efficiently processed by different resource.

At present, classification memory technology scheme, for the process of local network resource, is that the access temperature according to different resource or life cycle carry out different stage division, and each storage system has storage zone at different levels, analysis of central issue unit and reorientation unit. Each storage zone is provided with the mark representing and placing rank, and the height of described storage rank represents the height of storage media performance; Analysis of central issue unit carrys out the access temperature of Computed-torque control, and wherein the height of rank is placed in the height corresponding stored district of access to content temperature; Reorientation unit is used for the storage zone that content is put into corresponding level by the access temperature according to file. For the Web content that temperature height or life cycle are shorter, deposited in the fastest storage media (such as internal memory) of read or write speed, the Web content of temperature is low or life cycle length, is deposited in the slower storage media (such as machinery hard disk) of read-write.

For the access temperature of Web content, it is possible to by buffer memory server real time record and add up; And the life cycle for content, then mark by content distributed original website (abbreviation source station) completely, but internet content life cycle mark mode has multiple at present, and different.

In current existing buffer memory technology, when buffer memory server is from while the download network content of source station, according to classification storage policy, judging the life cycle of this resource, and select different storage medias according to cycle length. Owing to the mark mode of internet resource life cycle mainly contains two kinds of modes, a kind of directly mark mode, namely indicates the life cycle length that content is current;Another kind marks indirectly, i.e. only this version number of tag content or this renewal time, first the version number of this content or this renewal time are verified to source station when meaningful request, if source station version does not upgrade, then represent that content is still effectively, can directly use. When existing content classification technology carries out classification storage in initial downloaded content and in this locality, only the content that can directly obtain life cycle is carried out the coupling of classification strategy, and for adopting the content of indirect mark mode, owing to cannot directly obtain its life cycle, therefore this part content is not carried out classification strategy process, only content is directly deposited in the storage media of lowest speed, which seriously affects the efficiency that classification stores, run counter to the original intention using classification technique, the serious efficiency that have impact on classification storage.

Summary of the invention

The technical problem to be solved in the present invention there is provided a kind of data cached storage means and device, can the data cached life cycle of active probe, its life cycle is normalized, then it is carried out classification stores processor, it is to increase to the processing efficiency of different resource data.

In order to solve the problems of the technologies described above, the present invention adopts following technical scheme:

According to an aspect of the present invention, it provides a kind of data cached storage means, comprising:

Obtaining the first of the current cache data not downloading life cycle to upgrade the verification moment and the 2nd and upgrade the verification moment, wherein, described first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Upgrade verification moment and the 2nd renewal verification moment according to first, obtain the life cycle of current cache data;

According to described life cycle, current cache data are carried out classification storage.

Can selection of land, obtain do not download life cycle current cache data first upgrade verification the moment and the 2nd upgrade inspection the moment step comprise:

Carrying out life cycle to source station and upgrade verification, if content is updated, then record upgrades the renewable time verified and upgrades the verification moment as first;

The verification moment is upgraded, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment according to described first.

Can selection of land, according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is described data cached sequence number, and y upgrades the verification moment in described data cached the first renewal verification moment or the 2nd, and a, b, c are parameter.

Can selection of land, adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& {{\mathrm{\Σx}}_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Obtaining the parameter value of a, b, c, wherein, m is when this calculates, the described data cached number of times being updated, x_iFor described data cached sequence number, y_iFor described data cached renewal each time verifies the moment.

Can selection of land, upgrade the verification moment and the 2nd according to first and upgrade the verification moment, the step of life cycle obtaining current cache data comprises:

The life cycle of the time period between verification moment and the first renewal verification moment as current cache data is upgraded using the 2nd.

Can selection of land, according to described life cycle, the step that current cache data carry out classification storage comprises:

If described life cycle is in the default first renewal time interval, then by current cache data stored in one-level storage zone;

If described life cycle is in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

If described life cycle is in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

Can selection of land, described method also comprises: if current cache data are not updated, then according to the life cycle of current cache data that the last time calculates, current cache data are carried out classification storage.

According to another aspect of the present invention, additionally provide a kind of data cached storing device, comprising:

Acquisition module, upgrades the inspection moment for obtaining the first of the current cache data not downloading life cycle and upgrade the verification moment and the 2nd, and wherein, described first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Calculate module, for upgrading verification moment and the 2nd renewal verification moment according to first, obtain the life cycle of current cache data;

Memory module, for according to described life cycle, carrying out classification storage to current cache data.

Can selection of land, described acquisition module comprises:

Record cell, upgrades verification for carrying out life cycle to source station, if content is updated, then record upgrades the renewable time verified and upgrades the verification moment as first;

Projected unit, for upgrading the verification moment according to described first, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment.

Can selection of land, described projected unit is according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is described data cached sequence number, and y upgrades the verification moment in described data cached the first renewal verification moment or the 2nd, and a, b, c are parameter.

Can selection of land, described projected unit is further

Adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& {{\mathrm{\Σx}}_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Obtaining the parameter value of a, b, c, wherein, m is when this calculates, the described data cached number of times being updated, x_iFor described data cached sequence number, y_iFor described data cached renewal each time verifies the moment.

Can selection of land, described calculating module is further used for: upgrading the verification moment and first using the 2nd upgrades the time period between the verification moment as the life cycle of current cache data.

Can selection of land, described memory module comprises:

First storage unit, if for described life cycle in the default first renewal time interval, then by current cache data stored in one-level storage zone;

2nd storage unit, if for described life cycle in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

3rd storage unit, if for described life cycle in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

Can selection of land, be not updated if described memory module is also further used for current cache data, then according to the life cycle of current cache data that the last time calculates, current cache data carried out classification storage.

The invention has the beneficial effects as follows:

In embodiments of the present invention, according to classification foundation as content of the life cycle of content and replacement cycle, for the resource that cannot directly obtain life cycle, the mode adopting active probe and carrying out evolution reckoning obtains its life cycle, and then corresponding storage zone is stored in classification. When obtaining data cached life cycle and classification stores, first to be obtained the first of current cache data and upgrade verification moment and the 2nd renewal verification moment, wherein the first renewal verification moment was adjacent twice renewal verification moment with the 2nd renewal verification moment, then upgrade, according to first, the life cycle that the verification moment and the 2nd upgrades verification moment acquisition current cache data, and the life cycle according to current cache data carries out classification storage. The present invention effectively alleviates the load of server, to reduce network congested, it is to increase to the response speed of the access request of Web content, had both improve the processing efficiency to different resource data, and had improved again the rate of utilization of system performance.

Accompanying drawing explanation

Fig. 1 represents the schema of data cached storage means in the embodiment of the present invention;

Fig. 2 represents in the embodiment of the present invention data cached processing flow chart;

Fig. 3 represents the schema obtaining in the embodiment of the present invention and upgrading the verification moment;

Fig. 4 represents the schema of classification memory buffers data in the embodiment of the present invention;

Fig. 5 represents the structure block diagram of data cached storing device in the embodiment of the present invention;

Fig. 6 represents the structure block diagram of acquisition module in the embodiment of the present invention; And

Fig. 7 represents the structure block diagram of memory module in the embodiment of the present invention.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly, describe the present invention below in conjunction with the accompanying drawings and the specific embodiments.

Embodiment one

As shown in Figure 1, being the schema of storage means data cached in the embodiment of the present invention, method 100 comprises the steps:

Step S101, obtain the current cache data not downloading life cycle first upgrade the verification moment and the 2nd and upgrade the verification moment, and wherein, first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Step S103, according to first upgrade verification the moment and the 2nd upgrade verification the moment, obtain current cache data life cycle;

Step S105, according to life cycle, current cache data are carried out classification storage.

Further, in embodiments of the present invention, if current cache data are not updated, then current cache data are carried out classification storage by the life cycle of the current cache data calculated according to the last time.

Specifically, to data cached treatment scheme as shown in Figure 2, first data cached life cycle mark mode is resolved, if its mark mode is the life cycle directly marking current cache data, then directly according to life cycle length, corresponding storage unit is selected to store; If its mark mode is indirectly mark, i.e. only this version number of tag content or this renewal time, then initiate content update verification to source base station, twice adjacent time difference upgrading the verification moment is the life cycle of current cache data, according to its life cycle, corresponding storage unit is selected to store.

In embodiments of the present invention, as shown in Figure 3, for obtaining the schema in renewal verification moment, the first renewal verification moment and the 2nd renewal verification moment (step S101) that obtain the current cache data not downloading life cycle comprise the following steps:

Step S1011, carry out life cycle to source station upgrade verification, if content is updated, then record upgrade verification renewable time as first upgrade verification the moment;

Step S1013, upgrade the verification moment according to first, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment.

In embodiments of the present invention, when calculating its life cycle, this data cached history renewable time and number of times is recorded by this locality, extrapolate the curve that the conversion rule with its life cycle is close, adopt the method for least square conic fitting, thus calculate and obtain next renewable time, and then extrapolate the life cycle of current cache data.

Wherein, specifically, when calculating that using dynamic least square conic fitting method obtaining the 2nd renewal verifies the moment, according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is data cached sequence number, and the first renewal verification moment or the 2nd that y is data cached upgrades the verification moment, and a, b, c are parameter.When calculating the parameter value of a, b, c,

Adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& {{\mathrm{\Σx}}_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Wherein, m is when this calculates, the data cached number of times being updated, x_iFor data cached sequence number, can value be 1,2,3 ..., y_iFor data cached renewal each time verifies the moment, value is this data cached time experienced from the download of source station.

Data cached when upgrading for the m time, it is possible to calculate a, the parameter value of b, c according to history renewable time, the conic fitting expression formula y=ax of this data cached renewable time can be obtained according to the parameter value of a, b, c²+ bx+c, then substitute into x for (m+1), can obtain being updated the moment of verification next time, be designated as y_m+1, upgrade verification moment and this time difference upgrading the verification moment next time, it be the life cycle that this is data cached. Therefore upgrading according to first and verify moment and the 2nd renewal verification moment, the life cycle (step S103) of current cache data is specially: the 2nd renewal verification moment and the first renewal are verified the life cycle of the time period between the moment as current cache data. As shown in Figure 4, in embodiments of the present invention, according to life cycle, current cache data are carried out classification storage (step S105) and specifically comprises the steps:

If step S1051 life cycle is in the default first renewal time interval, then by current cache data stored in one-level storage zone;

If step S1053 life cycle is in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

If step S1055 life cycle is in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

By to above-mentioned to data cached life cycle active probe and carry out evolution reckoning, data cached life cycle is normalized, achieve and data cached classification is stored, effectively improve the processing efficiency to different resource data, improve the rate of utilization of system performance.

Embodiment two

As shown in Figure 5, being the structure block diagram of storing device data cached in the embodiment of the present invention, this device 500 comprises:

Acquisition module 501, upgrades the inspection moment for obtaining the first of the current cache data not downloading life cycle and upgrade the verification moment and the 2nd, and wherein, first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Calculate module 503, for upgrading verification moment and the 2nd renewal verification moment according to first, obtain the life cycle of current cache data;

Memory module 505, for according to life cycle, carrying out classification storage to current cache data.

Wherein, be not updated if memory module 505 is also further used for current cache data, then current cache data are carried out classification storage by the life cycle of the current cache data calculated according to the last time.

In embodiments of the present invention, as shown in Figure 6, acquisition module 501 comprises:

Record cell 5011, upgrades verification for carrying out life cycle to source station, if content is updated, then record upgrades the renewable time verified and upgrades the verification moment as first;

Projected unit 5013, for upgrading the verification moment according to first, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment.

Wherein, projected unit 5013 is according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is data cached sequence number, and the first renewal verification moment or the 2nd that y is data cached upgrades the verification moment, and a, b, c are parameter.

Further, when calculating the parameter value of a, b, c, projected unit 5013

Adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& {{\mathrm{\Σx}}_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Obtaining the parameter value of a, b, c, wherein, m is when this calculates, the data cached number of times being updated, x_iFor data cached sequence number, y_iFor data cached renewal each time verifies the moment.

In embodiments of the present invention, calculate module 503 to be further used for: upgrade the life cycle of the time period between verification moment and the first renewal verification moment as current cache data using the 2nd.

As shown in Figure 7, memory module 505 comprises:

First storage unit 5011, if for life cycle in the default first renewal time interval, then by current cache data stored in one-level storage zone;

2nd storage unit 5013, if for life cycle in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

3rd storage unit 5015, if for life cycle in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

Above-described is the preferred embodiment of the present invention; should be understood that the common personnel for the art; can also making some improvements and modifications under principle prerequisite of the present invention not departing from, these improvements and modifications are also in protection scope of the present invention.

Claims (14)

1. a data cached storage means, it is characterised in that, comprising:

Obtaining the first of the current cache data not downloading life cycle to upgrade the verification moment and the 2nd and upgrade the verification moment, wherein, described first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Upgrade verification moment and the 2nd renewal verification moment according to first, obtain the life cycle of current cache data;

According to described life cycle, current cache data are carried out classification storage.

2. data cached storage means as claimed in claim 1, it is characterised in that, the first renewal verification moment and the 2nd step upgrading the inspection moment that obtain the current cache data not downloading life cycle comprise:

Carrying out life cycle to source station and upgrade verification, if content is updated, then record upgrades the renewable time verified and upgrades the verification moment as first;

The verification moment is upgraded, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment according to described first.

3. data cached storage means as claimed in claim 2, it is characterised in that, according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is described data cached sequence number, and y upgrades the verification moment in described data cached the first renewal verification moment or the 2nd, and a, b, c are parameter.

4. data cached storage means as claimed in claim 3, it is characterised in that,

Adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Obtaining the parameter value of a, b, c, wherein, m is when this calculates, the described data cached number of times being updated, x_iFor described data cached sequence number, y_iFor described data cached renewal each time verifies the moment.

5. data cached storage means as claimed in claim 1, it is characterised in that, verification moment and the 2nd renewal verification moment is upgraded according to first, the step of the life cycle obtaining current cache data comprises:

The life cycle of the time period between verification moment and the first renewal verification moment as current cache data is upgraded using the 2nd.

6. data cached storage means as claimed in claim 1, it is characterised in that, according to described life cycle, the step that current cache data carry out classification storage comprises:

If described life cycle is in the default first renewal time interval, then by current cache data stored in one-level storage zone;

If described life cycle is in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

If described life cycle is in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

7. data cached storage means as claimed in claim 1, it is characterised in that, also comprise:

If current cache data are not updated, then current cache data are carried out classification storage by the life cycle of the current cache data calculated according to the last time.

8. a data cached storing device, it is characterised in that, comprising:

Acquisition module, upgrades the inspection moment for obtaining the first of the current cache data not downloading life cycle and upgrade the verification moment and the 2nd, and wherein, described first upgrades the verification moment, and upgrade the verification moment with the 2nd be that adjacent twice renewal verifies the moment;

Calculate module, for upgrading verification moment and the 2nd renewal verification moment according to first, obtain the life cycle of current cache data;

Memory module, for according to described life cycle, carrying out classification storage to current cache data.

9. data cached storing device as claimed in claim 8, it is characterised in that, described acquisition module comprises:

Record cell, upgrades verification for carrying out life cycle to source station, if content is updated, then record upgrades the renewable time verified and upgrades the verification moment as first;

Projected unit, for upgrading the verification moment according to described first, it may also be useful to dynamically least square conic fitting method calculates and obtained for the 2nd renewal verification moment.

10. data cached storing device as claimed in claim 9, it is characterised in that, described projected unit is according to formula: y=ax²+ bx+c, calculates and obtained for the 2nd renewal verification moment, wherein, x is described data cached sequence number, and y upgrades the verification moment in described data cached the first renewal verification moment or the 2nd, and a, b, c are parameter.

11. data cached storing devices as claimed in claim 10, it is characterised in that, described projected unit is further

Adopt following calculation formula: $\left[\begin{array}{ccc}m& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^2\\ \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x}_i& \mathrm{\Σ}{x_i}^3\\ \mathrm{\Σ}{x_i}^2& \mathrm{\Σ}{x_i}^3& \mathrm{\Σ}{x_i}^4\end{array}\right]\left[\begin{array}{c}c\\ b\\ a\end{array}\right]=\left[\begin{array}{c}\mathrm{\Σ}{y}_i\\ \mathrm{\Σ}{x}_i{y}_i\\ \mathrm{\Σ}{x_i}^2{y}_i\end{array}\right],$

Obtaining the parameter value of a, b, c, wherein, m is when this calculates, the described data cached number of times being updated, x_iFor described data cached sequence number, y_iFor described data cached renewal each time verifies the moment.

12. data cached storing devices as claimed in claim 8, it is characterised in that, described calculating module is further used for: upgrade the life cycle of the time period between verification moment and the first renewal verification moment as current cache data using the 2nd.

13. data cached storing devices as claimed in claim 8, it is characterised in that, described memory module comprises:

First storage unit, if for described life cycle in the default first renewal time interval, then by current cache data stored in one-level storage zone;

2nd storage unit, if for described life cycle in the default the 2nd renewal time interval, then by current cache data stored in secondary storage area;

3rd storage unit, if for described life cycle in the default the 3rd renewal time interval, then by current cache data stored in tertiary storage district.

14. data cached storing devices as claimed in claim 8, it is characterized in that, be not updated if described memory module is also further used for current cache data, then current cache data are carried out classification storage by the life cycle of the current cache data calculated according to the last time.