TW201901677A - Storage device, recording method and pre-fetch method - Google Patents

Abstract

A storage device includes a flash memory array and a controller. The flash memory array stores a plurality of user data. After an initial process is finished, the controller accesses the user data stored in the flash memory array according to host instructions and an H2F mapping table, and records a plurality of address information about the user data in a power-on access table.

Description

 Storage device and preloading method  

The invention relates to a flash memory device and a preloading method thereof.

When the system is powered up, solid-state hard drives are often powered up directly. However, after the SSD is powered on and the SSD is accessed for the first time, there is a free time. In order to effectively improve the performance of the SSD, it is necessary to use the idle time more efficiently. Increase the access speed of solid state drives.

In view of the above, the present invention provides a storage device coupled to a host, including: a flash memory array, a dynamic random access memory, and a controller. The flash memory array stores a first data. The dynamic random access memory stores a power-on access table, wherein the power-on access table is used to record address information of the first data accessed by the host after the power-on of the storage device. The controller preloads the first data into the dynamic random access memory according to the address information during an idle time of the storage device to improve the access efficiency of the host to the storage device. .

According to an embodiment of the invention, the idle time is a time before the storage device is powered on until the host starts accessing the storage device.

According to an embodiment of the invention, the host performs an initialization by using the first data preloaded into the dynamic random access memory.

According to another embodiment of the present invention, the power-on access table further records a sequence of the first data accessed by the host after the power-on of the storage device, wherein the controller is configured according to the sequence The first data of a predetermined size is preloaded into the dynamic random access memory, and the remaining first data is not preloaded into the dynamic random access memory, wherein the host accesses the dynamic random access memory The body and the flash memory array described above to complete an initialization.

According to an embodiment of the present invention, after the host completes the initialization, the host sends an access command for accessing a second data of the flash memory array, and the controller performs the above according to the access command. The second data is temporarily stored in the dynamic random access memory to improve the access efficiency of the host to the storage device.

The present invention further provides a preloading method for testing a flash memory array, comprising: powering up the flash memory array; and accessing the flash memory array by a host to perform an initialization. The address information of the first data is recorded in a power-on access list; the flash memory array is powered on again; and the first data is from the above information according to the address information of the power-on access table The flash memory array is preloaded with a dynamic random access memory; and the initialization of the host is completed by using the first data.

According to an embodiment of the present invention, the step of preloading the first data from the flash memory array into the dynamic random access memory according to the power-on access table is performed on the flash memory. After the body array is powered up, the host computer begins to access an idle time between the flash memory arrays.

According to an embodiment of the present invention, the power-on access table is further configured to record a sequence of the first data accessed by the host after the previous power-on of the flash memory array, wherein the foregoing The step of preloading the first data from the flash memory array into the dynamic random access memory further includes: preloading the first data of a predetermined size to the first order according to the foregoing sequence And the dynamic random access memory; and the portion of the first data that does not exceed the predetermined size is preloaded into the dynamic random access memory.

According to an embodiment of the present invention, after the initialization is completed, the second data of the flash memory array is moved to the dynamic random access memory according to an access instruction to improve the access efficiency of the host. The host accesses the second data according to the access instruction.

100‧‧‧ storage device

101‧‧‧Flash memory array

102‧‧‧ Dynamic Random Access Memory

103‧‧‧ Controller

10‧‧‧Host

200A, 200B‧‧‧Power access table

INS‧‧‧ access instructions

S10~S16‧‧‧Step procedure

S20~S26‧‧‧Step process

1 is a block diagram showing a storage device according to an embodiment of the present invention; FIG. 2A is a schematic view showing an upper power access table according to an embodiment of the present invention; and FIG. 2B is a view showing A schematic diagram of an upper power access table according to another embodiment of the present invention; FIG. 3A is a flowchart showing a method for establishing a power-on access list according to an embodiment of the present invention; and FIG. 3B is a diagram showing A flowchart of a method of using a power-on access list according to an embodiment of the present invention.

The following description is an embodiment of the present invention. The intent is to exemplify the general principles of the invention and should not be construed as limiting the scope of the invention, which is defined by the scope of the claims.

It is noted that the following disclosure may provide embodiments or examples for practicing various features of the present invention. The specific elements and arrangements of the elements described below are merely illustrative of the spirit of the invention and are not intended to limit the scope of the invention. In addition, the following description may reuse the same component symbols or characters in various examples. However, the re-use is for the purpose of providing a simplified and clear description, and is not intended to limit the relationship between the various embodiments and/or configurations discussed below. In addition, the description of one of the features described in the following description is connected to, coupled to, and/or formed on another feature, etc., and may include a plurality of different embodiments, including direct contact of the features, or other additional Features are formed between the features and the like such that the features are not in direct contact.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram showing a storage device in accordance with an embodiment of the present invention. As shown in FIG. 1, the storage device 100 includes a flash memory array 101, a dynamic random access memory 102, and a controller 103. The storage device 100 is coupled to the host 10, and the host 10 and the storage device 100 are combined. The storage device 100 generates and maintains a Host-Logical-Flash Physical (H2F) comparison table, and records the correspondence between the logical address and the physical address of the user data. According to an embodiment of the invention, the storage device 100 can be a solid state hard disk using a USB, SATA, PATA, PCIE physical interface or a USB, NVME, AHCI, SCSI communication protocol.

Assuming that the data storage capacity of the storage device 100 is 256 GB, after the system is powered on, the controller 103 provides the user data required by the host 10 according to the H2F comparison table, and records the logical address or physical address of the user data. Electric access table. Assuming that a logical address or a physical address is 2B in the power-on access table, and a logical address or a physical address corresponds to 4 KB of user data, the controller 103 can establish a power-on access of 8 MB. The table records the logical address of the 16 GB user data. The power-on access table records the user data that the host 10 wants to read or write. According to the operating characteristics of the host 10, the user data recorded by the power-on access table may be an operating system, an application, or a computer. The file of the game, and the power-on access table is mainly for recording the user data that the host 10 wants to read, and a small part is the user data that the host 10 wants to update/write.

When the host 10 starts Initializing, peripheral devices, such as the storage device 100, are also required to start initialization. However, the time required for the storage device 100 to complete initialization is shorter than the time required for the host 10 to complete initialization, and therefore, is stored. After the device 100 completes initialization and the host 10 has not completed initialization (also referred to as idle time), the controller 103 sets the user data from the flash memory array according to the address information recorded by the power-on access table. 101 is preloaded into the DRAM 102, and when the host 10 completes initialization and starts the loading of the operating system, that is, when the host 10 loads the file of the operating system from the storage device 100, the storage device 100 can be quickly and dynamically randomly selected. Accessing the memory 102, rather than from the flash memory array 101, provides a file (i.e., user profile) of the operating system to the host 10 to reduce the time required for data access. Since the storage device 100 can complete the initialization quickly, the idle time can also be regarded as the time until the host device 10 starts to access the storage device 100 after the storage device 100 is powered on.

According to another embodiment of the present invention, the recording of the power-on access table is preferably recorded in accordance with the reading order of the user data. As such, it is assumed that the controller 103 cannot load all the user data corresponding to the power-on access table into the DRAM 102 during the idle time, because the user data loaded first may be used by the host 10. First, during the data reading and processing, the controller 103 can continue to load the remaining user data corresponding to the power-on access table into the dynamic random access memory 102, waiting for the host 10 to read. The object of the invention is achieved.

According to another embodiment of the present invention, the controller 103 arranges the records of the power-on access table according to the parameters of the flash memory array 101, wherein the parameters of the flash memory array 101 include: a channel. The number or number of Chip Enables to speed up the reading of user data.

According to another embodiment of the present invention, the size of the user data corresponding to the power-on access table is preferably smaller than the data storage capacity of the dynamic random access memory 102, and thus, all the corresponding power-on access tables are User data can be stored in the DRAM 102.

According to another embodiment of the present invention, when the size of the user data corresponding to the power-on access table is greater than the data storage capacity of the dynamic random access memory 102, the controller 103 only corresponds to the power-on access table. Some user data is loaded into the dynamic random access memory 102. After the user data is read by the host 10, the remaining user data corresponding to the above power access table is replaced by the dynamic random access. User data of the memory 102.

2A is a schematic diagram showing an upper power access table according to an embodiment of the present invention. According to an embodiment of the present invention, the power-on access table 200A is used to record the logical address of the user data. According to the H2F comparison table, the controller 103 can learn that each user data is stored in the flash memory array 101. Address and access it. The explanation is explained here with only 10 access materials, and is not limited thereto in any way.

Figure 2B is a schematic diagram showing an over-powered access meter in accordance with another embodiment of the present invention. According to an embodiment of the present invention, the power-on access table 200B is like a small H2F look-up table, and records the logical address and physical address of the user data. Therefore, the controller 103 can learn, according to the power-on access table, that each user data is stored in the address of the flash memory array 101 and access it. The explanation is explained here with only 10 access materials, and is not limited thereto in any way.

FIG. 3A is a flow chart showing a method of establishing a power-on access list according to an embodiment of the present invention, and the method for establishing a power-on access list is applicable to the storage device 100. First, the storage device 100 performs initialization (step S10), wherein the initialization of the storage device 100 can be initiated by re-provisioning of the power source or host command of the host 10.

Then, the user data of the flash memory array is accessed according to the host command and the H2F comparison table (step S12). After the storage device 100 completes initialization, it waits to receive a host command from the host 10. Assuming host 10 also completes initialization and begins operating system loading, the host command is primarily a data read command to request storage device 100 to provide an archive of the operating system. The storage device 100 reads the user data (file of the operating system) of the flash memory array 101 according to the host command and the H2F comparison table, and outputs the user data to the host 10. In addition to the data read command, the host command may also be a data write command.

Then, the address information of the user data is recorded to the power-on access table (step S14) and it is determined whether the power-on access table has filled the address information (step S16), if the power-on access table has not yet filled the address The information continues to record the address information to the power-on access table; if the power-on access table has filled the address information (for example, 8 MB address information has been stored), the method of establishing the power-on access table of the present invention is terminated.

FIG. 3B is a flow chart showing a method of using a power-on access table according to an embodiment of the present invention, and the method for establishing a power-on access list is applicable to the storage device 100.

First, the storage device 100 performs initialization (step S20), wherein the initialization of the storage device 100 can be initiated by re-provisioning of the power source or host command of the host 10.

Next, it is judged whether or not the power-on access table exists (step S22), and if yes, step S24 is performed, otherwise the execution of the method of using the power-on access table of the present invention is ended.

Then, according to the H2F comparison table, the user data corresponding to the power-on access table is sequentially uploaded to the data temporary storage device (step S24), wherein the data temporary storage device can be the dynamic random access memory 102, if If the amount of data of the user data corresponding to the electric access table is greater than the data storage capacity of the data register, the user data corresponding to the power-on access table is first uploaded to the data register, and then the remaining power is uploaded. Access the user data corresponding to the table to the data register. If the record of the power-on access table is as shown in FIG. 2B, the controller 103 can sequentially upload the user data corresponding to the power-on access table to the data register without using the H2F comparison table.

Next, the storage device 100 is stored in accordance with the host command, the power-on access table, and the H2F look-up table (step S26). The storage device 100 receives the host command from the host 10, first determines whether the logical address in the host command is recorded in the power-on access table, and if so, the storage device 100 directly provides the user data stored in the data register to The host device 10; otherwise, the storage device 100 accesses the user data stored in the flash memory array 101 according to the host command and the H2F comparison table.

The above is an overview feature of the embodiment. Those having ordinary skill in the art should be able to use the present invention as a basis for design or adaptation to achieve the same objectives and/or achieve the same advantages of the embodiments described herein. It should be understood by those of ordinary skill in the art that the same configuration should not depart from the spirit and scope of the present invention, and various changes, substitutions and substitutions can be made without departing from the spirit and scope of the present invention. The illustrative methods are merely illustrative of the steps, but are not necessarily performed in the order presented. The steps may be additionally added, substituted, changed, and/or eliminated, as appropriate, and are consistent with the spirit and scope of the disclosed embodiments.

Claims (9)

 A storage device is coupled to a host, comprising: a flash memory array for storing a first data; and a dynamic random access memory for storing a power-on access table, wherein the power-on access table is used And recording, by the controller, the address information of the first data accessed by the host after the power-on of the storage device; and a controller, wherein the controller is in an idle time after the storage device is powered on, according to the bit The address information preloads the first data into the dynamic random access memory to improve the access efficiency of the host to the storage device.    The storage device of claim 1, wherein the idle time is a time before the storage device is powered on until the host starts to access the storage device.    The storage device of claim 1, wherein the host performs an initialization by using the first data preloaded into the dynamic random access memory.    The storage device of claim 1, wherein the power-on access table further records a sequence of the first data accessed by the host after the power-on of the storage device, wherein the controller is based on In the above sequence, the first data of a predetermined size is preloaded into the dynamic random access memory, and the remaining first data is not preloaded into the dynamic random access memory, wherein the host accesses the foregoing Dynamic random access memory and the above flash memory array to complete an initialization.    The storage device of claim 4, wherein, after the host completes the initialization, the host sends an access command for accessing a second data of the flash memory array, the controller is configured according to The access command temporarily stores the second data in the dynamic random access memory to improve access efficiency of the host to the storage device.    A preloading method, applicable to a flash memory array, comprising: powering on the flash memory array; and accessing a first data of the flash memory array by a host to perform an initialization The address information is recorded in a power-on access table; the flash memory array is powered on again; and the first data is saved from the flash memory according to the address information of the power-on access table. The body array is preloaded with a dynamic random access memory; and the initialization of the host is completed by using the first data.    The preloading method of claim 6, wherein the step of preloading the first data from the flash memory array into the dynamic random access memory according to the power-on access table is performed An idle time between when the flash memory array is powered up and when the host begins to access the flash memory array.    The preloading method of claim 6, wherein the power-on access table is further configured to record one of the first data accessed by the host after the previous power-on of the flash memory array. a sequence, wherein the step of preloading the first data from the flash memory array into the dynamic random access memory according to the power-on access table according to the foregoing power-on access table further comprises: performing a predetermined size according to the foregoing sequence The first data is preloaded into the dynamic random access memory; and the portion in which the first data exceeds the predetermined size is not preloaded into the dynamic random access memory.    The preloading method of claim 8, wherein, after the initializing is completed, moving a second data of the flash memory array to the dynamic random access memory according to an access command, The access efficiency of the host is improved, wherein the host accesses the second data according to the access instruction.