CN101493795A - Storage system, storage controller, and cache implementing method in the storage system - Google Patents
Storage system, storage controller, and cache implementing method in the storage system Download PDFInfo
- Publication number
- CN101493795A CN101493795A CNA2008100568071A CN200810056807A CN101493795A CN 101493795 A CN101493795 A CN 101493795A CN A2008100568071 A CNA2008100568071 A CN A2008100568071A CN 200810056807 A CN200810056807 A CN 200810056807A CN 101493795 A CN101493795 A CN 101493795A
- Authority
- CN
- China
- Prior art keywords
- disk
- read
- storage system
- write
- buffer memory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Memory System Of A Hierarchy Structure (AREA)
Abstract
The invention discloses a storage system. The interior of a storage controller (SC) is provided with a high-speed disk which is used as a virtual buffer memory, so that the buffer memory capacity can be infinitely expanded through increasing the capacity of the high-speed disk and the size of the equipped virtual buffer memory; therefore, the capacity of the buffer memory can be flexibly expanded without the limits of system hardware; and simultaneously the problem of large cost of data paths caused by repeated data reading and writing during the process of reading and writing the buffer can not appear, the problem of competing resources of the buffer memory with a disk array in a disk cabinet can not appear, and the potential troubles of data safety can be eliminated. That is to say, the invention can realize the capacity expansion of the buffer memory and meanwhile, can avoid the problems with probability of resulting in lowering the performance of the storage system in the existing scheme, and therefore, the performance of the storage system can be improved by the capacity expansion of the buffer memory. The invention further discloses an SC in the storage system and a buffer memory realizing method in the storage system.
Description
Technical field
The present invention relates to caching technology, and the memory controller in particularly a kind of storage system, a kind of storage system (Storage Controller, SC) and the buffer memory implementation method in a kind of storage system.
Background technology
Be used for providing the storage system of service to form by SC and these two parts of disk chassis usually to network host, wherein, be provided with in the disk chassis Redundant Array of Independent Disks (RAID) formed by a plurality of member's disks (Redundant Array of Independent Disks, RAID).
Fig. 1 is the structural representation of existing storage system.As shown in Figure 1, SC is by internet small computer interface (Internet Small Computer System Interface, iSCSI) or optical-fibre channel (Fibre Channel, FC) interface connects disk chassis, SC in the storage system comprises: board chip set (Chip Set), and the central processing unit that links to each other with board chip set (CPU), RAID controller, system disk, as the internal memory of buffer memory.
Wherein, the RAID controller can be a hardware independently, is called hard RAID controller, also can be a software program that runs among the CPU, is called soft RAID controller; System disk can be regarded local disk as, and wherein having carried SC provides service required software and data, for example operating system; Internal memory as herein described is meant physical memory, and for example, (Double DataRate SDRAM DDR) waits memory bar to the Double Data Rate synchronous DRAM, in order to promote the response speed to network host.
SC also has the business interface that links to each other with board chip set, and for example (Fibre Channel, FC) interface, business interface are used for external client visit storage system for Gigabit Ethernet interface, optical-fibre channel.Wherein, client can be the main frame that links to each other with storage system by network, can abbreviate network host as.
CPU among the SC receives the write request or the read request of client according to business interface, calls the RAID controller and sends write request so that some data is write among the RAID to the RAID controller, or send read request to obtain the data of storing among the RAID.
For the write request that CPU sends, the RAID controller is written to the write data of RAID to be written in the internal memory earlier, realizes the buffer memory to write data, then when default refreshing arrives constantly, with all the write data batch refresh in the internal memory to RAID.
For the read request that CPU sends, the RAID controller is searched in internal memory earlier, if find corresponding data, promptly hits, then can directly read apace very much, otherwise, from RAID, search and read corresponding data.
In storage system as shown in Figure 1, cache algorithm is reasonable, access rule is reasonable, and cache size just becomes the principal element of decision performance of storage system so.
Yet the number of slots amount of the internal memory among the existing storage system SC is that the capacity of limited and every root memory bar also is limited, thereby makes that the buffer memory capacity that is made of internal memory in the storage system is limited, thereby has limited the performance of storage system.
Suppose CPU and board chip set the memory slot figure place that can support mostly be 4 most, then storage system the max cap. of the largest buffered capacity=memory slot figure place * every root memory bar that can support.The max cap. of every DDRII internal memory is 2GB at present, and the largest buffered capacity of storage system support then shown in Figure 1 is exactly 8GB, dilatation further again.
In order to improve buffer memory capacity more further to improve performance of storage system, can also be in the prior art with the part member's disk among the RAID as buffer memory.Referring to Fig. 1, each circular pattern in the disk chassis is represented member's disk among the RAID, and dark solid circles pattern is wherein represented the member's disk as buffer memory.Though this scheme can realize the further dilatation of buffer memory capacity, has following problem:
1, the data in buffer transmission channel is long.For write request, all write datas all must be written in the member's disk that is used as buffer memory among the RAID by the RAID controller by iSCSI interface or FC interface, refreshing when constantly arriving then, after the write data of buffer memory has been read back to the RAID controller by iSCSI interface or FC interface, be written in corresponding member's disk among the RAID by iSCSI interface or FC interface once more, make that the expense of FC/iSCSI data path is very big.In like manner, in advance with the data read among the RAID to as in member's disk of buffer memory so that follow-up during according to the read request reading of data, also have same problem.Under such expense, even be that (its advantage at a high speed also can be cancelled for Solid State Disks, high speed disk such as SSD) for serial SCSI (SAS) disk or solid magnetic disc as member's disk of buffer memory.
2, buffer memory and RAID competitive resource: buffer memory and shared identical RAID controller of RAID and FC/iSCSI data path the situation of competitive resource can occur, thereby reduce the performance of storage system.
3, there is hidden danger in data security.If a plurality of disk chassis are arranged in the system, during as the residing disk chassis power down of member's disk of buffer memory, the data in the buffer memory can be lost, and can't be to the realization buffer memory that reads and writes data of RAID in other disk chassis, thereby reduce the performance of storage system.
As seen, the purpose of buffer memory dilatation is in order to improve the performance of storage system, and have now the member's disk among the RAID as the scheme of buffer memory with the dilatation of realization buffer memory, can cause the problem of various reduction performance of storage system to occur, thereby this scheme does not improve the performance of storage system by the dilatation of buffer memory.
Summary of the invention
In view of this, the invention provides SC in a kind of storage system, a kind of storage system and the buffer memory implementation method in a kind of storage system, can improve the performance of storage system by buffer memory dilatation flexibly.
A kind of storage system provided by the invention is used for providing service to network host, and this system comprises: memory controller SC, comprise the disk array of at least one low speed disk,
It is characterized in that comprise the hyperdisk as virtual cache among the described SC, its read or write speed is used for the data of the described disk array of the described network host read/write of buffer memory greater than described low speed disk.
Also comprise the local disk of read or write speed among the described SC, be used to deposit software and the data of described storage system in order to service to be provided less than described hyperdisk.
Also comprise the physical memory as buffer memory among the described SC, its read or write speed is greater than described hyperdisk.
Also comprise among the described SC: the virtual cache starter gear is used to control the opening and closing of described virtual cache.
SC in a kind of storage system provided by the invention is applied to network host and comprises between the disk array of at least one low speed disk, and provides service to network host, it is characterized in that,
Described SC comprises: as the hyperdisk of virtual cache, its read or write speed is used for the data of the described disk array of the described network host read/write of buffer memory greater than described low speed disk.
Also comprise the local disk of read or write speed among the described SC, be used to deposit software and the data of described SC in order to service to be provided less than described hyperdisk.
Also comprise the physical memory as buffer memory among the described SC, its read or write speed is greater than described hyperdisk.
Also comprise among the described SC: the virtual cache starter gear is used to control the opening and closing of described virtual cache.
Described virtual cache starter gear when read mode is visited described disk array at random, cuts out described virtual cache at described network host; When described network host is visited described disk array by the order read mode, open described virtual memory.
The described hyperdisk that is used as virtual cache is more than one, and described hyperdisk more than constitutes the 0th grade of other Redundant Array of Independent Disks (RAID) RAID0.
Buffer memory implementation method in a kind of storage system provided by the invention, described storage system is used for providing service to network host, and this system comprises: memory controller SC, comprise the disk array of at least one low speed disk, this method comprises:
Read or write speed is set greater than described low speed disk and as the hyperdisk of virtual cache at the memory controller SC of described storage system,
With the metadata cache of the described disk array of network host read/write in described hyperdisk.
If described network host is visited described disk array by read mode at random, then close described virtual cache;
If described network host is visited described disk array by the order read mode, then open described virtual memory.
The invention also discloses a kind of NSC, comprising:
Front-end interface is coupled by network and network host;
Back end interface is coupled with disk array;
Buffer memory is in order to promote network host request responding speed;
Local disk is in order to the necessary software of storage networking controller operation;
Described NSC also comprises the disk as virtual cache, and in order to promote network host request responding speed, wherein the read-write channel of the read-write channel of this disk and local disk is separate.
Described disk as virtual cache, its interface type is different from the interface type of local disk interface, and its read or write speed is greater than the local disk read or write speed.
Described disk as virtual cache, its read or write speed is greater than the read or write speed of disk array.
As seen from the above technical solution, the present invention is provided with hyperdisk in SC inside, and the hyperdisk that is provided with is used as virtual cache, makes that buffer memory capacity can the unlimited dilatation by the size of increase hyperdisk capacity and the virtual cache that is disposed.Like this, can not be subjected to the restriction of system hardware to expand the capacity of buffer memory neatly, simultaneously since data in buffer without the FC/iSCSI data path, do not carry out reading and writing data repeatedly in the read-write cache process and cause the very big problem of FC/iSCSI data path expense and can not appear at, the problem that buffer memory and RAID competitive resource also can not occur, the hidden danger that data security exists also can be eliminated.That is to say that the present invention can be when realizing the buffer memory dilatation, avoid the problem of the various reduction performance of storage system that may cause in the existing scheme, thereby can improve performance of storage system by the buffer memory dilatation.
And in the inner realization of SC, can there be the phase mutual interference in read-write cache between each components and parts among the SC fully yet; SC can also select to open or close virtual cache according to the access module of client, makes the user mode of virtual cache can be adaptive to different access modules; Can also be as the hyperdisk of virtual cache more than one, and constitute the 0th RAID level (RAID0), thus can further improve the speed of read-write cache, with the performance of further raising storage system.
Description of drawings
Fig. 1 is the structural representation of existing storage system.
Fig. 2 is the structural representation of storage system among the present invention.
Fig. 3 is the exemplary process diagram of the buffer memory implementation method in the storage system of the present invention.
Embodiment
For making purpose of the present invention, technical scheme and advantage clearer, below with reference to the accompanying drawing embodiment that develops simultaneously, the present invention is described in more detail.
The present invention is provided with hyperdisk in SC inside, and the hyperdisk that is provided with is used as virtual cache, makes that buffer memory capacity can the unlimited dilatation by the size of increase hyperdisk capacity and the virtual cache that is disposed.
Like this, neither can occur having now with the member's disk among the RAID as in the scheme of buffer memory with the dilatation of realization buffer memory, in the read-write cache process data repeatedly through and cause the very big problem of FC/iSCSI data path expense, the problem that buffer memory and RAID competitive resource also can not occur, the hidden danger that data security exists also can be eliminated.That is to say, the present invention can be when realizing the buffer memory dilatation, avoid the problem of the various reduction performance of storage system that may cause in the existing scheme, thereby can improve the bandwidth and the IOPS (I/O Per Second) of storage system, promptly improve performance of storage system by the buffer memory dilatation.
Need to prove that hyperdisk of the present invention is non-volatile storage medium, described " at a high speed " is not based on absolute standard and definite.In fact described " at a high speed " be meant that read or write speed is faster with respect to each member's disk among the system disk of regarding local disk as and the RAID, and the height of read or write speed can be differentiated according to the rotating speed of disk and the interface rate of disk usually.That is to say,, regard that each member's disk then can be described as the low speed disk among the system disk of local disk and the RAID as with respect to described " at a high speed ".
And because physical medium, the read or write speed of hyperdisk can be lower than physical memory usually.But in order to guarantee the efficient of read-write cache, preferably should select read or write speed near the hyperdisk of physical memory as virtual cache.
Certainly, the present invention not merely is the hyperdisk that is provided with in SC simply as virtual cache, and along with having increased hyperdisk among the SC, interactive mode among the SC between each components and parts and inter-process mode are also improved thereupon, below are elaborated.
Fig. 2 is the structural representation of storage system among the present invention.As shown in Figure 2, be used to network host to provide the SC in the storage system of service to comprise: board chip set, and the CPU that links to each other with board chip set, RAID controller, local disk be system disk, as the internal memory of buffer memory.The RAID that is arranged in the storage system disk chassis is made of the low speed disk.
Than existing storage system shown in Figure 1, SC in the storage system of the present invention also comprises the hyperdisk as virtual cache, this hyperdisk can be any hyperdisks such as SAS disk or SSD, and its read or write speed is higher than the member's disk among local disk and the RAID.
That is to say, the present invention's storage system as shown in Figure 2, difference than as shown in Figure 1 existing storage system is: comprise the hyperdisk as virtual cache among the SC, its read or write speed is used for the data of buffer memory network host read/write RAID greater than the low speed disk.
Specifically, CPU among the SC receives client, is the write request or the read request of network host according to business interface, call the RAID controller and send write request so that some data is write among the RAID, or send read request to obtain the data of storing among the RAID to the RAID controller.
For the write request that CPU sends, the RAID controller is written to the write data of RAID to be written in the internal memory when internal memory has idle storage space; When memory headroom is not enough, the write data of RAID to be written is written in the hyperdisk; When default refreshing arrives constantly, with the write data batch refresh in internal memory and the hyperdisk to RAID.
Above-mentionedly preferentially be: because the difference of physical medium with write data buffer memory to the reason of internal memory, the read or write speed of internal memory usually will be faster than hyperdisk, therefore, write data priority cache to internal memory can be guaranteed to write buffer efficiency to greatest extent, thereby guarantee the performance of storage system to greatest extent.Certainly, all can know for those skilled in the art, along with the development of hardware technology, physical medium as hyperdisk might make the read or write speed of hyperdisk be higher than internal memory, therefore, if the read or write speed of hyperdisk is higher than internal memory, then preferentially with the write data buffer memory to the high speed disk.
Certainly, even if the read or write speed of hyperdisk is not so good as internal memory, but because its read or write speed is higher than the member's disk among local disk and the RAID, thereby also the hyperdisk that is used as virtual cache can only be set in system, and do not need to be provided with again internal memory.
For the read request that CPU sends, the RAID controller is searched in internal memory and/or hyperdisk earlier, if find corresponding data, promptly hits, then can directly read apace very much, otherwise, from RAID, search and read corresponding data.
In the storage system as shown in Figure 2, local disk, be system disk can pass through integrated drive electronics equipment (Integrated Divers Electronics, IDE) interface links to each other with board chip set; Hyperdisk can based on the FC interface of two-forty or serial ATA (Serial Advanced TechnologyAttachment, SATA) interface links to each other with board chip set; RAID controller, business interface etc. then link to each other with board chip set based on Peripheral Component Interconnect standard (PCI) bus.Because each components and parts take different interface buss, thereby can guarantee enough processing poweies are arranged, and avoid the phase mutual interference that the contention resource causes between each components and parts as the hyperdisk of virtual cache.
This shows, the hyperdisk that is used as virtual cache among the present invention is physical disk independently, and the read-write channel of other components and parts such as this hyperdisk and local disk is separate, rather than the part of local disk is set to virtual cache, except the read or write speed of considering local disk is hanged down, a part of also considering local disk is set to virtual cache can cause CPU and the RAID controller unique interface by local disk, promptly simultaneously local disk is conducted interviews, thereby cause access conflict to influence system performance by unique read-write channel.
In the practical application, because CPU sends to the read-write requests of RAID controller usually from the network host as client, and network host may visit storage system by the different access pattern, thereby the present invention can also carry out corresponding processing at the different access pattern, with the performance of further raising storage system.
If the represented access module of the read request that CPU sends is for reading at random, consider that then the hit rate that reads at random is lower, thereby the RAID controller only searches corresponding data in internal memory, and the virtual cache of closing high-speed disk, i.e. read-write channel between shielding and the hyperdisk.
If the represented access module of the read request that CPU sends reads for order, then the RAID controller is opened the virtual cache of hyperdisk, i.e. read-write channel between gating and the hyperdisk, and in internal memory and hyperdisk, search corresponding data.When reading, can also carry out data read according to default pre-fetch policy, for example the data adjacent with these data that need read also be read in the buffer memory, thereby improve the follow-up cache hit rate of reading, promptly improve the speed that buffer memory is read in the storage system response, further improve performance of storage system.
That is to say that RAID can regard the starter gear of virtual cache as, be used to control the opening and closing of virtual cache.Certainly, another independently physical controller or increase by a software program realize control to the opening and closing of virtual cache in CPU also can be set.
In order to improve the speed of read-write cache, with the performance of further raising storage system, in the storage system as shown in Figure 2, can also be as the hyperdisk of virtual cache more than one, and constitute RAID0.Like this, when writing data, write data can be divided into data block that hyperdisk quantity equates after write to different hyperdisks respectively, thereby it is parallel to make that write operation is changed into by serial, improves the speed of writing buffer memory; In like manner, read buffer memory also can change into by serial parallel, thereby improve the speed read buffer memory.
Certainly in order to guarantee the reliability of virtual cache, also can constitute RAID, for example the 1st RAID level (RAID1), the 5th RAID level the 5, the 6th RAID level 6 etc. by a plurality of hyperdisks with redundance.
In the practical application, ordinary magnetic disc array with redundance also can be set in the disk chassis, be that RAID is not a necessary condition, therefore, RAID in the storage system of the present invention also can replace with does not have ordinary magnetic disc array redundance, that comprise at least one low speed disk, at this moment, the RAID controller among the SC of storage system can replace with the controller that corresponding realization is read and write disk array.
It more than is detailed description to SC in storage system of the present invention and the storage system.Below, again the buffer memory implementation method in the storage system of the present invention is described.
Buffer memory implementation method in the storage system of the present invention comprises: read or write speed is set greater than the low speed disk that constitutes RAID and as the hyperdisk of virtual cache at the SC of storage system, and with the metadata cache of network host read/write RAID in hyperdisk.
Fig. 3 is the exemplary process diagram of the buffer memory implementation method in the storage system of the present invention.As shown in Figure 3, the buffer memory implementation method in the storage system of the present invention comprises:
Step 301 is provided with in the SC of storage system and links to each other with board chip set and as the hyperdisk of virtual cache, makes to comprise as the internal memory of buffer memory and the hyperdisk that is used as virtual cache among the SC.
Step 302 judges to receive write request or read request that if write request, then execution in step 303, if read request execution in step 304 then.
Step 303 is according to the write request that receives, when internal memory has idle storage space, write data is written in the internal memory, when memory headroom is not enough, write data is written in the described hyperdisk, and process ends or return step 302 and continue the processing of visit next time.
After this step,, then the write data in internal memory and the hyperdisk is refreshed the RAID to the storage system if arrive default refreshing constantly.
In this step, if the represented access module of read request for reading at random, is then only searched corresponding data in described internal memory; If the represented access module of read request reads for order, then in described internal memory and described hyperdisk, search corresponding data.
So far, this flow process finishes.
By above-mentioned flow process as seen, this method is when read-write cache, the very big problem of FC/iSCSI data path expense neither can appear, the problem that buffer memory and RAID competitive resource also can not occur, the hidden danger that data security exists also can be eliminated, thereby when realizing the buffer memory dilatation, avoid the problem of the various reduction performance of storage system that may cause in the existing scheme, improved performance of storage system.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to and replace and improvement etc., all should be included within protection scope of the present invention.
Claims (15)
1, a kind of storage system is used for providing service to network host, and this system comprises: memory controller SC, comprise the disk array of at least one low speed disk,
It is characterized in that comprise the hyperdisk as virtual cache among the described SC, its read or write speed is used for the data of the described disk array of the described network host read/write of buffer memory greater than described low speed disk.
2, the system as claimed in claim 1 is characterized in that, also comprises the local disk of read or write speed less than described hyperdisk among the described SC, is used to deposit software and the data of described storage system in order to service to be provided.
3, system as claimed in claim 1 or 2 is characterized in that, also comprises the physical memory as buffer memory among the described SC, and its read or write speed is greater than described hyperdisk.
4, system as claimed in claim 3 is characterized in that, also comprises among the described SC: the virtual cache starter gear is used to control the opening and closing of described virtual cache.
5, the memory controller SC in a kind of storage system is applied to network host and comprises between the disk array of at least one low speed disk, and provides service to network host, it is characterized in that,
Described SC comprises: as the hyperdisk of virtual cache, its read or write speed is used for the data of the described disk array of the described network host read/write of buffer memory greater than described low speed disk.
6, SC as claimed in claim 5 is characterized in that, also comprises the local disk of read or write speed less than described hyperdisk among the described SC, is used to deposit software and the data of described SC in order to service to be provided.
As claim 5 or 6 described SC, it is characterized in that 7, also comprise the physical memory as buffer memory among the described SC, its read or write speed is greater than described hyperdisk.
8, SC as claimed in claim 7 is characterized in that, also comprises among the described SC: the virtual cache starter gear is used to control the opening and closing of described virtual cache.
9, system as claimed in claim 8 is characterized in that,
Described virtual cache starter gear when read mode is visited described disk array at random, cuts out described virtual cache at described network host; When described network host is visited described disk array by the order read mode, open described virtual memory.
10, SC as claimed in claim 5 is characterized in that, the described hyperdisk that is used as virtual cache is more than one, and described hyperdisk more than constitutes the 0th grade of other Redundant Array of Independent Disks (RAID) RAID0.
11, the buffer memory implementation method in a kind of storage system, described storage system is used for providing service to network host, and this system comprises: memory controller SC, comprise and the disk array of at least one low speed disk it is characterized in that this method comprises:
Read or write speed is set greater than described low speed disk and as the hyperdisk of virtual cache at the memory controller SC of described storage system,
With the metadata cache of the described disk array of network host read/write in described hyperdisk.
12, method as claimed in claim 11 is characterized in that,
If described network host is visited described disk array by read mode at random, then close described virtual cache; If described network host is visited described disk array by the order read mode, then open described virtual memory.
13, a kind of NSC comprises:
Front-end interface is coupled by network and network host;
Back end interface is coupled with disk array;
Buffer memory is in order to promote network host request responding speed;
Local disk is in order to the necessary software of storage networking controller operation;
It is characterized in that described NSC also comprises the disk as virtual cache, in order to promote network host request responding speed, wherein the read-write channel of the read-write channel of this disk and local disk is separate.
14, controller as claimed in claim 13 is characterized in that, described disk as virtual cache, and its interface type is different from the interface type of local disk interface, and its read or write speed is greater than the local disk read or write speed.
15, as claim 13 or 14 described controllers, it is characterized in that, described disk as virtual cache, its read or write speed is greater than the read or write speed of disk array.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100568071A CN101493795A (en) | 2008-01-24 | 2008-01-24 | Storage system, storage controller, and cache implementing method in the storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100568071A CN101493795A (en) | 2008-01-24 | 2008-01-24 | Storage system, storage controller, and cache implementing method in the storage system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101493795A true CN101493795A (en) | 2009-07-29 |
Family
ID=40924403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008100568071A Pending CN101493795A (en) | 2008-01-24 | 2008-01-24 | Storage system, storage controller, and cache implementing method in the storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101493795A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096556A (en) * | 2010-12-03 | 2011-06-15 | 成都市华为赛门铁克科技有限公司 | Method for copying data as well as method, device and system for reading data |
| CN102185874A (en) * | 2011-01-19 | 2011-09-14 | 杭州华三通信技术有限公司 | Method and device for processing commands based on iSCSI (internet small computer system interface) |
| CN102446261A (en) * | 2011-11-15 | 2012-05-09 | 浪潮电子信息产业股份有限公司 | Method for increasing writing efficiency of RAID (Redundant Array of Independent Disks) |
| CN103077149A (en) * | 2013-01-09 | 2013-05-01 | 厦门市美亚柏科信息股份有限公司 | Method and system for transmitting data |
| CN103631537A (en) * | 2013-11-29 | 2014-03-12 | 华为技术有限公司 | Method and device for managing virtual disk |
| CN103677155A (en) * | 2012-09-21 | 2014-03-26 | 英业达科技有限公司 | Server and server host |
| CN104375578A (en) * | 2014-11-24 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | High-speed large-capacity cache memory card |
| WO2015100653A1 (en) * | 2013-12-31 | 2015-07-09 | 华为技术有限公司 | Data caching method, device and system |
| CN105205017A (en) * | 2015-08-31 | 2015-12-30 | 浪潮(北京)电子信息产业有限公司 | Storage controller based on PCIE SSD |
| CN105743671A (en) * | 2014-12-10 | 2016-07-06 | 华为技术有限公司 | Capacity expanding method and system, and controller |
| CN107391037A (en) * | 2017-07-25 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of method and device of memory system data |
| CN107800730A (en) * | 2016-08-30 | 2018-03-13 | 阿里巴巴集团控股有限公司 | The expansion method and device of a kind of virtual disk |
| CN110196687A (en) * | 2019-05-20 | 2019-09-03 | 杭州宏杉科技股份有限公司 | Data read-write method, device, electronic equipment |
| CN113625956A (en) * | 2021-06-30 | 2021-11-09 | 苏州浪潮智能科技有限公司 | Cache setting method and device for RAID card, electronic equipment and storage medium |
-
2008
- 2008-01-24 CN CNA2008100568071A patent/CN101493795A/en active Pending
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096556A (en) * | 2010-12-03 | 2011-06-15 | 成都市华为赛门铁克科技有限公司 | Method for copying data as well as method, device and system for reading data |
| CN102096556B (en) * | 2010-12-03 | 2013-11-06 | 华为数字技术（成都）有限公司 | Method for copying data as well as method, device and system for reading data |
| CN102185874A (en) * | 2011-01-19 | 2011-09-14 | 杭州华三通信技术有限公司 | Method and device for processing commands based on iSCSI (internet small computer system interface) |
| CN102185874B (en) * | 2011-01-19 | 2014-04-02 | 浙江宇视科技有限公司 | Method and device for processing commands based on iSCSI (internet small computer system interface) |
| CN102446261A (en) * | 2011-11-15 | 2012-05-09 | 浪潮电子信息产业股份有限公司 | Method for increasing writing efficiency of RAID (Redundant Array of Independent Disks) |
| CN103677155A (en) * | 2012-09-21 | 2014-03-26 | 英业达科技有限公司 | Server and server host |
| CN103077149A (en) * | 2013-01-09 | 2013-05-01 | 厦门市美亚柏科信息股份有限公司 | Method and system for transmitting data |
| CN103631537A (en) * | 2013-11-29 | 2014-03-12 | 华为技术有限公司 | Method and device for managing virtual disk |
| CN103631537B (en) * | 2013-11-29 | 2016-09-14 | 华为技术有限公司 | The management method of a kind of virtual disk and device |
| WO2015100653A1 (en) * | 2013-12-31 | 2015-07-09 | 华为技术有限公司 | Data caching method, device and system |
| CN104375578A (en) * | 2014-11-24 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | High-speed large-capacity cache memory card |
| CN105743671B (en) * | 2014-12-10 | 2020-07-14 | 华为技术有限公司 | Capacity expansion method, controller and system |
| CN105743671A (en) * | 2014-12-10 | 2016-07-06 | 华为技术有限公司 | Capacity expanding method and system, and controller |
| CN105205017A (en) * | 2015-08-31 | 2015-12-30 | 浪潮(北京)电子信息产业有限公司 | Storage controller based on PCIE SSD |
| CN107800730A (en) * | 2016-08-30 | 2018-03-13 | 阿里巴巴集团控股有限公司 | The expansion method and device of a kind of virtual disk |
| CN107800730B (en) * | 2016-08-30 | 2021-01-29 | 阿里巴巴集团控股有限公司 | Capacity expansion method and device for virtual disk |
| CN107391037A (en) * | 2017-07-25 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of method and device of memory system data |
| CN110196687A (en) * | 2019-05-20 | 2019-09-03 | 杭州宏杉科技股份有限公司 | Data read-write method, device, electronic equipment |
| CN110196687B (en) * | 2019-05-20 | 2023-02-03 | 杭州宏杉科技股份有限公司 | Data reading and writing method and device and electronic equipment |
| CN113625956A (en) * | 2021-06-30 | 2021-11-09 | 苏州浪潮智能科技有限公司 | Cache setting method and device for RAID card, electronic equipment and storage medium |
| CN113625956B (en) * | 2021-06-30 | 2023-08-22 | 苏州浪潮智能科技有限公司 | Cache setting method and device of RAID card, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101493795A (en) | Storage system, storage controller, and cache implementing method in the storage system | |
| US9798472B1 (en) | Extent level cache destaging | |
| CN107656700B (en) | FLASH-DRAM hybrid memory module | |
| US10152423B2 (en) | Selective population of secondary cache employing heat metrics | |
| JP4866646B2 (en) | How to select commands to send to memory, memory controller, computer system | |
| CN101566927B (en) | Memory system, memory controller and data caching method | |
| US20120290786A1 (en) | Selective caching in a storage system | |
| CN102117248A (en) | Caching system and method for caching data in caching system | |
| CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
| CN109213693B (en) | Storage management method, storage system and computer program product | |
| US20130219122A1 (en) | Multi-stage cache directory and variable cache-line size for tiered storage architectures | |
| CN105897859B (en) | Storage system | |
| CN103076993A (en) | Storage system and method for concentration type system | |
| US7725654B2 (en) | Affecting a caching algorithm used by a cache of storage system | |
| CN113760796B (en) | SSD solid state disk based on HBM buffering | |
| CN110659305A (en) | High performance relational database service based on non-volatile storage system | |
| US9244693B2 (en) | System and method for booting multiple servers from snapshots of an operating system installation image while reducing access to unmodified portions of the installation image and reducing latencies between server ports | |
| US8108605B2 (en) | Data storage system and cache data—consistency assurance method | |
| US10942857B1 (en) | Dynamically adjusting a number of memory copy and memory mapping windows to optimize I/O performance | |
| US11016692B2 (en) | Dynamically switching between memory copy and memory mapping to optimize I/O performance | |
| EP4081905A1 (en) | Staging memory access requests | |
| CN104424124A (en) | Memory device, electronic equipment and method for controlling memory device | |
| EP4081906A1 (en) | Staging buffer arbitration | |
| CN101859232A (en) | Variable length data memory interface | |
| CN105353978B (en) | A kind of data of PCIE SSD arrays read method, system and reading/writing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20090729 |