WO2020073328A1

WO2020073328A1 - Data processing method and device

Info

Publication number: WO2020073328A1
Application number: PCT/CN2018/110111
Authority: WO
Inventors: 张伟; 刘威
Original assignee: 华为技术有限公司
Priority date: 2018-10-12
Filing date: 2018-10-12
Publication date: 2020-04-16
Also published as: CN112673359A

Abstract

Provided are a data processing method and device, relating to the technical field of computers, and resolving the issue of the prior art in which independent detection mechanisms of hot and cold data streams cause cold data streams to frequently replace hot data streams in a cache. The method comprises: receiving a request instruction, the request instruction carrying address information of data being requested; if the data being requested by the request instruction has not been stored in a cache and the data being requested by the request instruction is a hot data stream, querying, on the basis of the address information of the data being requested, a first information table, to determine in the cache a target address of the data being requested, wherein a first identifier is used to indicate that data corresponding to a second address can be replaced, and the target address is a second address corresponding to any one of the first addresses that has an attribute identified as a first identifier in the first information table; and copying the data being requested by the request instruction to the target address in the cache.

Description

Data processing method and device

Technical field

The embodiments of the present application relate to the field of computer technology, and in particular, to a data processing method and device.

Background technique

The cache memory is used to store frequently used data to facilitate quick access. When Cache is used, a copy of some content in the memory is saved in the Cache. When reading and writing data, the Central Processing Unit (CPU) first To access the Cache, the CPU only accesses the memory when there is no data required by the CPU in the Cache, and transfers the entire data block containing the storage unit to the Cache, so that several subsequent memory accesses can be converted to the Cache access. If the Cache memory is full, it is necessary to use the block information to replace the original block of information in the Cache according to a replacement algorithm / strategy under the control of the replacement control component.

In the existing Cache replacement strategy, the detection of hot data streams (data that is frequently accessed) and cold data streams (data that is not frequently accessed) are independent of each other, and each has a set of detection mechanisms, which consumes more resources, and This independent detection mechanism will often cause cold data streams to replace hot data streams in the Cache, reducing the efficiency of data processing.

Summary of the invention

Embodiments of the present application provide a data processing method and device, which can associate a detection mechanism of a hot data stream and a cold data stream, and the two cooperate with each other, which can effectively reduce the probability of the cold data stream replacing the hot data stream in the cache.

To achieve the above purpose, the embodiments of the present application adopt the following technical solutions:

A first aspect of an embodiment of the present application provides a data processing method. The method includes: receiving a request instruction, where the request instruction carries address information of the requested data; and determining whether the data requested by the request instruction has been saved in the cache memory In the Cache, and to determine whether the data requested by the request instruction is a hot data stream, if it is determined that the data requested by the request instruction is not stored in the above Cache, and the data requested by the request instruction is a hot data stream, according to the address of the requested data Information, query the first information table to determine the target address of the requested data in the cache, where the hot data stream refers to data whose access frequency is greater than or equal to a preset threshold; the first information table includes one or more first Address, the first address corresponds one-to-one with the second address in the cache, the first information table further includes an attribute identifier corresponding to each first address, the attribute identifier includes a first identifier and a second identifier, the first The identifier is used to indicate that the data at the corresponding second address can be replaced, and the second identifier is used to indicate the corresponding second The data at the address cannot be replaced. The target address is the second address corresponding to any first address whose attribute identifier is the first identifier in the first information table; copy the data requested by the request instruction to the Cache. On the above target address. Based on this solution, by identifying the second address corresponding to any first address of the first identifier in the attribute of the first information table, it is determined as the target address of the hot data stream not stored in the cache, so that the When data is replaced, only the data that can be replaced will be replaced, and the data that cannot be replaced will not be replaced, so the probability of the cold data stream replacing the hot data stream in the Cache can be reduced.

With reference to the first aspect, in a possible implementation manner, the method further includes: if the data requested by the request instruction is successfully copied to the target address in the cache, the first information table corresponds to the target address The attribute identifier of the first address is reset to the second identifier. Based on this solution, by successfully copying the hot data stream not hit in the Cache to the Cache, and resetting the first address attribute identifier corresponding to the target address in the first information table to the second identifier, the request data (Hot data stream) The data identified as non-replaceable data can effectively degrade the probability that the cold data stream replaces the hot data stream.

With reference to the first aspect and the foregoing possible implementation manner, in another possible implementation manner, the address information of the request data includes row address information, and the first information table includes one or more rows, each of which includes one Or a plurality of first addresses, querying the first information table according to the address information of the request data, and determining the target address of the request data, including: searching the row indicated by the row address information in the first information table; in the row In the row indicated by the address information, select any one of the attribute addresses as the first address of the first identifier, and determine that the second address in the Cache corresponding to the first address is the target address. Based on this solution, the first address that identifies any attribute in the row indicated by the row address information as the first identifier can be determined as the target address, so that when the data in the cache is replaced, only the data that can be replaced is replaced, The data that cannot be replaced will not be replaced, and the efficiency of data processing can be improved while reducing the probability of the cold data stream replacing the hot data stream in the cache.

With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, the method further includes: if it is determined that the data requested by the request instruction is stored in the cache, and the data requested by the request instruction is hot data Search for the second address of the requested data in the cache; reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier. Based on this solution, by resetting the first address in the first information table corresponding to the second address of the hot data stream in the Cache to the second identifier, the hot data stream is identified as data that cannot be replaced, It can reduce the probability that the cold data stream replaces the hot data stream in the cache.

With reference to the first aspect and the foregoing possible implementation manner, in another possible implementation manner, the method further includes: if it is determined that the data requested by the request instruction is a cold data stream, according to the address information of the request data, the The second information table replaces the address information in any third address with the address information of the above-mentioned request data; wherein, the cold data stream refers to data whose access frequency is less than the preset threshold; the second information table contains one or more A third address, the third address includes address information. Based on this solution, the address information in any third address in the second information table can be replaced with the address information of the request data, so that when the access frequency of the request instruction is high, the data requested by the request instruction can be detected as hot data flow.

With reference to the first aspect and the foregoing possible implementation manner, in another possible implementation manner, the second information table includes one or more rows, and each row includes one or more third addresses. For the address information of the data, replace the address information in any third address with the address information of the request data in the second information table, including: according to the address information of the request data, replace the above information in the second information table The address information in any third address in the line indicated by the line address information is replaced with the address information of the above-mentioned request data. Based on this solution, while improving the efficiency of data processing, it can be ensured that when the access frequency of the request instruction is high, the data requested by the request instruction can be detected as a hot data stream.

With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, if the attribute identifiers of any row in the first information table are all second identifiers, the rows in which all are second identifiers All attribute identifiers are reset to the first identifier. Based on this solution, when the attribute identifier of a row in the first information table is all the second identifier, the attribute identifier of the row can be reset to the first identifier, and a new round of detection can be restarted to improve the efficiency of the cache Sex.

With reference to the first aspect and the above possible implementation manner, in another possible implementation manner, the above method further includes: searching for the row indicated by the above row address information in the second information table according to the address information of the above request data; if the The address information of the requested data is the same as the address information included in any third address in the row indicated by the row address information in the second information table, and it is determined that the data requested by the request instruction is a hot data stream; if the address of the requested data The information is different from the address information included in all the third addresses in the row indicated by the row address information in the second information table, and it is determined that the data requested by the request instruction is a cold data stream. Based on this solution, it can be determined according to the second information table that the data requested by the request instruction is a hot data stream or a cold data stream.

A second aspect of an embodiment of the present application provides a chip including a processor, a memory controller, and a cache memory, wherein the processor is used to send a request instruction, and the request instruction carries the address of the requested data Information; the memory controller, used to receive the request command sent by the processor; determine whether the data requested by the request command has been saved in the cache memory, and determine whether the data requested by the request command is a hot data stream If it is determined that the data requested by the request instruction is not stored in the Cache, and the data requested by the request instruction is a hot data stream, query the first information table according to the address information of the request data to determine the requested data in the Cache Target address, where the hot data stream refers to data with an access frequency greater than or equal to a preset threshold; the first information table includes one or more first addresses, and the first addresses correspond one-to-one with the second addresses in the cache , The above-mentioned first information table further includes an attribute identifier corresponding to each first address, and the attribute identifier The identification includes a first identification and a second identification. The first identification is used to indicate that the data at the corresponding second address can be replaced, and the second identification is used to indicate that the data at the corresponding second address cannot be replaced. The target address is a second address corresponding to any first address whose attribute identifier is the first identifier in the first information table; copy the data requested by the request instruction to the target address in the cache.

With reference to the second aspect, in a possible implementation manner, the memory controller is further configured to, if the data requested by the request instruction is successfully copied to the target address in the cache, add the first information table to the The attribute identifier of the first address corresponding to the target address is reset to the second identifier.

With reference to the second aspect and the foregoing possible implementation manner, in another possible implementation manner, the address information of the request data includes row address information, and the first information table includes one or more rows, and each row includes one Or a plurality of first addresses, the memory controller is also used to search for the row indicated by the row address information in the first information table; select any attribute identifier as the first identifier in the row indicated by the row address information For the first address, determine that the second address in the Cache corresponding to the first address is the target address.

With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, the memory controller is further configured to determine that the data requested by the request instruction is stored in the cache and the data requested by the request instruction For hot data flow, search for the second address of the requested data in the Cache; reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier.

With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, the memory controller is further configured to determine that the data requested by the request instruction is a cold data stream, based on the address information of the request data, In the second information table, replace the address information in any third address with the address information of the request data; wherein, the cold data stream refers to data whose access frequency is less than the preset threshold; the second information table contains One or more third addresses, the third address including address information.

With reference to the second aspect and the foregoing possible implementation manner, in another possible implementation manner, the memory controller is specifically configured to indicate the row address information in the second information table according to the address information of the request data The address information in any third address in the row of is replaced with the address information of the above request data.

With reference to the second aspect and the foregoing possible implementation manner, in another possible implementation manner, the foregoing memory controller is further configured to, if the attribute identifiers of any row in the first information table are all second identifiers, Reset all attribute identifiers in the row of the second identifier to the first identifier.

With reference to the second aspect and the foregoing possible implementation manner, in another possible implementation manner, the foregoing memory controller is further configured to search for the row indicated by the row address information in the second information table according to the address information of the request data If the address information of the requested data is the same as the address information included in any third address in the row indicated by the row address information in the second information table, it is determined that the data requested by the request instruction is a hot data stream; if the above request The address information of the data is different from the address information included in all the third addresses in the row indicated by the row address information in the second information table, and it is determined that the data requested by the request instruction is a cold data stream.

For the description of the second aspect and the effects of various implementation manners of the second aspect, reference may be made to the description of the corresponding effects of the first aspect, and details are not described herein again.

In a third aspect of the embodiments of the present application, a server is provided. The structure of the server includes a processor and a memory. The memory is used to couple with the processor to store necessary program instructions and data of the server. The processor is used to execute The program instructions stored in the memory cause the server to execute the above method.

According to a fourth aspect of the embodiments of the present application, a computer storage medium is provided, in which computer program code is stored, and when the computer program code runs on a processor, the processor is caused to execute the first aspect Or the data processing method described in any one of the possible implementation manners of the first aspect.

A fifth aspect of the embodiments of the present application provides a computer program product that stores computer software instructions executed by the processor, and the computer software instructions include a program for executing the solution described in the above aspect.

According to a sixth aspect of the embodiments of the present application, an apparatus is provided. The apparatus exists in the form of a chip product. The structure of the apparatus includes a processor and a memory. The memory is used to couple with the processor to store necessary programs of the apparatus Instruction and data, the processor is used to execute the program instructions stored in the memory, so that the device performs the function of the data processing device in the above method.

BRIEF DESCRIPTION

FIG. 1 is an architectural diagram of a computer system provided by an embodiment of the present application;

2 is a flowchart of a data processing method provided by an embodiment of the present application;

3 is a schematic diagram of a second information table provided by an embodiment of this application;

4 is a schematic diagram of a first information table provided by an embodiment of this application;

5 is an application schematic diagram of a data processing method provided by an embodiment of the present application;

6 is a flowchart of another data processing method provided by an embodiment of the present application;

7 is an application schematic diagram of another data processing method provided by an embodiment of the present application;

8 is a flowchart of another data processing method provided by an embodiment of the present application;

9 is a flowchart of another data processing method provided by an embodiment of the present application;

10 is a schematic diagram of application of another data processing method provided by an embodiment of the present application;

11 is a flowchart of another data processing method provided by an embodiment of this application;

12 is an application schematic diagram of another data processing method provided by an embodiment of the present application;

13 is a schematic diagram of the composition of a data processing device provided by an embodiment of the present application;

14 is a schematic diagram of the composition of another data processing device provided by an embodiment of the present application;

15 is a schematic diagram of the composition of a chip provided by an embodiment of the present application.

detailed description

An embodiment of the present application provides a data processing method, which can be applied to the computer system 100 shown in FIG. 1. The computer system shown in FIG. 1 may include a processor 101, a memory controller 102, and a first-level memory 103之间的 Secondary storage 104.

The processor 101 is a control center of the computer system 100, and is used to run an operating system of the computer system 100 and applications (including system applications and third-party applications) on the computer 100.

In the embodiments of the present application, the processor 101 may specifically be a central processing unit (CPU), which may implement or execute various exemplary logical blocks, modules, and modules described in conjunction with the contents disclosed in the embodiments of the present application. The circuit; the processor may also be a combination that realizes a calculation function, for example, including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc. This embodiment of the present application does not limit this.

The memory controller 102 is used to manage data in the first-level storage and the second-level storage. Exemplarily, the memory controller may be a cache controller, which is used to manage data in the cache.

The first-level memory 103 is used to cache data in the second-level memory, and is also used to support cache access. Exemplarily, the first-level memory 103 may be a cache memory Cache between the processor and the memory, which is used to improve the read and write performance of the system. The data in the cache can be part of the data in the memory. Further, if the Cache contains data to be accessed, the processor 101 can obtain the data to be accessed from the Cache instead of obtaining the data to be accessed from the memory, thereby speeding up the reading speed. It should be noted that, in the following embodiments of the present application, the first-level memory is used as the cache memory Cache as an example for description.

The second level memory 104 is used to store all instructions and data. Exemplarily, the second-level memory 104 may be the memory of the computer system 100.

In the embodiment of the present application, the second-level memory 104 may specifically include volatile memory (volatile memory), such as random-access memory (RAM); the memory may also include non-volatile memory (non-volatile memory). volatile memory), such as a read-only memory (Read Only Memory, ROM); the memory may also include a combination of the above types of memory.

It is understandable that FIG. 1 is only an exemplary description. In practical applications, the computer system 100 may include more or fewer components than those shown in FIG. 1, and the structure shown in FIG. 1 does not constitute the computer system provided by the embodiments of the present application. Any restrictions.

In order to solve the problem that the detection mechanism of the hot data stream and the cold data stream are independent of each other in the background technology, the situation that the cold data stream replaces the hot data stream in the Cache often occurs, which reduces the efficiency of data processing. A data processing method. This method can associate the detection mechanism of the hot data stream and the cold data stream. The two cooperate with each other to reduce the probability that the cold data stream replaces the hot data stream in the cache.

With reference to FIG. 1 and FIG. 2, the data processing method provided in this embodiment of the present application may include steps S201-S203.

S201. The memory controller receives the request instruction.

Wherein, the above request instruction carries the address information of the request data, and the address of the request data is a logical address.

Exemplarily, the memory controller may receive a request instruction sent by the processor, and the request instruction may be any access request sent by the processor.

Exemplarily, the above request instruction may be a read request or a write request sent by the processor, and the read request or the write request is used to read or write data in the first-level memory and / or the second-level memory.

Exemplarily, the address information of the request data carried in the above request instruction may include row address information, and the row address information may be row address information divided according to a preset rule, for example, the attributes of the data may be classified into rows. This application The embodiment does not limit this, and here is only an exemplary description. Exemplarily, the row address information may be recorded as RowID.

S202. If it is determined that the data requested by the request instruction is not stored in the cache and the data requested by the request instruction is a hot data stream, query the first information table according to the address information of the requested data to determine the target address of the requested data in the cache.

The hot data stream refers to data whose access frequency is greater than or equal to a preset threshold, that is, data that the processor frequently accesses. The specific value of the preset threshold is not limited in this embodiment of the present application, and can be determined according to actual applications. .

Exemplarily, the method for determining whether the data requested by the request instruction is a hot data stream in the embodiment of the present application may be implemented in various ways, for example, the number of accesses of the requested data within a preset time may be compared with the preset number of accesses, If the number of accesses to the requested data within a preset time is greater than or equal to the preset number of accesses, the requested data can be determined to be a hot data stream; for another example, the access frequency of the requested data can be compared with the preset threshold by counting the access frequency of the requested data For comparison, if the access frequency of the requested data is greater than or equal to the preset threshold, the requested data can be determined to be a hot data stream; for another example, the address information of the requested data in the second information table can be found by searching for the address information of the requested data It matches the address information of any requested data in the second information table, and it can be determined that the requested data is a hot data stream. The embodiment of the present application does not limit the specific method for determining whether the requested data is a hot data stream. It is an example.

Exemplarily, the second information table may include one or more third addresses, and each third address includes address information. The above-mentioned determination of whether the data requested by the request instruction is a hot data stream by searching the second information table may specifically include : Search in the second information table according to the address information of the requested data. If the address information of the requested data is the same as the address information included in any third address in the second information table, determine that the data requested by the request instruction is a hot data stream; If the address information of the requested data is different from the address information included in all the third addresses in the second information table, it is determined that the data requested by the request instruction is a cold data stream.

Exemplarily, the second information table may include one or more rows, and each row includes one or more third addresses. As shown in FIG. 3, the second information table includes multiple rows, and each row includes 8 pieces of information. Each information space records address information corresponding to 8 addresses (third address). The third address marked as black in FIG. 3 stores the address information of the requested data, and the third address marked as white Store empty address information. The embodiment of the present application does not limit the specific specifications of the second information table. Here, only eight information spaces are included in each row as an example for description.

Further, in this embodiment of the present application, the address information of the requested data can be matched with the address information included in the third address in the second information table. If the address information of the requested data includes row address information, and the address information of the requested data is Any third address in the row indicated by the row of address information in the second information table includes the same address information. It can be considered that the data requested by the request instruction is data frequently accessed by the processor, and the data requested by the request instruction is determined to be hot Data flow; if the address information of the requested data is different from the address information included in all the third addresses in the row indicated by the row of address information in the second information table, it can be considered that the data requested by the request instruction is not frequently accessed by the processor Data, it is determined that the data requested by the request instruction is a cold data stream.

Exemplarily, the above target address refers to the target address in the cache of the data requested by the request instruction. It can be understood that the data requested by the request instruction and the requested data in the embodiment of the present application are the same data.

Exemplarily, the above-mentioned first information table is used to record the currently available Cache resources. The first information table may include one or more first addresses, and the first addresses correspond one-to-one with the second addresses in the Cache. The first information table may further include an attribute identifier corresponding to each first address. The attribute identifier includes a first identifier and a second identifier. The first identifier is used to indicate that data on the corresponding second address can be replaced. The second identifier is used to indicate that the data at the corresponding second address cannot be replaced.

Exemplarily, the target address of the foregoing request data may be a second address corresponding to any first address whose attribute identifier is the first identifier in the first information table. Understandably, the second address corresponding to the first address of any attribute identified as the first identification in the first information table may be determined as the target address, because the first address whose attribute is identified as the first identification corresponds to the first The data stored at the second address is data that can be replaced, so the data stored at the target address of the requested data is data that can be replaced, and the requested data can be stored at the target address to replace the original data at the target address .

Further, the first information table may include one or more rows, and each row includes one or more first addresses. As shown in FIG. 4, each row includes 8 information spaces, and each information space records Attribute identifiers corresponding to 8 addresses (first addresses), the attribute identifier corresponding to each first address may be the first identifier or the second identifier, the first identifier may be recorded as 0, and the second identifier may be recorded as 1. The attribute ID of 0 means that the data of the first address in the corresponding second address in the Cache can be replaced, and the attribute ID of 1 means that the data of the first address in the corresponding second address in the Cache cannot be replaced Where the first address corresponding to the hot data stream in the embodiment of the present application in the first information table may be identified as 1. The embodiment of the present application does not limit the specifications of the first information table and the specific expression forms of the first identifier and the second identifier. Here, only each row in the first information table includes 8 information spaces, and the first identifier is recorded as 0, the second mark is 1 as an example for description.

Exemplarily, the target address of the foregoing request data may be a row indicated by row address information in the first information table, and the attribute identifier is a second address corresponding to any first address of the first identifier. Understandably, the second address corresponding to the first address whose first attribute is identified as the first identification in the row corresponding to the above-mentioned row address information in the first information table may be determined as the target address, since the first attribute identified as 0 The data stored at the second address corresponding to the address in the cache is data that can be replaced, so the data stored at the target address of the requested data is data that can be replaced.

Exemplarily, the above querying the first information table according to the address information of the requested data to determine the target address of the requested data in the cache may include: searching the first information table for the row indicated by the row address information, where the row address information In the indicated line, select any one of the attribute addresses as the first address of the first identifier, and determine that the second address in the Cache corresponding to the first address is the target address.

For example, as shown in FIG. 5, according to the row address RowID carried in the request instruction, if the row address is 2, in the row indicated by the row address RowID 2, randomly select any first address in the row whose attribute ID is 0 (Location), as shown in Fig. 5, there are 5 first addresses with attribute identification 0, and a first address (location) can be randomly selected from the 5 first addresses (location) with attribute identification 0, as shown in Fig. 5 The rowID indicates a first address with a column ID of 1 in the row indicated by 2 (the position identified as 0 in the dotted frame), and the second address corresponding to the first address in the cache is determined as the target address. Since the data of the first address with the attribute ID of 0 at the corresponding second address in the Cache can be replaced, the data of the first address with the attribute ID of 1 at the corresponding second address of the Cache cannot be replaced, so this In the embodiment of the application, by selecting the second address corresponding to any one of the first addresses with the attribute identifier of 0 as the target address, it can be ensured that the data stored on the selected target address is data that can be replaced, not not Replaced data.

It is understandable that the data of the first address in the first information table with the attribute ID of 0 in the corresponding second address in the Cache is data that can be replaced, and the first address in the first information table with the attribute ID of 1 in the Cache The data at the corresponding second address in the data is data that cannot be replaced, wherein the hot data stream can be identified as 1, and the data processing method in this embodiment of the present application can identify any attribute in the first information table as 0 A second address corresponding to a first address in the Cache is determined to be the target address, that is, the data identified as 0 is replaced, and the data identified as 1 is not replaced. Therefore, when replacing the data in the cache, only the data that can be replaced will be replaced, and the data that is marked as 1 that cannot be replaced will not be replaced, so the data processing method in the embodiment of the present application can reduce the current There is a probability that a cold data stream replaces a hot data stream in the technology.

S203. The memory controller copies the data requested by the request instruction to the target address in the cache.

Exemplarily, the target address is a second address of data that can be replaced in the cache. The data requested by the above request instruction is not stored in the cache and is a hot data stream, so the requested data can be copied from the memory to the target address in the cache. Since the target address is the second address corresponding to any first address whose attribute identifier is the first identifier in the first information table, when copying the requested data from the memory to the target address on the cache, the target address The previously stored data that can be replaced is replaced with the request data, and during the replacement, only the data that can be replaced will be replaced, and the data that is identified as 1 that cannot be replaced will not be replaced.

Exemplarily, before step S203, it may further include: the memory controller obtains the requested data from the memory according to the address information of the requested data.

In the data processing method provided in the embodiment of the present application, by receiving a request instruction, the request instruction carries the address information of the requested data; if it is determined that the data requested by the request instruction is not stored in the cache, and the data requested by the request instruction is a hot data stream, According to the address information of the requested data, query the first information table to determine the target address of the requested data in the cache, where the target address is the second corresponding to any one of the first addresses whose attribute identifier is the first identifier in the first information table Address; copy the data requested by the request instruction to the target address in the Cache. In the embodiment of the present application, by identifying the attribute in the first information table as the second address corresponding to any first address of the first identifier, the target address of the hot data stream not hit in the cache is determined, so that the When data is replaced, only the data that can be replaced will be replaced, and the data that cannot be replaced will not be replaced. Therefore, the probability of the cold data stream replacing the hot data stream in the cache can be effectively reduced.

This application also provides an embodiment. As shown in FIG. 6, if step S203 successfully copies the requested data to the cache, the above method further includes step S204; if the requested data is not successfully copied to the chip, a failure message is returned.

S204. The memory controller resets the attribute identifier of the first address corresponding to the target address in the first information table to the second identifier.

Exemplarily, as shown in FIG. 7, after the request data is successfully copied to the Cache, the attribute identifier of the first address corresponding to the target address in the first information table is reset to 1, so the request data (hot data stream ) The attribute identifier of the corresponding first address in the first information table is a second identifier that cannot be replaced.

Exemplarily, the request data reset to the second identifier is data that cannot be replaced. Therefore, when the data requested by the next request instruction is a hot data stream and is not hit in the cache, the above cannot be reset to The requested data of the second identifier is replaced, thereby reducing the probability that the cold data stream replaces the hot data stream in the cache.

In this way, the second information table can be used to determine whether the data requested by the request instruction is a hot data stream or a cold data stream, the attribute identifier in the first information table can be used to determine whether the resources in the cache can be replaced, and the request instruction Is not stored in the cache and is a hot data stream, after successfully copying the data requested by the request instruction to the second address (target address) in the cache that can be replaced, the first information table corresponds to the target address The attribute identifier of the first address is reset to a second identifier that cannot be replaced, so that the detection mechanism of the cold data stream and the hot data stream are unified and coordinated; and because the embodiment of the present application replaces the data in the cache, only The data with the identifier 0 can be replaced, but the data with the identifier 1 cannot be replaced. Therefore, the probability that the cold data stream can replace the hot data stream can be effectively reduced.

In the data processing method provided by the embodiment of the present application, by receiving a request instruction, the request instruction carries the address information of the requested data; if it is determined that the data requested by the request instruction is not stored in the cache, and the requested data is a hot data stream According to the address information of the requested data, query the first information table to determine the target address of the requested data in the cache, where the target address is the second corresponding to any one of the first addresses whose attribute identifier is the first identifier in the first information table Address; copy the data requested by the request instruction to the target address in the Cache; reset the attribute identifier of the first address corresponding to the target address in the first information table to the second identifier. In the embodiment of the present application, the attribute in the first information table is identified as the second address corresponding to any one of the first addresses in the first identifier, and is determined as the target address of the hot data stream not hit in the cache. When data is replaced, only the data that can be replaced will be replaced, and the data that cannot be replaced will not be replaced. Therefore, the probability of the cold data stream replacing the hot data stream in the cache can be effectively reduced.

This application also provides an embodiment. As shown in FIG. 8, the data processing method further includes S205-S206.

S205. If it is determined that the data requested by the request instruction is stored in the cache, and the data requested by the request instruction is a hot data stream, search for the second address of the requested data in the cache.

Exemplarily, in the case that the data requested by the request instruction is stored in the cache and is a hot data stream, the second address of the request data in the cache may be found according to the address information of the request data carried in the request instruction, the The second address corresponds to the first address in the first information table.

S206. The memory controller resets the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier.

Exemplarily, when the request data is a hot data stream and hits in the Cache, the first address corresponding to the second address may be searched in the first information table according to the second address of the request data in the Cache, and The first address is reset to the second identification. Exemplarily, before resetting the attribute identifier of the first address corresponding to the second address in the first information table, the attribute identifier of the first address corresponding to the second address may be the first identifier or the second ID, regardless of whether the attribute ID of the first address corresponding to the target address before resetting is the first ID or the second ID, the attribute ID needs to be reset to the second ID.

Understandably, after resetting the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier, the request data is data that cannot be replaced.

Exemplarily, according to the second address of the request data in the cache, the attribute identifier of the first address corresponding to the second address is set to 1 in the first information table, and the attribute identifier is 1 indicating that the second address is stored on the second address The data cannot be replaced. Therefore, when the data requested by the next request instruction is a hot data stream and is stored in the cache, the above-mentioned request data reset to the second identification cannot be replaced.

In this way, the second information table can be used to determine whether the data requested by the request instruction is a hot data stream or a cold data stream, and the attribute identifier in the first information table can be used to determine whether the resources in the Cache can be replaced, and then determine the request instruction request When the data is stored in the cache and is a hot data stream, the attribute identifier of the corresponding address in the first information table can be reset to a second identifier that cannot be replaced, so that the detection mechanism of the cold data stream and the hot data stream is unified Coordination; and because the embodiments of the present application replace the data in the Cache, only the data with the identifier 0 can be replaced, and the data with the identifier 1 can not be replaced, so it can be effectively downgraded The probability of a cold data stream replacing a hot data stream.

In the data processing method provided by the embodiment of the present application, by receiving the request instruction, the request instruction carries the address information of the requested data; if it is determined that the data requested by the request instruction is stored in the Cache and the requested data is a hot data stream, search for Request the second address of the data in the Cache; reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier. In the embodiment of the present application, by resetting the first address in the first information table corresponding to the second address of the hot data stream in the cache to the second identifier, the data in the cache can be replaced without replacing the first address The data stream whose attribute identifier is the second identifier in the information table is replaced, so the probability of the cold data stream replacing the hot data stream in the cache can be effectively reduced.

This application also provides an embodiment. As shown in FIG. 9, the data processing method further includes S207.

S207. If it is determined that the data requested by the request instruction is a cold data stream, the memory controller replaces the address information in any third address in the second information table with the address information of the requested data in the second information table.

The cold data stream refers to data whose access frequency is less than a preset threshold, that is, data that the processor does not access frequently (less access times), and the specific value of the preset threshold is not limited in the embodiment of the present application. It can be determined according to the actual application.

Exemplarily, replacing the address information in any third address with the address information of the request data in the second information table according to the address information of the request data may include: according to the address information of the request data, in the second In the information table, the address information in any third address in the row indicated by the row address information is replaced with the address information of the requested data. For example, as shown in FIG. 10, if it is determined that the data requested by the request instruction is a cold data stream (infrequently accessed data), any third address in the row indicated by the row address information may be replaced with the address information of the requested data , So that when the access frequency of the request instruction is relatively high, the data requested by the request instruction can be detected as a hot data stream.

Exemplarily, replacing the instruction address information included in any third address in the row indicated by the row address information with the instruction address information of the request data may include: if any third address in the row indicated by the row address information If the address information of the requested data is stored, the address information of the requested data stored in the third address can be replaced with the address information of the data requested by the request instruction; if any one of the rows indicated by the address information of the row is third The address stores empty address information, and the empty address information stored in the third address can be replaced with the address information of the data requested by the request instruction.

Exemplarily, after replacing the instruction address information included in any third address in the row indicated by the row address information with the instruction address information of the request data, if the request instruction initiates a request again, the data requested by the request instruction is not saved In the Cache and is a hot data stream, the request data can be processed in steps S201-S204; if the data requested by the request instruction is successfully copied to the Cache, when the request instruction initiates a request again, the request instruction requests the The data is stored in the cache and is a hot data stream. The data processing method of steps S205-S206 can be used to reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier. During the data processing, the probability of the cold data stream replacing the hot data stream can be reduced.

In the data processing method provided in the embodiment of the present application, by receiving a request instruction, the request instruction carries address information of the requested data, and the address information of the requested data includes row address information; if it is determined that the data requested by the request instruction is a cold data stream, according to the request For the address information of the data, replace the address information in any third address in the row indicated by the row address information with the address information of the requested data in the second information table. In the embodiment of the present application, by replacing the address information in any third address in the row indicated by the row address information in the second information table with the address information of the requested data, when the request instruction is initiated again, the data requested by the request instruction can be Is detected as a hot data stream and misses in the cache, and then the attribute in the first information table is identified as the second address corresponding to any first address in the first information table according to step S202, and is determined as the data requested by the request instruction Target address, so that when replacing the data in the Cache, only the data that can be replaced will be replaced, and the data that cannot be replaced will not be replaced. Therefore, the situation of replacing the cold data flow with the hot data flow in the Cache can be effectively reduced The probability of occurrence.

This application also provides an embodiment. As shown in FIG. 11, the data processing method further includes S208.

S208. If the attribute identifiers of any row in the first information table are all second identifiers, the memory controller resets all attribute identifiers in the rows that are all second identifiers to the first identifier.

Exemplarily, as shown in FIG. 12, if the attribute identifiers of any row in the first information table are all 1, then reset the row of all 1s to 0, and then start a new round of detection and processing.

Exemplarily, if the attribute identifier of any row in the first information table is all the second identifier, it can be determined that the data of the corresponding second address in the cache of each first address in the row cannot be replaced. The data corresponding to the stored row will become fixed data. In this way, if the data accessed by the processor is no longer the hot data stream stored in the current Cache, then the processor will need to continuously read data from memory when accessing The efficiency of Cache will not be used. Therefore, when the attribute identifiers of a row in the first information table are all the second identifiers, all the attribute identifiers of the row can be reset to the first identifier, so that the data stored in the Cache is replaced and a new one is restarted Wheel detection.

In the data processing method provided by the embodiment of the present application, by receiving a request instruction, the request instruction carries the address information of the requested data, and the address information of the requested data includes the row address information; if it is determined that the data requested by the request instruction is not stored in the cache, and The data requested by the request instruction is a hot data stream. According to the address information of the requested data, the first information table is queried to determine the target address of the requested data in the Cache, where the target address is the attribute identifier in the first information table as the first identifier The second address corresponding to any one of the first addresses; when the attribute identifiers of any row in the first information table are all the second identifiers, the memory controller resets all attribute identifiers in the rows that are all the second identifiers to The first logo. In the embodiment of the present application, the attribute in the first information table is identified as the second address corresponding to any one of the first addresses in the first identifier, and is determined as the target address of the hot data stream not hit in the cache. When the data is replaced, only the data that can be replaced will be replaced, and the data that cannot be replaced will not be replaced. Therefore, the probability of the cold data flow replacing the hot data flow in the Cache can be effectively reduced; When the attribute identifier of a row in the table is all the second identifier, reset the attribute identifier of the row to the first identifier, start a new round of detection, and improve the efficiency of the cache.

The above mainly introduces the solutions provided by the embodiments of the present invention from the perspective of method steps. It can be understood that, in order to realize the above-mentioned functions, the computer includes a hardware structure and / or a software module corresponding to each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the present application can be implemented in a combination of hardware and computer software. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

In the embodiments of the present application, the computer may be divided into function modules according to the above method examples. For example, each function module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or software function modules. It should be noted that the division of the modules in the embodiment of the present invention is schematic, and is only a division of logical functions. In actual implementation, there may be another division manner.

In the case where each functional module is divided corresponding to each function, FIG. 13 shows a possible structural schematic diagram of the data processing device involved in the above embodiment. The data processing device 1300 includes: a receiving unit 1301 and a processing unit 1302和 RESET 单元 1303。 And reset unit 1303. The receiving unit 1301 is used to support the data processing device 1300 to execute S201 in FIG. 2; the processing unit 1302 is used to support the data processing device 1300 to execute S202-S203 in FIG. 2, or S205 in FIG. 8, or S207 in FIG. 9. The reset unit 1303 is used to support the data processing device 1300 to execute S204 in FIG. 6, or S206 in FIG. 8, or S208 in FIG. 11. Wherein, all relevant content of each step involved in the above method embodiments can be referred to the function description of the corresponding function module, which will not be repeated here. With reference to FIG. 1, the above data processing device 1300 may be the computer system 100 shown in FIG. 1, and at least one unit in the data processing device 1300 may be implemented in software, hardware, or a combination of software and hardware.

In the case of using an integrated unit, FIG. 14 shows a possible structural schematic diagram of the data processing apparatus 1400 involved in the foregoing embodiment. The data processing device 1400 includes a memory 1401 and a processor 1402. The processor 1402 is used to control and manage the operation of the data processing device. For example, the processor 1402 is used to support the data processing device to execute S201-S203 in FIG. 2, S201-S204 in FIG. 6, and S201-S206 in FIG. 8. , S201-S207 in FIG. 9, S201-S208 in FIG. 11, and / or other processes used in the techniques described herein. The memory 1401 is used to store program codes and data corresponding to the data processing method performed by the data processing device 1400 according to any of the data processing methods provided above. The data processing device 1400 may be the computer system 100 shown in FIG. 1. The description of all relevant contents of the components involved in FIG. 1 above may be referred to the functional description of the corresponding component in FIG. 14, which will not be repeated here.

An embodiment of the present application further provides a chip. As shown in FIG. 15, the chip 1500 includes: a processor 1501, a memory controller 1502, and a cache memory Cache 1503, where the processor 1501 is used to send a request instruction; the memory controller 1502 is used to execute S201-S203 in FIG. 2, S201-S204 in FIG. 6, S201-S206 in FIG. 8, S201-S207 in FIG. 9, S201-S208 in FIG. 11; cache Cache1503 is used in Support cache access to improve the read and write performance of the system. The chip 1500 may be a chipset in the computer system 100 shown in FIG. 1. The description of all relevant contents of the components involved in FIG. 1 above can be referred to the functional description of the corresponding component in FIG. 15, which will not be repeated here.

The steps of the method or algorithm described in conjunction with the disclosure of the present application may be implemented by hardware, or by a processor executing software instructions. Software instructions can be composed of corresponding software modules, which can be stored in random access memory (Random Access Memory, RAM), flash memory, erasable programmable read-only memory (Erasable Programmable ROM, EPROM), electrically erasable Programmably read only memory (Electrically EPROM, EEPROM), registers, hard disk, removable hard disk, CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in the ASIC. In addition, the ASIC may be located in the core network interface device. Of course, the processor and the storage medium may also exist as discrete components in the core network interface device.

Those skilled in the art should be aware that in one or more of the above examples, the functions described in the present invention may be implemented by hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media includes computer storage media and communication media, where communication media includes any medium that facilitates transfer of a computer program from one place to another. The storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

The specific embodiments described above further describe the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the scope of the present invention. The scope of protection, any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention, shall be included in the scope of protection of the present invention.

Claims

A data processing method, characterized in that the method includes:

Receiving a request instruction, where the request instruction carries the address information of the requested data;

Determine whether the data requested by the request instruction has been saved in the cache memory, and determine whether the data requested by the request instruction is a hot data stream, if it is determined that the data requested by the request instruction is not saved in the Cache , And the data requested by the request instruction is a hot data stream, query the first information table according to the address information of the request data, and determine the target address of the requested data in the cache, where the hot data stream refers to Data whose access frequency is greater than or equal to a preset threshold; the first information table includes one or more first addresses, the first address corresponds one-to-one with the second address in the cache, and the first information table further includes An attribute identifier corresponding to each first address. The attribute identifier includes a first identifier and a second identifier. The first identifier is used to indicate that data on the corresponding second address can be replaced. The second identifier is used To indicate that the data on the corresponding second address cannot be replaced, the target address is any first place where the attribute identifier in the first information table is the first identifier Corresponding to the second address;

Copy the data requested by the request instruction to the target address in the cache.
The data processing method according to claim 1, wherein the method further comprises:

If the data requested by the request instruction is successfully copied to the target address in the Cache, the attribute identifier of the first address corresponding to the target address in the first information table is reset to the second identifier.
The data processing method according to claim 1 or 2, wherein the address information of the requested data includes row address information, the first information table includes one or more rows, and each row includes one or more rows A first address, the querying the first information table according to the address information of the request data to determine the target address of the request data in the cache includes:

Searching the row indicated by the row address information in the first information table;

In the row indicated by the row address information, select any one of the first addresses whose attribute identifier is the first identifier, and determine that the second address in the Cache corresponding to the first address is the target address.
The data processing method according to any one of claims 1 to 3, wherein the method further comprises:

If it is determined that the data requested by the request instruction is stored in the cache, and the data requested by the request instruction is a hot data stream, look up the second address of the requested data in the cache;

Reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier.
The data processing method according to any one of claims 1 to 4, wherein the method further comprises:

If it is determined that the data requested by the request instruction is a cold data stream, according to the address information of the requested data, the address information in any third address is replaced with the address information of the requested data in the second information table ; Wherein, the cold data stream refers to data whose access frequency is less than the preset threshold; the second information table includes one or more third addresses, and the third address includes one address information.
The data processing method according to claim 5, wherein the second information table includes one or more rows, each row includes one or more third addresses, and the address according to the request data Information, replacing the address information in any third address with the address information of the requested data in the second information table includes:

According to the address information of the request data, the address information in any third address in the row indicated by the row address information is replaced with the address information of the request data in the second information table.
The data processing method according to any one of claims 1 to 6, characterized in that, if the attribute identifiers of any row in the first information table are all second identifiers, the attributes in the rows that are all second identifiers All attribute identifiers are reset to the first identifier.
The data processing method according to any one of claims 1-7, wherein the determining whether the data requested by the request instruction is a hot data stream includes:

Searching the row indicated by the row address information in the second information table according to the address information of the request data;

If the address information of the request data is the same as the address information included in any third address in the row indicated by the row address information in the second information table, it is determined that the data requested by the request instruction is the hot data flow;

If the address information of the request data is different from the address information included in all the third addresses in the row indicated by the row address information in the second information table, it is determined that the data requested by the request instruction is a cold data stream.
A chip, characterized in that the chip includes a processor, a memory controller and a cache memory Cache, wherein,

The processor is used to send a request instruction, and the request instruction carries address information of the requested data;

The memory controller is configured to receive the request instruction sent by the processor; determine whether the data requested by the request instruction has been saved in the cache memory, and determine whether the data requested by the request instruction is Hot data stream, if it is determined that the data requested by the request instruction is not stored in the cache, and the data requested by the request instruction is a hot data stream, query the first information table according to the address information of the requested data to determine The target address of the requested data in the cache, wherein the hot data stream refers to data whose access frequency is greater than or equal to a preset threshold; the first information table includes one or more first addresses, the first The addresses correspond one-to-one with the second addresses in the cache. The first information table further includes an attribute identifier corresponding to each first address. The attribute identifier includes a first identifier and a second identifier. The first identifier is used In order to indicate that the data at the corresponding second address can be replaced, the second identifier is used to indicate that the data at the corresponding second address cannot be replaced. Identification of the first attribute information table is any one of a first identifier corresponding to the first address a second address; copy of the requested data to the instruction request on the destination address of the Cache.
The chip according to claim 9, characterized in that

The memory controller is further configured to, if the data requested by the request instruction is successfully copied to the target address in the cache, copy the first address corresponding to the target address in the first information table The attribute identification is reset to the second identification.
The chip according to claim 9 or 10, wherein the address information of the request data includes row address information, the first information table includes one or more rows, and each row includes one or more An address,

The memory controller is also used to search for the row indicated by the row address information in the first information table; among the rows indicated by the row address information, select any one of the attribute identifiers as the first Address, it is determined that the second address in the cache corresponding to the first address is the target address.
The chip according to any one of claims 9-11, characterized in that

The memory controller is further configured to, if it is determined that the data requested by the request instruction is stored in the cache, and the data requested by the request instruction is a hot data stream, search for the second address of the requested data in the cache Reset the attribute identifier of the first address corresponding to the second address in the first information table to the second identifier.
The chip according to any one of claims 9-12, characterized in that

The memory controller is further configured to, if it is determined that the data requested by the request instruction is a cold data stream, according to the address information of the requested data, address information in any third address in the second information table Replaced by the address information of the requested data; wherein, the cold data stream refers to data whose access frequency is less than the preset threshold; the second information table includes one or more third addresses, and the third address Include an address information.
The chip according to claim 13, wherein the second information table includes one or more rows, and each row includes one or more third addresses,

The memory controller is specifically configured to replace the address information in any third address in the row indicated by the row address information with the address information of the request data in the second information table by the The address information of the requested data.
The chip according to any one of claims 9 to 14, wherein

The memory controller is further configured to, if the attribute identifiers of any row in the first information table are all second identifiers, reset all the attribute identifiers in the rows that are all second identifiers to the first identifier.
The chip according to any one of claims 9 to 15, wherein

The memory controller is further configured to search for the row indicated by the row address information in the second information table according to the address information of the requested data; if the address information of the requested data is different from that in the second information table Any third address in the row indicated by the address information includes the same address information, and it is determined that the data requested by the request instruction is the hot data stream; if the address information of the requested data is the same as the second information table All third addresses in the row indicated by the row address information in the row include different address information, and it is determined that the data requested by the request instruction is a cold data stream.
A computer storage medium storing computer program code in the computer storage medium, characterized in that when the computer program code runs on a processor, the processor is caused to execute any one of claims 1-8 The data processing method.