公开(公告)号：US06732292B2

公开(公告)日：2004-05-04

申请号：US10090876

申请日：2002-03-04

申请人： Mark D. Hertz ,  Philip E. Cloud ,  Travis D. Fox ,  Edwin S. Olds

发明人： Mark D. Hertz ,  Philip E. Cloud ,  Travis D. Fox ,  Edwin S. Olds

IPC分类号： G06F1120

CPC分类号： G06F3/0659 ,  G06F3/0613 ,  G06F3/064 ,  G06F3/0676

摘要： Method and apparatus for transferring data between a host device and a data storage device having a first memory space and a second memory space. The host issues access commands to store and retrieve data. The device stores write data associated with write commands in the first memory space pending transfer to the second memory space. An interface circuit evaluates relative proximity of first and second sets of LBAs associated with pending first and second write commands, and forms an adaptive bi-directional write skip mask when the sets of LBAs fit within a predetermined mask interval. The mask is preferably slidably adjustable to encompass both sets of LBAs and is preferably expandable to accommodate additional sets of LBAs. The data within the mask are treated as a combined write command and are thereafter written in an order corresponding to an order in which the data appear in the mask.

摘要翻译： 用于在主机设备和具有第一存储器空间和第二存储器空间的数据存储设备之间传送数据的方法和装置。 主机发出访问命令来存储和检索数据。 该设备将与写命令相关联的写数据存储在第一存储器空间中，等待传送到第二存储器空间。 接口电路评估与待决的第一和第二写命令相关联的第一组和第二组LBA的相对接近度，并且当所述LBA组合在预定的屏蔽间隔内时，形成自适应双向写入跳过掩码。 掩模优选地可滑动地调节以包含两组LBA，并且优选地可扩展以容纳另外的一组LBA。 掩码内的数据被视为组合写入命令，然后以与数据出现在掩码中的顺序相对应的顺序写入。

公开(公告)号：US06934802B2

公开(公告)日：2005-08-23

申请号：US10345476

申请日：2003-01-15

申请人： Stephen R. Cornaby ,  Travis D. Fox ,  Edwin S. Olds

发明人： Stephen R. Cornaby ,  Travis D. Fox ,  Edwin S. Olds

IPC分类号： G06F12/08 ,  G06F12/12

CPC分类号： G06F12/0862 ,  G06F12/0866 ,  G06F12/122 ,  G06F12/123 ,  G06F2212/312

摘要： A data storage device with a cache memory in communication with a control processor programmed with a routine to effect data throughput with a host device. The data storage device includes a head-disc assembly responsive to the control processor retrieving a host data along with a speculative data in a form of a read data. The control processor adjusting a range for acquisition of the speculative data based on data stored in cache memory fragments of the cache memory. The cache memory storing the read data in an unused cache memory fragment, while the control processor updates a history queue and a band count table based on the acquired read data and releases an oldest cache memory fragment with a lowest count value in the band count table from the cache memory in response to a need for cache memory space.

摘要翻译： 一种数据存储设备，其具有与用例程一起编程的控制处理器通信的高速缓冲存储器，以利用主机设备来实现数据吞吐量。 数据存储设备包括头盘组件，其响应于控制处理器以读取数据的形式检索主机数据以及推测数据。 控制处理器基于存储在高速缓冲存储器的高速缓存存储器片段中的数据来调整用于获取推测数据的范围。 高速缓存存储器将读取的数据存储在未使用的高速缓冲存储器片段中，而控制处理器基于所获取的读取数据更新历史队列和频带计数表，并且在频带计数表中释放具有最低计数值的最旧的高速缓存存储器片段 从高速缓冲存储器响应需要缓存的内存空间。

公开(公告)号：US06957300B2

公开(公告)日：2005-10-18

申请号：US10143235

申请日：2002-05-10

申请人： Edwin S. Olds ,  Jack A. Mobley ,  Mark D. Hertz ,  Kenny T. Coker

发明人： Edwin S. Olds ,  Jack A. Mobley ,  Mark D. Hertz ,  Kenny T. Coker

IPC分类号： G06F3/06 ,  G06F12/00 ,  G06F12/08

CPC分类号： G06F3/0601 ,  G06F2003/0697

摘要： Method and apparatus for transferring data between a host device and a data storage device having a first memory space and a second memory space. The host issues access commands to store and retrieve data. The device stores commands in the first memory space pending transfer to the second memory space. An interface circuit evaluates relative proximity of first and second sets of LBAs associated with pending first and second commands, and delays promotion of later pending commands in front of earlier pending commands during an overlap condition. If the overlap is caused by performance enhancing features (PEF) the PEFs are disabled so the commands can be scheduled for disc access. Indicators are set in the commands to signal that a PEF has caused the overlap and that PEF can be disabled. Values are added to indicators in the commands such that the PEFs can be modified and avoid overlaps.

摘要翻译： 用于在主机设备和具有第一存储器空间和第二存储器空间的数据存储设备之间传送数据的方法和装置。 主机发出访问命令来存储和检索数据。 设备将命令存储在第一存储器空间中，等待传送到第二存储器空间。 接口电路评估与等待的第一和第二命令相关联的第一组和第二组LBA的相对接近度，并且在重叠条件期间延迟提前稍早的待命命令。 如果重叠是由性能增强功能（PEF）引起的，则PEF被禁用，因此命令可以被调度用于光盘访问。 在命令中设置指示信号，指示PEF已经引起重叠，并且可以禁用PEF。 将值添加到命令中的指示符中，以便可以修改PEF并避免重叠。

公开(公告)号：USRE44128E1

公开(公告)日：2013-04-02

申请号：US12642157

申请日：2009-12-18

申请人： Edwin S. Olds ,  Travis D. Fox ,  Mark A. Thiessen

发明人： Edwin S. Olds ,  Travis D. Fox ,  Mark A. Thiessen

IPC分类号： G06F13/00

CPC分类号： G11B20/10527 ,  G06F12/0804 ,  G06F12/0866 ,  G06F12/123 ,  G11B2020/10675

摘要： A method for determining an aging period for retaining a write-back data in a cache memory prior to writing the write-back data to a storage media is determined through use of a write-back aging routine. The aging period is based on a proportional utilization level of the cache memory by the write-back data, the higher the memory utilization level, the shorter the period for aging the write-back data. The aging period takes a form of an aging threshold, which differs depending on the memory utilization level, i.e., the amount of cache memory utilized by the write-back data.The method includes, identifying the memory utilization level, selecting the data aging threshold based on the memory utilization level; and writing the data from the cache memory to the storage media when an age of the data in the memory exceeds the selected data aging threshold.

摘要翻译： 通过使用回写老化例程来确定在将写回数据写入存储介质之前，将缓存数据保留在高速缓冲存储器中的老化周期的方法。 老化周期基于回写数据的高速缓冲存储器的比例利用率，存储器利用率水平越高，回写数据的老化周期越短。 老化周期采用老化阈值的形式，其根据存储器利用率，即由回写数据使用的高速缓冲存储器的量而不同。 该方法包括：识别内存利用水平，根据内存利用水平选择数据老化阈值; 以及当所述存储器中的数据的年龄超过所选择的数据老化阈值时，将所述数据从所述高速缓冲存储器写入所述存储介质。

公开(公告)号：US07464218B2

公开(公告)日：2008-12-09

申请号：US10930474

申请日：2004-08-31

申请人： Mark A. Thiessen ,  Edwin S. Olds ,  Travis D. Fox

发明人： Mark A. Thiessen ,  Edwin S. Olds ,  Travis D. Fox

IPC分类号： G06F13/14

CPC分类号： G11B20/10 ,  G06F3/0601 ,  G06F2003/0692

摘要： A method for improving throughput performance of a data storage device by executing an execution critical write-back data priority routine programmed into a controller of the data storage device. The method includes, determining a write-back data aging threshold limit; identifying and executing a pending command; recognizing write-back data exceeding the aging threshold limit as an ending position aged data execution command; determining a read/write head following an execution of a pending command; determining a latency portion of the aged data execution command based on the determined head position; ascertaining a total execution time for each of a plurality of commands based on the determined head position; selecting from the plurality of commands an additional command executable within the determined latency portion; and executing both the additional command and the aged data execution command following execution of the pending command, when the total execution time of the aged data execution command is unaffected.

摘要翻译： 一种通过执行编程到数据存储设备的控制器中的执行关键回写数据优先级程序来提高数据存储设备的吞吐量性能的方法。 该方法包括：确定回写数据老化阈值限制; 识别和执行待命命令; 识别超过老化阈值的回写数据作为数据执行命令的结束位置; 在执行待命命令之后确定读/写头; 基于确定的头位置确定老化数据执行命令的延迟部分; 基于所确定的头位置确定多个命令中的每一个的总执行时间; 从所述多个命令中选择在所确定的等待时间部分内可执行的附加命令; 并且当老化的数据执行命令的总执行时间不受影响时，在执行挂起命令之后执行附加命令和老化的数据执行命令。

公开(公告)号：US07346740B2

公开(公告)日：2008-03-18

申请号：US10602254

申请日：2003-06-23

申请人： Travis D. Fox ,  Edwin S. Olds ,  Mark A. Gaertner ,  Abbas Ali

发明人： Travis D. Fox ,  Edwin S. Olds ,  Mark A. Gaertner ,  Abbas Ali

IPC分类号： G06F13/00

CPC分类号： G06F3/0611 ,  G06F3/0656 ,  G06F3/0659 ,  G06F3/0676 ,  G06F12/0862 ,  G06F12/0866

摘要： Method and apparatus for transferring speculative data in lieu of requested data in a data transfer operation. First data are transferred in response to an execution of a first pending command. Speculative data are transferred instead of second data associated with a second pending command during a next available latency period for the second data, preferably when the speculative data are adjudged as having a utility greater than a utility of the second data. The first and second commands are preferably received in a queue and a command execution algorithm identifies the second command as a next best command to be executed after execution of the first command. The above steps are preferably carried out by a controller of a data storage device.

摘要翻译： 用于在数据传送操作中传送推测数据代替所请求的数据的方法和装置。 响应于第一未决命令的执行而传送第一数据。 在第二数据的下一可用等待时间期间优选地，当推测数据被判定为具有大于第二数据的效用的实用程序时，传送投机数据而不是与第二挂起命令相关联的第二数据。 第一和第二命令优选地被接收在队列中，并且命令执行算法将第二命令标识为在执行第一命令之后执行的下一个最佳命令。 上述步骤优选地由数据存储装置的控制器执行。

公开(公告)号：US07310707B2

公开(公告)日：2007-12-18

申请号：US10439266

申请日：2003-05-15

申请人： Edwin S. Olds ,  Travis D. Fox ,  Mark A Thiessen

发明人： Edwin S. Olds ,  Travis D. Fox ,  Mark A Thiessen

IPC分类号： G06F12/12

CPC分类号： G11B20/10527 ,  G06F12/0804 ,  G06F12/0866 ,  G06F12/123 ,  G11B2020/10675

摘要： A method for determining an aging period for retaining a write-back data in a cache memory prior to writing the write-back data to a storage media is determined through use of a write-back aging routine. The aging period is based on a proportional utilization level of the cache memory by the write-back data, the higher the memory utilization level, the shorter the period for aging the write-back data. The aging period takes a form of an aging threshold, which differs depending on the memory utilization level, i.e., the amount of cache memory utilized by the write-back data.The method includes, identifying the memory utilization level, selecting the data aging threshold based on the memory utilization level; and writing the data from the cache memory to the storage media when an age of the data in the memory exceeds the selected data aging threshold.

摘要翻译： 通过使用回写老化例程来确定在将写回数据写入存储介质之前，将缓存数据保留在高速缓冲存储器中的老化周期的方法。 老化周期基于回写数据的高速缓冲存储器的比例利用率，存储器利用率水平越高，回写数据的老化周期越短。 老化周期采用老化阈值的形式，其根据存储器利用率，即由回写数据使用的高速缓冲存储器的量而不同。 该方法包括：识别内存利用水平，根据内存利用水平选择数据老化阈值; 以及当所述存储器中的数据的年龄超过所选择的数据老化阈值时，将所述数据从所述高速缓冲存储器写入所述存储介质。

公开(公告)号：US08595432B1

公开(公告)日：2013-11-26

申请号：US13237683

申请日：2011-09-20

申请人： Wayne H. Vinson ,  Edwin S. Olds

发明人： Wayne H. Vinson ,  Edwin S. Olds

IPC分类号： G06F12/00

CPC分类号： G06F3/0689 ,  G06F3/0611 ,  G06F3/0619 ,  G06F3/0659 ,  G06F3/0665

摘要： A disk drive is disclosed comprising a head actuated over a disk. The disk drive receives a plurality of access commands including a first access command out of a group G of access commands. The access commands are stored in a command queue. A completion status is received for a number of access commands out of the G access commands that are assigned to other disk drives, and the first access command is selected from the command queue based on the completion status.

摘要翻译： 公开了一种磁盘驱动器，包括在盘上致动的磁头。 磁盘驱动器从接入命令组G接收包括第一访问命令的多个访问命令。 访问命令存储在命令队列中。 从分配给其他磁盘驱动器的G访问命令中接收到多个访问命令的完成状态，并且基于完成状态从命令队列中选择第一访问命令。

公开(公告)号：US20090055620A1

公开(公告)日：2009-02-26

申请号：US11842653

申请日：2007-08-21

申请人： Timothy R. Feldman ,  Jonathan W. Haines ,  Craig W. Miller ,  Edwin S. Olds

发明人： Timothy R. Feldman ,  Jonathan W. Haines ,  Craig W. Miller ,  Edwin S. Olds

IPC分类号： G06F12/00

CPC分类号： G06F12/0246 ,  B82Y25/00 ,  G01R33/093 ,  G06F2212/1032 ,  G06F2212/7201 ,  G11B5/012

摘要： The application relates to defect management using mutable logical to physical association. Embodiments disclosed utilize mutable mapping between logical blocks and physical blocks. Dynamically updated mapping data, which mutably associates the logical blocks and the physical blocks, also includes physical block defect allocations.

摘要翻译： 该应用涉及使用可变逻辑与物理关联的缺陷管理。 公开的实施例利用逻辑块和物理块之间的可变映射。 动态更新的映射数据，其可变地关联逻辑块和物理块，还包括物理块缺陷分配。

公开(公告)号：US08699175B1

公开(公告)日：2014-04-15

申请号：US13554994

申请日：2012-07-20

申请人： Edwin S. Olds ,  William B. Boyle

发明人： Edwin S. Olds ,  William B. Boyle

IPC分类号： G11B5/596

CPC分类号： G11B5/012 ,  G11B20/1217

摘要： A drive is disclosed comprising a head actuated over a disk comprising a plurality of data tracks. A random access write zone is defined on the disk comprising a first plurality of the data tracks, and a circular buffer write zone is defined on the disk comprising a second plurality of the data tracks. Write commands are received comprising logical block addresses (LBAs). First frequency and second frequency write LBAs are identified, where the second frequency is higher than the first frequency. Data associated with the first frequently write LBAs is stored in the circular buffer write zone, and data associated with the second frequency write LBAs is stored in the random access write zone.

摘要翻译： 公开了一种驱动器，包括在包括多个数据轨道的盘上致动的头部。 在包括第一多个数据轨道的磁盘上定义随机访问写入区，并且在包括第二多个数据轨道的盘上定义循环缓冲区写入区。 接收包括逻辑块地址（LBA）的写命令。 识别第一频率和第二频率写入LBA，其中第二频率高于第一频率。 与第一频繁写入LBA相关联的数据被存储在循环缓冲区写入区中，并且与第二频率写入LBA相关联的数据被存储在随机存取写入区中。