TW201835774A - Method for controlling the writing flow of a solid state disk - Google Patents

Method for controlling the writing flow of a solid state disk Download PDF

Info

Publication number
TW201835774A
TW201835774A TW106109235A TW106109235A TW201835774A TW 201835774 A TW201835774 A TW 201835774A TW 106109235 A TW106109235 A TW 106109235A TW 106109235 A TW106109235 A TW 106109235A TW 201835774 A TW201835774 A TW 201835774A
Authority
TW
Taiwan
Prior art keywords
data
write
user data
written
traffic
Prior art date
Application number
TW106109235A
Other languages
Chinese (zh)
Inventor
曹祖英
黃聰明
方子維
Original Assignee
廣明光電股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 廣明光電股份有限公司 filed Critical 廣明光電股份有限公司
Priority to TW106109235A priority Critical patent/TW201835774A/en
Publication of TW201835774A publication Critical patent/TW201835774A/en

Links

Abstract

The present invention is to disclose a method for controlling the writing flow of a solid state drive. The method reduces the writing flow of inner data when concurrently writing user data and inner data, and learns a delay time from the writing flow of user data at concurrency stage when the inner data have been written. The method reduces the writing flow of the user data with the delay time to stabilize the writing efficiency.

Description

固態硬碟控制寫入流量的方法  Solid state hard disk control method for writing traffic  

本發明有關一種固態硬碟,尤其關於固態硬碟為穩定主機寫入資料的效能,控制寫入流量的方法。 The present invention relates to a solid state hard disk, and more particularly to a method for controlling write traffic by a solid state hard disk for stabilizing the performance of writing data to a host.

固態硬碟(Solid State Drive,簡稱SSD)是由反及閘快閃記憶體陣列(NAND Flash Memory Array)整合成為單一的儲存裝置。因快閃記憶體以資料區塊為資料抹除單位,資料區塊包含有效資料頁及被廢棄的無效資料頁,過多的無效資料頁佔據快閃記憶體有效的儲存空間。因此固態硬碟利用垃圾回收機制(Garbage Collection,簡稱GC),對無效資料頁超出閥值的資料區塊進行回收,將資料區塊中的有效資料頁,先讀出再搬移至空白資料區塊,然後抹除整個回收資料區塊的資料,形成備用的空白資料區塊,以利固態硬碟存取資料。 Solid State Drive (SSD) is a single storage device that is integrated into a NAND Flash Memory Array. Because the flash memory erases the unit with the data block as the data, the data block contains the valid data page and the discarded invalid data page, and the excessive invalid data page occupies the effective storage space of the flash memory. Therefore, the solid state hard disk uses the Garbage Collection (GC) to recover the data blocks whose invalid data pages exceed the threshold, and the valid data pages in the data block are read and then moved to the blank data block. Then, erase the data of the entire recycling data block to form an alternate blank data block for the solid state hard disk to access the data.

如圖6所示,先前技術固態硬碟的緩衝記憶體1,除暫存接收主機的使用者資料(User Data)2外,尚包含垃圾回收機制讀出而需寫回的有效資料,及需要備份的邏輯實體對照表等內部資料3,需由各快閃記憶體的先進先出(FIFO)的資料管道,依序寫入快閃記憶體陣列4。先前技術固態硬碟為了避免影響緩衝記憶體1中使用者資料2的寫入效能,通常利用使用者 資料寫入的空檔,將內部資料3寫入快閃記憶體陣列4。但寫入內部資料3時,主機時常指令固態硬碟存取使用者資料2。如果寫入的內部資料3不多,固態硬碟可及時回應主機的指令,不致過度影響使用者資料2的存取效能。一旦寫入的內部資料3過多,固態硬碟需要較長的時間寫入內部資料3,無法回應主機的指令,主機端使用者資料2的存取效能會嚴重受到影響,甚至存取效能降為零,導致使用者須等待較久時間或空等。 As shown in FIG. 6, the buffer memory 1 of the prior art solid state hard disk includes, in addition to the user data (User Data) 2 of the temporary receiving host, the valid data that needs to be written back by the garbage collection mechanism, and the need The backed up logical entity comparison table and other internal data 3 need to be sequentially written into the flash memory array 4 by the data stream of the first in first out (FIFO) of each flash memory. In order to avoid affecting the writing performance of the user data 2 in the buffer memory 1, the prior art solid state hard disk usually writes the internal data 3 to the flash memory array 4 by using the space written by the user data. However, when writing internal data 3, the host often instructs the solid state hard disk to access the user data 2. If there is not much internal data written, the solid state hard disk can respond to the host's instructions in time without affecting the access performance of the user data 2. Once the internal data 3 is written too much, the solid state hard disk takes a long time to write to the internal data 3, and cannot respond to the host's instructions. The access performance of the host user data 2 is seriously affected, and even the access performance is reduced to Zero, causing the user to wait for a long time or empty.

因此另有先前技術的固態硬碟,將內部資料3與使用者資料2同時寫入快閃記憶體陣列1。雖然使用者資料3的存取效能與內部資料2的寫入效能都會受到影響,但是可維持使用者資料3的存取效能在一定的水準,且可防止使用者資料3的存取效能降為零,讓使用者無空等的感覺,以避免使用者產生固態硬碟低效能的印象。 Therefore, in addition to the prior art solid state hard disk, the internal data 3 and the user data 2 are simultaneously written into the flash memory array 1. Although the access performance of User Data 3 and the write performance of Internal Data 2 will be affected, the access performance of User Data 3 can be maintained at a certain level, and the access performance of User Data 3 can be prevented from being reduced to Zero, let the user feel like no space, to avoid the user's impression of low-performance solid-state hard disk.

然而,前述的先前技術固態硬碟,控制內部資料與使用者資料同時寫入時,如圖7所示,顯示使用者資料的存取效能曲線P,在內部資料與使用者資料存取的同時寫入時段A,固態硬碟必須分擔控制處理能力寫入內部資料,排擠造成使用者資料的存取效能降低,且導致效能曲線激烈上下起伏。而在完成寫入內部資料後,僅寫入使用者資料的單獨寫入時段B,因無內部資料寫入的干擾,存取效能升高,不僅無法維持前後時段在相同存取效能水準,且單獨寫入時段B短時間大量寫入使用者資料,造成使用者資料間雍塞排擠,使效能曲線出現上下拉長的影線L,同樣無法維持在相同存取效能水準。因此,固態硬碟在寫入流量的控制方法上,仍有問題亟待解決。 However, in the foregoing prior art solid state hard disk, when the internal data and the user data are simultaneously written, as shown in FIG. 7, the access performance curve P of the user data is displayed, and the internal data and the user data are accessed simultaneously. During the writing period A, the solid state hard disk must share the control processing capability to write internal data, and the performance of the user data is reduced due to the crowding, and the performance curve is violently up and down. After the internal data is written, only the individual write period B of the user data is written, and the access performance is increased because there is no internal data write interference, and the same access performance level cannot be maintained in the before and after periods, and The write time period B is written into the user data in a short time, causing the user data to be crowded and squeezed, so that the performance curve has a long and downward shadow line L, and the same access performance level cannot be maintained. Therefore, there is still a problem in the control method of the write capacity of the solid state hard disk.

本發明的目的提供一種固態硬碟控制寫入流量的方法,在內部資料與使用者資料的同時寫入時段,藉由延長內部資料的寫入時間,降低內部資料的寫入流量,增加使用者資料寫入流量,以提升使用者資料寫入效能。 The object of the present invention is to provide a method for controlling write traffic by a solid state hard disk, which can reduce the write flow of internal data and increase the write flow of internal data by increasing the write time of internal data and user data. Data is written to the traffic to improve user data write performance.

本發明的另一目的提供一種固態硬碟控制寫入流量的方法,利用在同時寫入時段學習使用者資料的寫入流量,形成延遲時間,在單獨寫入時段利用延遲時間延遲寫入使用者資料,降低使用者資料的寫入流量,以穩定使用者資料的寫入效能。 Another object of the present invention is to provide a method for controlling write traffic by a solid state hard disk, which utilizes a write traffic for learning user data during a simultaneous write period to form a delay time, and writes the user with a delay time delay in a separate write period. Data, reducing the write traffic of user data to stabilize the write performance of user data.

為了達到前述發明的目的,本發明固態硬碟控制寫入流量的方法,首先控制減少內部資料的寫入流量,內部資料包含邏輯實體對照表的備份或垃圾回收資料區塊的有效資料頁,接著在同時寫入時段,同時寫入使用者資料及內部資料,檢查未完成寫入內部資料時,繼續同時寫入使用者資料及內部資料。檢查已完成寫入內部資料時,由同時寫入時段的使用者資料寫入流量學習延遲時間,在單獨寫入時段,利用延遲時間控制減少使用者資料的寫入流量。 In order to achieve the object of the foregoing invention, the method for controlling write traffic of the solid state hard disk of the present invention firstly controls to reduce the write traffic of the internal data, and the internal data includes a backup of the logical entity comparison table or a valid data page of the garbage collection data block, and then At the same time, the user data and internal data are written at the same time, and when the internal data is not completed, the user data and internal data are simultaneously written. When checking that the internal data has been written, the user data of the simultaneous writing period is written into the traffic learning delay time, and in the separate writing period, the delay time control is used to reduce the writing traffic of the user data.

本發明固態硬碟控制寫入流量的方法,將內部資料的寫入時間增加至預設延長寫入時間,或利用固定或不固定的間隔時間,間斷式的寫入內部資料,使內部資料在預設延長寫入時間完成寫入,以減少內部資料的寫入流量。在同時寫入時段,記錄內部資料的延長寫入時間及使用者資料的寫入資料量,由該寫入資料量除以延長寫入時間,計算使用者資料在同時寫入時段的寫入流量。再將單獨寫入時段使用者資料剩下的寫入資料量,除以使用者資料在同時寫入時段的寫入流量,學習出延遲時間。 The solid-state hard disk of the invention controls the method of writing the traffic, increases the writing time of the internal data to a preset extended writing time, or intermittently writes the internal data by using a fixed or unfixed interval time, so that the internal data is The write is completed by a preset extended write time to reduce the write flow of internal data. During the simultaneous writing period, the extended writing time of the internal data and the written data amount of the user data are recorded, and the writing data amount is divided by the extended writing time, and the writing traffic of the user data in the simultaneous writing period is calculated. . Then, the amount of data written in the user data of the time period is separately written, divided by the write traffic of the user data in the simultaneous writing period, and the delay time is learned.

本發明固態硬碟控制寫入流量的方法,利用延遲時間延長使用者資料的寫入時間,或利用固定或不固定的間隔時間,間斷式的寫入使用者資料,讓使用者資料在延遲時間完成寫入,以減少使用者資料的寫入流量。 The solid-state hard disk of the present invention controls the method of writing traffic, and uses the delay time to extend the writing time of the user data, or uses a fixed or unfixed interval time to intermittently write the user data, so that the user data is delayed. Write is completed to reduce the write traffic of user data.

10‧‧‧電子裝置儲存系統 10‧‧‧Electronic device storage system

11‧‧‧主機 11‧‧‧Host

12‧‧‧中央處理器 12‧‧‧Central processor

13‧‧‧動態隨機記憶體 13‧‧‧Dynamic random memory

14‧‧‧傳輸介面 14‧‧‧Transport interface

15‧‧‧固態硬碟 15‧‧‧ Solid State Drive

16‧‧‧控制器 16‧‧‧ Controller

17‧‧‧緩衝記憶體 17‧‧‧ Buffer memory

18‧‧‧快閃記憶體陣列 18‧‧‧Flash Memory Array

20‧‧‧使用者資料 20‧‧‧ User data

21‧‧‧內部資料 21‧‧‧ Internal information

A‧‧‧同時寫入時段 A‧‧‧ simultaneous writing period

B‧‧‧單獨寫入時段 B‧‧‧Single writing period

T1‧‧‧寫入時間 T1‧‧‧ write time

T2‧‧‧延長寫入時間 T2‧‧‧Extended write time

T3‧‧‧延遲時間 T3‧‧‧Delayed time

圖1 為本發明電子裝置儲存系統的功能方塊圖。 1 is a functional block diagram of an electronic device storage system of the present invention.

圖2 本發明固態硬碟寫入資料的示意圖。 Figure 2 is a schematic diagram of the solid state hard disk writing data of the present invention.

圖3 為本發明使用者資料的存取效能的曲線圖。 FIG. 3 is a graph showing the access performance of the user data of the present invention.

圖4 為本發明另一實施例的固態硬碟寫入資料的示意圖。 FIG. 4 is a schematic diagram of a solid state hard disk write data according to another embodiment of the present invention.

圖5 為本發明固態硬碟控制寫入流量的方法的流程圖。 FIG. 5 is a flow chart of a method for controlling write traffic by a solid state hard disk according to the present invention.

圖6 為先前技術固態硬碟寫入資料的示意圖。 Figure 6 is a schematic diagram of prior art solid state hard disk write data.

圖7 為先前技術使用者資料的存取效能的曲線圖。 Figure 7 is a graph of the access performance of prior art user data.

有關本發明為達成上述目的,所採用之技術手段及其功效,茲舉較佳實施例,並配合圖式加以說明如下。 The technical means and the efficacies of the present invention for achieving the above objects are as follows, and the preferred embodiments are described below with reference to the drawings.

請同時參閱圖1至圖3,圖1為本發明電子裝置儲存系統的功能方塊圖,圖2為本發明固態硬碟寫入資料的示意圖,圖3為本發明使用者資料的存取效能的曲線圖。圖1中,本發明電子裝置儲存系統10,在主機11設中央處理器(CPU)12及動態隨機記憶體(DRAM)13,由中央處理器12配合動態隨機記憶體13將寫入的使用者資料20(參圖2),傳輸至連接在傳輸介面14的固態硬碟(SSD)15,固態硬碟15內設控制器16及緩衝記憶體17,由控制 器16配合緩衝記憶體17接收主機11要寫入的使用者資料20,將使用者資料20暫存至緩衝記憶體17,以待寫入快閃記憶體陣列18。 Please refer to FIG. 1 to FIG. 3 . FIG. 1 is a functional block diagram of a storage system of an electronic device according to the present invention. FIG. 2 is a schematic diagram of data written by a solid state hard disk according to the present invention, and FIG. 3 is an access performance of user data according to the present invention. Graph. In FIG. 1, the electronic device storage system 10 of the present invention has a central processing unit (CPU) 12 and a dynamic random access memory (DRAM) 13 in the host 11, and a user to be written by the central processing unit 12 in conjunction with the dynamic random memory 13. The data 20 (see FIG. 2) is transmitted to a solid state hard disk (SSD) 15 connected to the transmission interface 14, and the controller 16 and the buffer memory 17 are disposed in the solid state hard disk 15. The controller 16 cooperates with the buffer memory 17 to receive the host. 11 User data 20 to be written, the user data 20 is temporarily stored in the buffer memory 17 to be written to the flash memory array 18.

為了管理資料的邏輯位址與儲存在快閃記憶體陣列18的實體位址的相對關係,固態硬碟15在啟動時,先讀取快閃記憶體陣列18中各資料區塊的管理資料,形成資料的邏輯位址與實體位址的邏輯實體對照表,儲存在緩衝記憶體17中。且為了防止揮發性的緩衝記憶體17在斷電時邏輯實體對照表的記憶資料消失,因此定期的將邏輯實體對照表寫入非揮發的快閃記憶體陣列18進行備份。 In order to manage the relative relationship between the logical address of the data and the physical address stored in the flash memory array 18, the solid state hard disk 15 first reads the management data of each data block in the flash memory array 18 at startup. A logical entity comparison table of logical addresses and physical addresses forming the data is stored in the buffer memory 17. In order to prevent the memory data of the logical entity table from disappearing when the volatile buffer memory 17 is powered off, the logical entity comparison table is periodically written into the non-volatile flash memory array 18 for backup.

此外,固態硬碟15利用垃圾回收機制,對快閃記憶體陣列18中無效資料頁過多的資料區塊,將資料區塊中的有效資料頁,陸續讀出至緩衝記憶體17,以待寫入搬移至快閃記憶體陣列18的備用空白資料區塊,進行回收資料區塊。因此固態硬碟15主要除了處理寫入主機的使用者資料20外,尚需處理固態硬碟15本身的內部資料21(參圖2),內部資料21包含前述寫入邏輯實體對照表備份及垃圾回收資料區塊的有效資料頁等,將暫存在緩衝記憶體17的資料,由各先進先出資料管道,依序寫入快閃記憶體陣列18。 In addition, the solid state hard disk 15 uses the garbage collection mechanism to read out the valid data pages in the data block to the buffer memory 17 for the data block with too many invalid data pages in the flash memory array 18 to be written. The data is moved to the spare blank data block of the flash memory array 18 to recover the data block. Therefore, the solid state hard disk 15 mainly needs to process the internal data 21 of the solid state hard disk 15 itself (see FIG. 2) in addition to the user data 20 written to the host computer, and the internal data 21 includes the foregoing written logical entity comparison table backup and garbage. The valid data page of the data block is recovered, and the data temporarily stored in the buffer memory 17 is sequentially written into the flash memory array 18 by the FIFO data pipes.

如圖2所示,由於固態硬碟15的控制器16,受機器本身效能的限制,單位時間處理的資料有一定流量的上限。固態硬碟15在同時處理使用者資料20及內部資料21的同時寫入時段A,為了維持使用者資料20存取效能在一定水準,並避免過度處理內部資料21,而影響使用者資料20的存取效能。本發明在同時處理內部資料21與使用者資料20的同時寫入時段A,將內部資料21的寫入時間T1,增加至預設延長寫入時間T2,藉由拉長處理 的時間,降低內部資料21的寫入流量。讓控制器16減少處理寫入的內部資料21,空出較多的時間寫入使用者資料20,以提升使用者資料20的存取效能。 As shown in FIG. 2, since the controller 16 of the solid state hard disk 15 is limited by the performance of the machine itself, the data processed per unit time has an upper limit of a certain flow rate. The solid state drive 15 writes the user data 20 and the internal data 21 while simultaneously writing the time period A, in order to maintain the user data 20 access performance at a certain level, and avoid excessive processing of the internal data 21, which affects the user data 20 Access performance. The present invention simultaneously writes the internal data 21 and the user data 20 simultaneously into the period A, increases the writing time T1 of the internal data 21 to the preset extended writing time T2, and reduces the internal time by lengthening the processing time. The write traffic of the data 21. The controller 16 is caused to reduce the internal data 21 processed and written, and spend more time writing to the user data 20 to improve the access performance of the user data 20.

因使用者資料20的存取效能是以單位時間作為評估標準,本發明將內部資料21延長時間寫入,降低內部資料21單位時間的寫入流量,分散處理內部資料21,減少單位時間處理的內部資料21,而可同時避免排擠使用者資料20的處理時間,因此可增加單位時間內處理的使用者資料20。比較先前技術圖5與本發明圖3同時寫入時段A的使用者資料20存取效能曲線P,本發明大致提高使用者資料20的存取效能,且使用者資料20存取效能曲線P激烈起伏的現象及上下影線L也大幅減少,顯示使用者資料20存取效能更加穩定。 Since the access performance of the user data 20 is based on the unit time as the evaluation standard, the present invention writes the internal data 21 for an extended period of time, reduces the write flow of the internal data 21 units, disperses the internal data 21, and reduces the processing time per unit time. The internal data 21 can avoid the processing time of the user data 20 at the same time, so that the user data 20 processed per unit time can be increased. Comparing the prior art FIG. 5 with the user data 20 access performance curve P of the simultaneous writing period A of FIG. 3 of the present invention, the present invention substantially improves the access performance of the user data 20, and the user data 20 access performance curve P is intense. The undulation phenomenon and the upper and lower shadow lines L are also greatly reduced, indicating that the user data 20 access performance is more stable.

請再參圖2,由於寫入的內部資料21優先度高於寫入的使用者資料20,當固態硬碟15完成寫入內部資料21後,在僅剩寫入使用者資料20的單獨寫入時段B時,固態硬碟15的控制器16將全面處理寫入使用者資料20,使用者資料20的寫入流量將大增,雖然可大幅提升使用者資料20的存取效能,但會造成前後時段使用者資料20的存取效能差異過大,而無法維持穩定的效能。此外,寫入的使用者資料20流量大增,也需依序經由各先進先出的資料管道,才能處理寫入快閃記憶體陣列18。而各先進先出的資料管道處理資料的快慢,將造成寫入的使用者資料20間的相互排擠,進而拖累降低使用者資料20的存取效能。 Referring to FIG. 2 again, since the written internal data 21 has higher priority than the written user data 20, when the solid state hard disk 15 finishes writing the internal data 21, only the write of the user data 20 is separately written. When entering the time period B, the controller 16 of the solid state hard disk 15 will fully process the write data to the user data 20. The write traffic of the user data 20 will be greatly increased, although the access performance of the user data 20 can be greatly improved, but The access performance of the user data 20 before and after the time difference is too large to maintain stable performance. In addition, the written user data 20 traffic is greatly increased, and it is also necessary to sequentially process the write flash memory array 18 via the first-in-first-out data pipeline. The speed of processing data by each of the first-in-first-out data pipelines will result in the mutual exclusion of the written user data 20, thereby dragging down the access performance of the user data 20.

為了維持穩定寫入效能,本發明固態硬碟15同時處理寫入使用者資料20及內部資料21時,在同時寫入時段A記錄內部資料21的延長寫入 時間T2,以及使用者資料20的寫入資料量,由資料量除以延長寫入時間T2,計算使用者資料20在同時寫入時段A的寫入流量。接著藉由使用者資料20剩下的寫入資料量除以同時寫入時段A的寫入流量,學習出延遲時間T3的參數。由固態硬碟15的控制器16利用延遲時間T3,延長使用者資料20寫入時間,控制單獨寫入時段B的使用者資料20的寫入流量,減少寫入流量至約等於同時寫入時段A的寫入流量。雖然單獨寫入時段B無內部資料21寫入的干擾,使用者資料20的寫入效能會提高。比較先前技術圖5與本發明圖3單獨寫入時段B的使用者資料20存取效能曲線P,本發明大致提高使用者資料20的存取效能,由於減少使用者資料20的寫入流量,讓固態硬碟15更充裕能力處理寫入使用者資料20,避免使用者資料20相互排擠,而讓存取效能曲線P上下影線L也大幅縮短,顯示使用者資料20存取效能更加穩定。 In order to maintain stable write performance, the solid state hard disk 15 of the present invention simultaneously processes the extended write time T2 of the internal data 21 and the user data 20 at the simultaneous write period A when the user data 20 and the internal data 21 are simultaneously written. The amount of data written is divided by the amount of data divided by the extended writing time T2, and the write flow rate of the user data 20 at the simultaneous writing period A is calculated. Then, by dividing the amount of write data remaining in the user data 20 by the write flow rate of the simultaneous write period A, the parameter of the delay time T3 is learned. The controller 16 of the solid state hard disk 15 uses the delay time T3 to extend the write time of the user data 20, control the write traffic of the user data 20 of the individual write period B, and reduce the write traffic to approximately equal to the simultaneous write period. A write traffic. Although there is no interference written by the internal data 21 in the separate writing period B, the writing performance of the user data 20 is improved. Comparing the prior art FIG. 5 with the user data 20 access performance curve P of the separate writing period B of FIG. 3 of the present invention, the present invention substantially improves the access performance of the user data 20, and reduces the write traffic of the user data 20, The solid state hard disk 15 is more capable of processing and writing the user data 20, so as to prevent the user data 20 from being mutually crowded, and the upper and lower shadow lines L of the access performance curve P are also greatly shortened, and the user data 20 is more stable in access performance.

如圖4所示,為本發明另一實施例的固態硬碟控制寫入流量的方法。本實施例的固態硬碟控制寫入流量的方法與前實施例基本架構相同,為簡化說明,相同的構件沿用前實施例的件號。本實施例與前實施例主要不同在延長寫入時間T2與延遲時間T3的實施方式。本實施例對延長寫入時間T2,是在保持原內部資料21寫入流量下,利用固定或不固定的間隔時間m,間斷式的寫入內部資料21,使內部資料21在延長寫入時間T2完成寫入,以控制減少單位時間內部資料21寫入流量。本實施例對延遲時間T3,在保持原使用者資料20寫入流量下,利用固定或不固定的間隔時間n,間斷式的寫入使用者資料20,讓使用者資料20在延遲時間T3完成寫入,以控制減少單位時間使用者資料20寫入流量。同樣可提供較充裕的處理時間,使本實施例達到穩定使用者資料存取效能的目的。 As shown in FIG. 4, a solid state hard disk control method for controlling write traffic according to another embodiment of the present invention. The method of controlling the write traffic by the solid state hard disk of this embodiment is the same as that of the previous embodiment. For simplification of description, the same components follow the part number of the previous embodiment. This embodiment differs from the previous embodiment mainly in the implementation of extending the write time T2 and the delay time T3. In the present embodiment, when the write time T2 is extended, the internal data 21 is intermittently written by using the fixed or unfixed interval time m while maintaining the write flow rate of the original internal data 21, so that the internal data 21 is extended in the write time. T2 completes the write to control the reduction of internal data 21 write traffic per unit time. In the embodiment, for the delay time T3, the user data 20 is intermittently written by using the fixed or unfixed interval time n while maintaining the original user data 20 write flow rate, so that the user data 20 is completed at the delay time T3. Write to control the reduction of user data 20 write traffic per unit time. The same processing time can be provided, so that the embodiment can achieve the purpose of stabilizing user data access performance.

如圖5所示,為本發明固態硬碟控制寫入流量的方法的流程圖。本發明固態硬碟控制寫入流量的詳細步驟說明如下:首先步驟S1,將內部資料的寫入時間增加至預設延長寫入時間,以控制減少內部資料的寫入流量;在步驟S2,在同時寫入時段同時寫入使用者資料及內部資料;在步驟S3,檢查是否完成寫入內部資料?假如未完成寫入內部資料,則回到步驟S2繼續同時寫入,假如已完成寫入內部資料,則至步驟S4,由同時寫入時段的使用者資料寫入流量,學習延遲時間;最後在步驟S5,利用延遲時間,控制單獨寫入時段的使用者資料的寫入流量,減少使用者資料的寫入流量。 As shown in FIG. 5, it is a flowchart of a method for controlling write traffic of a solid state hard disk according to the present invention. The detailed steps of controlling the write traffic by the solid state hard disk of the present invention are as follows: First, in step S1, the write time of the internal data is increased to a preset extended write time to control the reduction of the write traffic of the internal data; in step S2, At the same time, the user data and the internal data are simultaneously written in the writing period; in step S3, it is checked whether the internal data is written. If the internal data is not completed, return to step S2 to continue the simultaneous writing. If the internal data has been written, then to step S4, the user data is written by the user of the simultaneous writing period, and the delay time is learned; In step S5, the delay time is used to control the write traffic of the user data in the separate writing period, and the write traffic of the user data is reduced.

因此,本發明固態硬碟控制寫入流量的方法,在內部資料與使用者資料的同時寫入時段,就可藉由延長內部資料的寫入時間,降低內部資料的寫入流量,增加使用者資料寫入流量,達到提升使用者資料寫入效能的目的。並在同時寫入時段,學習使用者資料的寫入流量,形成單獨寫入時段的延遲時間的參數,在單獨寫入時段延遲寫入使用者資料,控制降低使用者資料的寫入流量,達到穩定使用者資料的寫入效能的目的。 Therefore, the method for controlling the write traffic of the solid-state hard disk of the present invention can increase the write time of the internal data and increase the write flow of the internal data by increasing the write time of the internal data and the user data at the same time. The data is written into the traffic to improve the performance of the user data. And in the simultaneous writing period, learning the writing flow of the user data, forming a parameter of the delay time of the separate writing period, delay writing the user data in the separate writing period, and controlling the writing flow of the user data is reduced. Stabilize the write performance of user data.

以上所述者,僅為用以方便說明本發明之較佳實施例,本發明之範圍不限於該等較佳實施例,凡依本發明所做的任何變更,於不脫離本發明之精神下,皆屬本發明申請專利之範圍。 The above is only a preferred embodiment for facilitating the description of the present invention, and the scope of the present invention is not limited to the preferred embodiments, and any changes made in accordance with the present invention may be made without departing from the spirit of the present invention. All of them are within the scope of the patent application of the present invention.

Claims (9)

一種固態硬碟控制寫入流量的方法,其步驟包含:控制減少內部資料的寫入流量;在同時寫入時段,同時寫入使用者資料及內部資料;檢查已完成寫入內部資料時,由同時寫入時段的使用者資料寫入流量,學習延遲時間;在單獨寫入時段,利用延遲時間控制減少使用者資料的寫入流量。  A method for controlling write traffic by a solid state hard disk, the method comprising: controlling to reduce write traffic of internal data; simultaneously writing user data and internal data during a simultaneous write period; and checking when writing internal data is completed At the same time, the user data of the writing period is written into the traffic, and the delay time is learned; in the separate writing period, the delay time control is used to reduce the writing traffic of the user data.   如申請專利範圍第1項所述之固態硬碟控制寫入流量的方法,其中該內部資料包含邏輯實體對照表的備份或垃圾回收資料區塊的有效資料頁。  The method of claim 1, wherein the internal data includes a backup of a logical entity comparison table or a valid data page of a garbage collection data block.   如申請專利範圍第1項所述之固態硬碟控制寫入流量的方法,其中將內部資料的寫入時間增加至預設延長寫入時間,以減少內部資料的寫入流量。  The solid-state hard disk control write flow method according to claim 1, wherein the write time of the internal data is increased to a preset extended write time to reduce the write flow of the internal data.   如申請專利範圍第3項所述之固態硬碟控制寫入流量的方法,其中利用固定或不固定的間隔時間,間斷式的寫入內部資料,使內部資料在預設延長寫入時間完成寫入。  The method for controlling write traffic by the solid state hard disk according to claim 3, wherein the internal data is intermittently written by using a fixed or unfixed interval, so that the internal data is written at a preset extended write time. In.   如申請專利範圍第3項所述之固態硬碟控制寫入流量的方法,其中在同時寫入時段,記錄內部資料的延長寫入時間及使用者資料的寫入資料量,由該寫入資料量除以延長寫入時間,計算使用者資料在同時寫入時段的寫入流量。  The method for controlling write traffic by the solid state hard disk according to claim 3, wherein in the simultaneous writing period, the extended writing time of the internal data and the written data amount of the user data are recorded, and the written data is written. The amount is divided by the extended write time to calculate the write traffic of the user data during the simultaneous write period.   如申請專利範圍第5項所述之固態硬碟控制寫入流量的方法,其中將單獨寫入時段使用者資料剩下的寫入資料量,除以使用者資料在同時寫入時段的寫入流量,學習出延遲時間。  A method for controlling write traffic by a solid state hard disk as described in claim 5, wherein the amount of write data remaining in the user data of the time period is separately divided, and the user data is written in the simultaneous write period. Traffic, learn the delay time.   如申請專利範圍第6項所述之固態硬碟控制寫入流量的方法,其中利用延 遲時間延長使用者資料的寫入時間。  A method of controlling write traffic by a solid state hard disk as described in claim 6 wherein the delay time is used to extend the write time of the user data.   如申請專利範圍第6項所述之固態硬碟控制寫入流量的方法,其中利用固定或不固定的間隔時間,間斷式的寫入使用者資料,讓使用者資料在延遲時間完成寫入。  The method of controlling the write traffic by the solid state hard disk according to claim 6 of the patent application, wherein the user data is intermittently written by using a fixed or unfixed interval, and the user data is written in the delay time.   如申請專利範圍第1項所述之固態硬碟控制寫入流量的方法,其中檢查未完成寫入內部資料時,繼續同時寫入使用者資料及內部資料。  The method for controlling the write traffic of the solid state hard disk according to the first aspect of the patent application, wherein when the internal data is not completed, the user data and the internal data are continuously written.  
TW106109235A 2017-03-17 2017-03-17 Method for controlling the writing flow of a solid state disk TW201835774A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW106109235A TW201835774A (en) 2017-03-17 2017-03-17 Method for controlling the writing flow of a solid state disk

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW106109235A TW201835774A (en) 2017-03-17 2017-03-17 Method for controlling the writing flow of a solid state disk

Publications (1)

Publication Number Publication Date
TW201835774A true TW201835774A (en) 2018-10-01

Family

ID=64797277

Family Applications (1)

Application Number Title Priority Date Filing Date
TW106109235A TW201835774A (en) 2017-03-17 2017-03-17 Method for controlling the writing flow of a solid state disk

Country Status (1)

Country Link
TW (1) TW201835774A (en)

Similar Documents

Publication Publication Date Title
US10191688B2 (en) Memory system and information processing system
US8140741B2 (en) Semiconductor storage device and control method thereof
US20150378888A1 (en) Controller, flash memory apparatus, and method for writing data into flash memory apparatus
US10860494B2 (en) Flushing pages from solid-state storage device
TWI759699B (en) Flash controllers, methods, and corresponding storage devices capable of rapidly/fast generating or updating contents of valid page count table
US20170228191A1 (en) Systems and methods for suppressing latency in non-volatile solid state devices
US10754785B2 (en) Checkpointing for DRAM-less SSD
TW201415233A (en) Fast execution of flush commands using adaptive compaction ratio
KR102106261B1 (en) Method of operating memory controller and methods for devices having same
US11385831B2 (en) Memory controller and storage device including the same
WO2019000982A1 (en) Storage system, solid state disk, and data storage method
US10635356B2 (en) Data management method and storage controller using the same
US20190042153A1 (en) Mass storage device capable of fine grained read and/or write operations
US20160110107A1 (en) Method for writing data into flash memory apparatus, flash memory apparatus, and storage system
TWI454922B (en) Memory storage device and memory controller and data writing method thereof
US9904622B2 (en) Control method for non-volatile memory and associated computer system
AU2015203828B2 (en) Implementing enhanced performance with read before write to phase change memory to avoid write cancellations
US11150809B2 (en) Memory controller and storage device including the same
CN110515861B (en) Memory device for processing flash command and method thereof
TW201835774A (en) Method for controlling the writing flow of a solid state disk
US10331385B2 (en) Cooperative write-back cache flushing for storage devices
TW202242664A (en) Method of performing garbage collection with partial clean operation and related controller and storage system
US20170153994A1 (en) Mass storage region with ram-disk access and dma access
US20200142623A1 (en) Storage device and electronic device managing data transmitted to storage device
TW201832085A (en) Operating method of memory system