WO2009027398A1 - Mass storage system with improved usage of buffer capacity - Google Patents

Mass storage system with improved usage of buffer capacity Download PDF

Info

Publication number
WO2009027398A1
WO2009027398A1 PCT/EP2008/061148 EP2008061148W WO2009027398A1 WO 2009027398 A1 WO2009027398 A1 WO 2009027398A1 EP 2008061148 W EP2008061148 W EP 2008061148W WO 2009027398 A1 WO2009027398 A1 WO 2009027398A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
mass storage
real
storage array
time
Prior art date
Application number
PCT/EP2008/061148
Other languages
French (fr)
Inventor
Stefan Abeling
Johann Maas
Wolfgang Klausberger
Thomas Brune
Original Assignee
Thomson Licensing
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 Thomson Licensing filed Critical Thomson Licensing
Priority to US12/733,338 priority Critical patent/US8285932B2/en
Priority to EP08787494A priority patent/EP2183663B1/en
Priority to CN2008801051342A priority patent/CN101796479B/en
Priority to KR20107003596A priority patent/KR101491687B1/en
Priority to JP2010522347A priority patent/JP5244909B2/en
Publication of WO2009027398A1 publication Critical patent/WO2009027398A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0658—Controller construction arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0608—Saving storage space on storage systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0656—Data buffering arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671—In-line storage system
    • G06F3/0683—Plurality of storage devices
    • G06F3/0689—Disk arrays, e.g. RAID, JBOD

Definitions

  • the present invention relates to a mass storage system with improved usage of buffer capacity, and more specifically to a mass storage system for real-time data storage with an embedded controller.
  • Embedded stand-alone systems designed for storing real-time data streams with a high data rate coming directly from a data stream source can generally be split up into two kinds of data processing:
  • Non real-time data storage such as file system information managed by an embedded processor.
  • Both data paths have to be connected to an I/O controller, which implements the I/O communication with storage units forming a mass storage array.
  • a high guaranteed data rate is reached by parallelization of single storage units such as hard disks, optical disks, or solid state based memories.
  • the above types of storage units often introduce significant latency at the start of a transfer. Since latency is a problem especially when real-time behavior is needed, memory with no or at least no significant access latency such as SRAM and/or SDRAM is introduced. Other requirements, particularly for mobile applications, are that the whole storage system has to be as small as possible and to have a reduced power consumption.
  • this object is achieved by a mass storage system having a first data path between a real-time data interface and a mass storage array, the first data path including a data buffer without access latency, and a second data path between an embedded processor and the mass storage array, wherein the data buffer without access latency is also used as a data buffer for non real-time data transfers between the embedded processor and the mass storage array.
  • the invention proposes a multi-purpose use of the 'no access latency' memory unit in the real-time data path. This is a more efficient use of the memory and additionally a more efficient way to transfer non real-time data between an embedded processor and a mass storage array consisting of single storage devices.
  • the invention is not only applicable to systems designed for real-time data. It can likewise be used in other environments where a high guaranteed data rate is needed and latency is a problem.
  • the 'no access latency' memory is integrated into the real-time data path as a realtime data buffer to compensate any latencies of the target mass storage array.
  • the mass storage array consists of two or more single storage units. Each of the storage units is controlled by an I/O controller. Since also the embedded processor has to access and modify data on the mass storage array, there is a second path implemented between the real-time data buffer and the main memory of the processor. Therefore, a data switch is implemented for switching between data from or to the real-time interface and data from or to the embedded processor. Using the necessary real-time data buffer also for non real-time data transfers between the embedded processor and the mass storage array saves an extra buffer for non real-time transfers. Therefore, space and power consumption is reduced.
  • the hardware blocks are preferably configured by the embedded processor through a control line to be in the required mode.
  • the embedded processor transfers data from and to the mass storage array by initiating a direct memory access transfer (DMA transfer) between its main memory and the real-time data buffer.
  • DMA transfer direct memory access transfer
  • access of the embedded processor to the mass storage array via the real-time data buffer is faster than accessing the mass storage array directly.
  • the mass storage system has a data processing block for splitting up a data stream and sending it in parallel to the two or more storage units. This allows to easily implement RAID or RAID-like systems.
  • one or more thresholds can be set for the 'no access latency' data buffer, the one or more thresholds controlling the instant when data are streamed out of the data buffer to the mass storage array, to the real-time interface, and to the main memory. In this way the performance behavior of the different data targets can be adapted.
  • Fig. 1 shows an exemplary mass storage system according to the invention.
  • Fig. 1 shows the architectural design of a stand-alone mass storage system 1 with an embedded processor 2 for real-time data streams in form of a block diagram.
  • a real-time interface block 3 is connected to a real-time data source or sink (not shown) , i.e. a camera or a display.
  • the embedded processor 2 reacts on user input like 'record video' or 'playback video'.
  • the embedded processor 2 configures the hardware blocks to be in the requested mode through control connections 4 (indicated by the dotted lines) .
  • control connections 4 indicated by the dotted lines
  • a simple 'mode' register is modified by the embedded processor 2 in a write access, which configures the hardware blocks.
  • the embedded processor 2 For a real-time data streaming demand the embedded processor 2 first configures the hardware blocks.
  • a data switch 5 selects the real-time interface for data streaming.
  • a plurality of I/O control units 6 receive information about the direction (read or write) and the storage address of the real-time data stream. The latter is independent of the direction and the kind of data stream.
  • the main difference between real-time data and non real-time data is the large amount of real-time data that is to be expected in comparison to the usually rather small amount of non real-time data. In order to cope with this difference the number of bytes to be transferred is included in the read/write command to the I/O control units 6.
  • the I/O control units 6 allow to read/write an amount of data that is sufficiently large to ensure that the embedded processor 2 is able to generate and send the next read/write command. This is necessary to guarantee the real-time data rate, as a relatively large amount of time passes between two read/write commands of the embedded processor 2.
  • the amount of data allowed to be read/written with a single read/write command is rather small, as often only a few data have to be transferred.
  • the real-time data stream passes the real-time interface 3, the data switch 5, a real-time data buffer 7 and a data processing unit 8, where the real-time data stream is split up and sent in parallel to the I/O control units 6.
  • the I/O control units 6 write the data to a plurality of storage units 9 of a mass storage array 10. As long as the storage units 9 of the mass storage array 10 are not able to receive data, the real-time data stream is stored into the real- time data buffer 7.
  • the data transfer to the mass storage array 10 continues when the storage units 9 are ready to receive data. For example, such latencies arise if hard disks have to reposition their read/write heads. Due to the mechanical operations a data transfer is paused. Only after finishing the positioning procedure the hard disks are able to run at their media transfer data rate.
  • the data are read in parallel from the storage units 9 and are written into the real-time data buffer 7 after passing the data processing unit 8, where the single data streams are merged into one real-time data stream.
  • the transmission of the real-time data stream through the realtime interface 3 to a data sink is preferably initiated only when the real-time data buffer 7 is full. This measure helps to compensate for latencies produced by the mass storage array 10, and to fulfill the real-time requirements. This behavior is advantageously changeable by setting a threshold, which will be explained later.
  • a non real-time data transfer is executed.
  • the main difference to a real-time data stream is that the real-time interface 3 is logically replaced by the processor's main memory 11. Data are transferred from the main memory 11 to the real-time data buffer 7 in case data are written to the mass storage array 10, or from the real-time data buffer 7 to the main memory 11 in case data are read from the storage array 10. Therefore, a direct memory access (DMA) transfer is initiated by the embedded processor 2 between the main memory 11 and the real-time data buffer 7.
  • DMA direct memory access
  • the data transfer from the real-time data buffer 7 to the mass storage array 10 is automatically executed. Consequently, after initializing the DMA transfer the embedded processor 2 is free to work on other tasks until a message of a successful or unsuccessful DMA transfer to the mass storage array 10 is received.
  • latencies of the mass storage array 10 are compensated for by the real-time data buffer 7 as if a real-time data stream was processed. This offers a higher performance than direct single word accesses of the embedded processor 2 to the mass storage array 10.
  • a further improvement is that the control mechanism for a non real-time data stream is the same as for a real-time data stream.
  • the data switch 6 has to be configured differently to replace the realtime interface 3 with the main memory 11.
  • the amount of data to be written with a single read/write command is modified.
  • the amount of data per command is relatively small, but can likewise be adapted to the needs of the embedded processor 2, so that each DMA transfer is performed with a single command to the I/O control units 6.
  • a parallelization of the data is executed in the data processing unit 8, allowing to write to, respectively to read from, all single storage units 9 of the mass storage array 10 in parallel.
  • Other RAID or RAID-like systems can be implemented as well.
  • Thresholds of the real-time data buffer 7 can be dynamically configured by the embedded processor 2, preferably even during two accesses to the mass storage array 10 or when different mass storage units 9 are added to the mass storage array 10.
  • the thresholds control the instant when data are streamed out of the real-time data buffer 7 to the mass storage array 10, to the real-time interface 3, or to the main memory 11. In this way the performance behavior of the different data targets can be adapted. For example, a DMA transfer from the real-time data buffer 7 to the main memory 11 is first initiated when there are at least a number 'n' of bytes available in the real-time data buffer 7, the number 'n' being determined by a special threshold.
  • the main memory 11 is locked by the DMA transfer as short as possible and the embedded processor 2 has only to perform its tasks without its main memory 11 for the same short time.
  • the dynamic modification of the thresholds allows to optimally adapt the system to different types of mass storage units 9 or to different types of data accesses, such as data streaming or transmission of smaller data blocks. For example, for some types of solid state disks a pause in the data transfer at an unfavorable instant leads to a drop in the data rate. During recording this is compensated by first collecting data in the real-time data buffer 7 before transferring them to the mass storage units 9. Though this reduces the latencies that can be compensated, the overall data rate is increased.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The present invention relates to a mass storage system (1) with improved usage of buffer capacity, and more specifically to a mass storage system (1) for real-time data storage with an embedded controller (2). According to the invention, the mass storage system (1) has a first data path between a real-time data interface (3) and a mass storage array (10), the first data path including a data buffer (7) without access latency, and a second data path between an embedded processor (2) and the mass storage array (10), wherein the data buffer (7) without access latency is also used as a data buffer for non real-time data transfers between the embedded processor (2) and the mass storage array (10).

Description

MASS STORAGE SYSTEM WITH IMRPOVED USAGE OF BUFFER CAPACITY
The present invention relates to a mass storage system with improved usage of buffer capacity, and more specifically to a mass storage system for real-time data storage with an embedded controller.
Embedded stand-alone systems designed for storing real-time data streams with a high data rate coming directly from a data stream source, e.g. a film camera, can generally be split up into two kinds of data processing:
1. Real-time video data storage with a guaranteed data rate, and
2. Non real-time data storage such as file system information managed by an embedded processor.
Both data paths have to be connected to an I/O controller, which implements the I/O communication with storage units forming a mass storage array.
A high guaranteed data rate is reached by parallelization of single storage units such as hard disks, optical disks, or solid state based memories. The above types of storage units often introduce significant latency at the start of a transfer. Since latency is a problem especially when real-time behavior is needed, memory with no or at least no significant access latency such as SRAM and/or SDRAM is introduced. Other requirements, particularly for mobile applications, are that the whole storage system has to be as small as possible and to have a reduced power consumption.
It is an object of the invention to propose a mass storage system with an improved usage of the available buffer capacity.
According to the invention, this object is achieved by a mass storage system having a first data path between a real-time data interface and a mass storage array, the first data path including a data buffer without access latency, and a second data path between an embedded processor and the mass storage array, wherein the data buffer without access latency is also used as a data buffer for non real-time data transfers between the embedded processor and the mass storage array.
The invention proposes a multi-purpose use of the 'no access latency' memory unit in the real-time data path. This is a more efficient use of the memory and additionally a more efficient way to transfer non real-time data between an embedded processor and a mass storage array consisting of single storage devices. Of course, the invention is not only applicable to systems designed for real-time data. It can likewise be used in other environments where a high guaranteed data rate is needed and latency is a problem.
The 'no access latency' memory, preferably with a small footprint, is integrated into the real-time data path as a realtime data buffer to compensate any latencies of the target mass storage array. The mass storage array consists of two or more single storage units. Each of the storage units is controlled by an I/O controller. Since also the embedded processor has to access and modify data on the mass storage array, there is a second path implemented between the real-time data buffer and the main memory of the processor. Therefore, a data switch is implemented for switching between data from or to the real-time interface and data from or to the embedded processor. Using the necessary real-time data buffer also for non real-time data transfers between the embedded processor and the mass storage array saves an extra buffer for non real-time transfers. Therefore, space and power consumption is reduced.
The hardware blocks are preferably configured by the embedded processor through a control line to be in the required mode. The embedded processor transfers data from and to the mass storage array by initiating a direct memory access transfer (DMA transfer) between its main memory and the real-time data buffer. This is an efficient implementation because the processor is able to work on other tasks while the data transfer is executed. In addition, access of the embedded processor to the mass storage array via the real-time data buffer is faster than accessing the mass storage array directly.
Advantageously, the mass storage system has a data processing block for splitting up a data stream and sending it in parallel to the two or more storage units. This allows to easily implement RAID or RAID-like systems.
Preferably, one or more thresholds can be set for the 'no access latency' data buffer, the one or more thresholds controlling the instant when data are streamed out of the data buffer to the mass storage array, to the real-time interface, and to the main memory. In this way the performance behavior of the different data targets can be adapted.
For a better understanding the invention shall now be explained in more detail in the following description with reference to the figures. It is understood that the invention is not limited to this exemplary embodiment and that specified features can also expediently be combined and/or modified without departing from the scope of the present invention. In the figures:
Fig. 1 shows an exemplary mass storage system according to the invention.
Fig. 1 shows the architectural design of a stand-alone mass storage system 1 with an embedded processor 2 for real-time data streams in form of a block diagram. A real-time interface block 3 is connected to a real-time data source or sink (not shown) , i.e. a camera or a display. The embedded processor 2 reacts on user input like 'record video' or 'playback video'. The embedded processor 2 configures the hardware blocks to be in the requested mode through control connections 4 (indicated by the dotted lines) . For this purpose, a simple 'mode' register is modified by the embedded processor 2 in a write access, which configures the hardware blocks.
For a real-time data streaming demand the embedded processor 2 first configures the hardware blocks. A data switch 5 selects the real-time interface for data streaming. In addition, a plurality of I/O control units 6 (I/O Ctrl) receive information about the direction (read or write) and the storage address of the real-time data stream. The latter is independent of the direction and the kind of data stream. The main difference between real-time data and non real-time data is the large amount of real-time data that is to be expected in comparison to the usually rather small amount of non real-time data. In order to cope with this difference the number of bytes to be transferred is included in the read/write command to the I/O control units 6. In the real-time mode the I/O control units 6 allow to read/write an amount of data that is sufficiently large to ensure that the embedded processor 2 is able to generate and send the next read/write command. This is necessary to guarantee the real-time data rate, as a relatively large amount of time passes between two read/write commands of the embedded processor 2. In the non real-time mode the amount of data allowed to be read/written with a single read/write command is rather small, as often only a few data have to be transferred.
In case of a 'record' user input the real-time data stream passes the real-time interface 3, the data switch 5, a real-time data buffer 7 and a data processing unit 8, where the real-time data stream is split up and sent in parallel to the I/O control units 6. The I/O control units 6 write the data to a plurality of storage units 9 of a mass storage array 10. As long as the storage units 9 of the mass storage array 10 are not able to receive data, the real-time data stream is stored into the real- time data buffer 7. The data transfer to the mass storage array 10 continues when the storage units 9 are ready to receive data. For example, such latencies arise if hard disks have to reposition their read/write heads. Due to the mechanical operations a data transfer is paused. Only after finishing the positioning procedure the hard disks are able to run at their media transfer data rate.
In case of a 'playback' user input the data are read in parallel from the storage units 9 and are written into the real-time data buffer 7 after passing the data processing unit 8, where the single data streams are merged into one real-time data stream. The transmission of the real-time data stream through the realtime interface 3 to a data sink is preferably initiated only when the real-time data buffer 7 is full. This measure helps to compensate for latencies produced by the mass storage array 10, and to fulfill the real-time requirements. This behavior is advantageously changeable by setting a threshold, which will be explained later.
If the embedded processor 2 has to access the mass storage array 10, e.g. to update some file system information or to read, write or modify descriptive data belonging to a real-time data stream stored on the mass storage array 10, a non real-time data transfer is executed. The main difference to a real-time data stream is that the real-time interface 3 is logically replaced by the processor's main memory 11. Data are transferred from the main memory 11 to the real-time data buffer 7 in case data are written to the mass storage array 10, or from the real-time data buffer 7 to the main memory 11 in case data are read from the storage array 10. Therefore, a direct memory access (DMA) transfer is initiated by the embedded processor 2 between the main memory 11 and the real-time data buffer 7. The data transfer from the real-time data buffer 7 to the mass storage array 10 is automatically executed. Consequently, after initializing the DMA transfer the embedded processor 2 is free to work on other tasks until a message of a successful or unsuccessful DMA transfer to the mass storage array 10 is received. During a DMA transfer, latencies of the mass storage array 10 are compensated for by the real-time data buffer 7 as if a real-time data stream was processed. This offers a higher performance than direct single word accesses of the embedded processor 2 to the mass storage array 10. A further improvement is that the control mechanism for a non real-time data stream is the same as for a real-time data stream. However, the data switch 6 has to be configured differently to replace the realtime interface 3 with the main memory 11. Also, the amount of data to be written with a single read/write command is modified. The amount of data per command is relatively small, but can likewise be adapted to the needs of the embedded processor 2, so that each DMA transfer is performed with a single command to the I/O control units 6.
A parallelization of the data is executed in the data processing unit 8, allowing to write to, respectively to read from, all single storage units 9 of the mass storage array 10 in parallel. This corresponds to a RAID-O system. Other RAID or RAID-like systems can be implemented as well.
Thresholds of the real-time data buffer 7 can be dynamically configured by the embedded processor 2, preferably even during two accesses to the mass storage array 10 or when different mass storage units 9 are added to the mass storage array 10. The thresholds control the instant when data are streamed out of the real-time data buffer 7 to the mass storage array 10, to the real-time interface 3, or to the main memory 11. In this way the performance behavior of the different data targets can be adapted. For example, a DMA transfer from the real-time data buffer 7 to the main memory 11 is first initiated when there are at least a number 'n' of bytes available in the real-time data buffer 7, the number 'n' being determined by a special threshold. Consequently, the main memory 11 is locked by the DMA transfer as short as possible and the embedded processor 2 has only to perform its tasks without its main memory 11 for the same short time. The dynamic modification of the thresholds allows to optimally adapt the system to different types of mass storage units 9 or to different types of data accesses, such as data streaming or transmission of smaller data blocks. For example, for some types of solid state disks a pause in the data transfer at an unfavorable instant leads to a drop in the data rate. During recording this is compensated by first collecting data in the real-time data buffer 7 before transferring them to the mass storage units 9. Though this reduces the latencies that can be compensated, the overall data rate is increased.

Claims

Claims
1. Mass storage system (1) having a first data path between a real-time data interface (3) and a mass storage array (10), the first data path including a data buffer (7) without access latency, and a second data path between an embedded processor (2) and the mass storage array (10), wherein the data buffer (7) without access latency is also used as a data buffer for non real-time data transfers between the embedded processor (2) and the mass storage array (10) .
2. Mass storage system (1) according to claim 1, further having a data switch (5) for switching between data from or to the real-time interface (3) and data from or to the embedded processor (2) .
3 Mass storage system (1) according to claim 1 or 2, wherein the mass storage array (10) includes two or more storage units (9) .
4. Mass storage system (1) according to claim 3, wherein the two or more storage units (9) are hard disks, optical disks, or solid state memories.
5. Mass storage system (1) according to claim 3 or 4, further having a data processing block (8) for splitting up a data stream and sending it in parallel to the two or more storage units (9) .
6. Mass storage system (1) according to one of claims 1 to 5, wherein a threshold can be set for the data buffer (7) without access latency, the threshold controlling the instant when data are streamed out of the data buffer (7) to the mass storage array (10), to the real-time interface (3) , or to the main memory (11) .
7. Mass storage system (1) according to claim 6, wherein different thresholds can be set for data transfers to the mass storage array (10), to the real-time interface (3), and to the main memory (11) .
8. Method for data transfer in a mass storage system (1) having a first data path between a real-time data interface
(3) and a mass storage array (10), the first data path including a data buffer (7) without access latency, and a second data path between an embedded processor (2) and the mass storage array (10), having the step of performing non real-time data transfers between the embedded processor (2) and the mass storage array (10) using the data buffer (7) without access latency.
PCT/EP2008/061148 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity WO2009027398A1 (en)

Priority Applications (5)

Application Number Priority Date Filing Date Title
US12/733,338 US8285932B2 (en) 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity
EP08787494A EP2183663B1 (en) 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity
CN2008801051342A CN101796479B (en) 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity
KR20107003596A KR101491687B1 (en) 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity
JP2010522347A JP5244909B2 (en) 2007-08-31 2008-08-26 Mass storage system with improved buffer capacity utilization

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP07115446.2 2007-08-31
EP07115446 2007-08-31

Publications (1)

Publication Number Publication Date
WO2009027398A1 true WO2009027398A1 (en) 2009-03-05

Family

ID=39944446

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2008/061148 WO2009027398A1 (en) 2007-08-31 2008-08-26 Mass storage system with improved usage of buffer capacity

Country Status (7)

Country Link
US (1) US8285932B2 (en)
EP (1) EP2183663B1 (en)
JP (1) JP5244909B2 (en)
KR (1) KR101491687B1 (en)
CN (1) CN101796479B (en)
TW (1) TWI461909B (en)
WO (1) WO2009027398A1 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI410870B (en) * 2010-03-26 2013-10-01 Novatek Microelectronics Corp Computer system architecture
KR102238652B1 (en) * 2014-11-12 2021-04-09 삼성전자주식회사 Data storage devce, method thereof, and method for operating data processing system having the same
CN104579575B (en) * 2015-01-23 2018-09-14 合肥联宝信息技术有限公司 A kind of method and device of embedded controller safety communication
CN107491266A (en) * 2016-06-13 2017-12-19 恩智浦美国有限公司 Mass-storage system and block date storage method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1046981A1 (en) * 1997-12-17 2000-10-25 Matsushita Electric Industrial Co., Ltd. Magnetic disk apparatus and method of data transfer
US6430660B1 (en) * 1999-05-21 2002-08-06 International Business Machines Corporation Unified memory hard disk drive system
US20060015682A1 (en) * 2004-07-15 2006-01-19 Hitachi Global Storage Technologies Netherlands B.V. Disk drive having real time performance improvement
EP1777614A1 (en) * 2005-10-17 2007-04-25 Hitachi, Ltd. Storage system

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996032674A2 (en) * 1995-04-13 1996-10-17 Cirrus Logic, Inc. Semiconductor memory device for mass storage block access applications
US20010047473A1 (en) * 2000-02-03 2001-11-29 Realtime Data, Llc Systems and methods for computer initialization
US7039770B1 (en) * 2002-03-05 2006-05-02 Juniper Networks, Inc. Low latency request dispatcher
US7350059B2 (en) * 2004-05-21 2008-03-25 Via Technologies, Inc. Managing stack transfers in a register-based processor
US7523327B2 (en) * 2005-03-05 2009-04-21 Intel Corporation System and method of coherent data transfer during processor idle states
WO2007023975A1 (en) * 2005-08-22 2007-03-01 Ssd Company Limited Multi-processor, direct memory access controller, and serial data transmitting/receiving apparatus
JP4671353B2 (en) * 2005-12-14 2011-04-13 株式会社日立製作所 Storage apparatus and control method thereof
US20080301376A1 (en) * 2007-05-31 2008-12-04 Allison Brian D Method, Apparatus, and System Supporting Improved DMA Writes
WO2009013877A1 (en) * 2007-07-20 2009-01-29 Panasonic Corporation Memory controller, memory card, and nonvolatile memory system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1046981A1 (en) * 1997-12-17 2000-10-25 Matsushita Electric Industrial Co., Ltd. Magnetic disk apparatus and method of data transfer
US6430660B1 (en) * 1999-05-21 2002-08-06 International Business Machines Corporation Unified memory hard disk drive system
US20060015682A1 (en) * 2004-07-15 2006-01-19 Hitachi Global Storage Technologies Netherlands B.V. Disk drive having real time performance improvement
EP1777614A1 (en) * 2005-10-17 2007-04-25 Hitachi, Ltd. Storage system

Also Published As

Publication number Publication date
JP5244909B2 (en) 2013-07-24
JP2010537339A (en) 2010-12-02
TW200910086A (en) 2009-03-01
KR20100054134A (en) 2010-05-24
CN101796479B (en) 2012-04-18
CN101796479A (en) 2010-08-04
US8285932B2 (en) 2012-10-09
EP2183663B1 (en) 2012-10-10
US20100211738A1 (en) 2010-08-19
EP2183663A1 (en) 2010-05-12
TWI461909B (en) 2014-11-21
KR101491687B1 (en) 2015-02-09

Similar Documents

Publication Publication Date Title
CA3062336C (en) Apparatus and method for controlling data acceleration
US8140747B2 (en) Operating method for a memory subsystem and devices for executing the operating method
KR102145401B1 (en) Apparatuses and methods for storing and writing multiple parameter codes for memory operating parameters
KR20090019593A (en) Ecc control circuit and multi channel memory system icluding the same
US20120278539A1 (en) Memory apparatus, memory control apparatus, and memory control method
JP2015136073A (en) Recorder, imaging apparatus, control method and program of recorder
US8285932B2 (en) Mass storage system with improved usage of buffer capacity
US7809877B2 (en) Host apparatus for controlling memory cards which minimizes interruption of writing to memory cards
US11150809B2 (en) Memory controller and storage device including the same
US11194493B2 (en) Data storage system capable of using high speed channel to access data, and using low speed channel to manage data exchanging, copying, and moving
US9990142B2 (en) Mass storage system and method of storing mass data
US20100125704A1 (en) Storage control apparatus and storage system
US20190187901A1 (en) Memory system and operating method thereof
WO2017090517A1 (en) Software storage unit, backup method, and recording medium on which backup control program is recorded
JP4622871B2 (en) Data processing system, access control method, apparatus thereof, and program thereof
KR20150083558A (en) Method for processing data on storage device and storage device
CN102063269B (en) Method and system for realizing communication of double controllers of disk array
KR20050004157A (en) Data transfer unit with support for multiple coherency granules
US20070174738A1 (en) Disk device, method of writing data in disk device, and computer product
JPH076093A (en) Storage controller
JPH08179894A (en) Memory system, and memory system applied to disk recording and reproducing device
JP2009017201A (en) Data recording device
KR20080025231A (en) Device having shared memory and method for displaying data
JP2007026125A (en) Recording and reproducing apparatus
JP2004199720A (en) Disk array subsystem

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200880105134.2

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08787494

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2008787494

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20107003596

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 12733338

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2010522347

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE