公开(公告)号：US07793167B2

公开(公告)日：2010-09-07

申请号：US11843791

申请日：2007-08-23

申请人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Conner ,  Krishnakumar R. Surugucchi

发明人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Conner ,  Krishnakumar R. Surugucchi

IPC分类号： G06F11/00

CPC分类号： G06F11/1076 ,  G06F11/004 ,  G06F2211/104 ,  G06F2211/109 ,  G11B20/1833 ,  G11B2020/1843 ,  G11B2220/2516 ,  G11B2220/415

摘要： Methods are provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a “new data, unconfirmed” state. The state of written data is periodically checked, such as with a timer. If the data is in the “new data, unconfirmed” state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a “single dropped write confirmed” state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a “confirmed good” state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an “unquantifiable” state.

摘要翻译： 提供了用于检测和校正在存储系统中的丢弃写入的方法。 将数据和校验和写入存储设备，如RAID阵列。 数据的状态被分类为“新数据，未确认”状态。 周期性地检查写入数据的状态，例如使用定时器。 如果数据处于“新数据，未确认”状态，则检查是否丢弃写入。 如果发生了丢弃的写入，则将数据的状态改变为“单个丢弃的写入确认”状态，并且优选地校正丢弃的写入错误。 如果没有检测到丢弃的写入，则状态被改变为“已确认的”状态。 如果在检查丢失的写入事件之前通过读取 - 修改 - 写入更新数据，则其状态将更改为“无法量化”状态。

公开(公告)号：US20080126850A1

公开(公告)日：2008-05-29

申请号：US11467869

申请日：2006-08-28

申请人： Robert A. Kubo ,  Richard A. Ripberger

发明人： Robert A. Kubo ,  Richard A. Ripberger

IPC分类号： G06F11/14

CPC分类号： G06F11/1092 ,  G06F11/1096 ,  G06F11/2094

摘要： A storage array disposed in a data storage system is reconfigured. A data storage system is supplied comprising a plurality of data storage devices, wherein each of the plurality of data storage devices is assigned to one of a plurality of data storage arrays, or is assigned as a spare device. An unassigned data storage device is detected. If any of the plurality of data storage arrays comprises an intermixed data storage array, the storage array is reconfigured to restore the intermixed data storage array to a homogeneous technology class comprising the unassigned device. If any of the plurality of data storage arrays comprises a degraded data storage array, the storage array is reconfigured to restore the degraded storage array to optimal reliability, optimal performance, and optimal efficiency.

摘要翻译： 配置在数据存储系统中的存储阵列被重新配置。 提供一种数据存储系统，包括多个数据存储装置，其中多个数据存储装置中的每一个被分配给多个数据存储阵列中的一个，或被分配为备用装置。 检测到未分配的数据存储设备。 如果多个数据存储阵列中的任一个包括混合数据存储阵列，则存储阵列被重新配置以将混合数据存储阵列恢复到包括未分配设备的均匀技术类别。 如果多个数据存储阵列中的任一个包括劣化的数据存储阵列，则重新配置存储阵列以将劣化的存储阵列恢复到最佳可靠性，最佳性能和最佳效率。

公开(公告)号：US07996609B2

公开(公告)日：2011-08-09

申请号：US11613636

申请日：2006-12-20

申请人： Kevin J. Ash ,  Robert A. Kubo

发明人： Kevin J. Ash ,  Robert A. Kubo

IPC分类号： G06F12/00 ,  G06F13/00 ,  G06F13/28

CPC分类号： G06F12/0871 ,  G06F12/0804 ,  G06F2212/222 ,  G06F2212/262

摘要： A method of dynamic allocation of non-volatile memory in a cache storage subsystem includes constraining a redundant array of independent disks (RAID) 6 array in an unavailable state, the RAID 6 array having been newly created in the subsystem, destaging data from a write cache to a storage medium to free memory space, reserving the freed memory space for allocation to data-persistent storage, allocating the memory space to an adapter in the subsystem as a target for data in the event of a power loss, and changing the RAID 6 array to an available state. A system for dynamic allocation of non-volatile memory in a cache storage subsystem includes a computer processing device which constrains a newly-created redundant array of independent disks (RAID) 6 array in an unavailable state to execute a reconfiguration method.

摘要翻译： 在高速缓存存储子系统中动态分配非易失性存储器的方法包括限制在不可用状态下的独立磁盘（RAID）6阵列的冗余阵列，已经在子系统中新创建的RAID 6阵列，从写入 缓存到存储介质以释放存储器空间，保留释放的存储器空间用于分配到数据永久存储器，将存储器空间分配给子系统中的适配器作为电源丢失的数据的目标，并且改变RAID 6阵列到可用状态。 用于在高速缓存存储子系统中动态分配非易失性存储器的系统包括限制在不可用状态下新创建的独立磁盘冗余阵列（RAID）6阵列以执行重新配置方法的计算机处理设备。

公开(公告)号：US07451354B2

公开(公告)日：2008-11-11

申请号：US11326650

申请日：2006-01-06

申请人： Matthew J. Kalos ,  Robert A. Kubo ,  Michael P. Vageline

发明人： Matthew J. Kalos ,  Robert A. Kubo ,  Michael P. Vageline

IPC分类号： G06F11/00

CPC分类号： G06F3/0629 ,  G06F3/0607 ,  G06F3/0683

摘要： An apparatus and method are disclosed to configure, format, and test, a data storage subsystem product. The method supplies a data storage subsystem product comprising one or more host computer ports, a processor, one or more data storage device ports, and one or more data storage devices interconnected to the one or more data storage device ports. The method further supplies a configuration appliance comprising a storage configuration. The method connects the configuration appliance to one of the one or more storage device ports, boots up the data storage subsystem product, discovers the configuration appliance by the data storage subsystem product, imports storage configuration data into the data storage subsystem product, formats the one or more data storage device, and tests the input and output data transfer rates for the data storage subsystem product, wherein the formatting and testing are initiated concurrently.

摘要翻译： 公开了一种用于配置，格式化和测试数据存储子系统产品的装置和方法。 该方法提供包括一个或多个主计算机端口，处理器，一个或多个数据存储设备端口以及与一个或多个数据存储设备端口互连的一个或多个数据存储设备的数据存储子系统产品。 该方法还提供包括存储配置的配置设备。 该方法将配置设备连接到一个或多个存储设备端口之一，启动数据存储子系统产品，通过数据存储子系统产品发现配置设备，将存储配置数据导入数据存储子系统产品，格式化 或更多数据存储设备，并测试数据存储子系统产品的输入和输出数据传输速率，其中格式化和测试同时启动。

公开(公告)号：US20080195807A1

公开(公告)日：2008-08-14

申请号：US11673367

申请日：2007-02-09

申请人： Robert A. Kubo ,  Jeremy M. Pinson

发明人： Robert A. Kubo ,  Jeremy M. Pinson

IPC分类号： G06F12/16 ,  G06F12/00

CPC分类号： G06F12/126

摘要： A method of destage management of redundant data copies in a cache storage subsystem includes providing a plurality of target storage devices, each of the plurality of target storage devices capable of storing a complete copy of a data image. The data image and the plurality of target storage devices are subdivided into a plurality of sub-regions, each of the plurality of sub-regions organized such that a minimum cumulative total of the plurality of sub-regions on the target storage devices equals a number of sub-regions required to form the complete copy of the data image. Data is destaged from a write cache to the plurality of sub-regions on the target storage devices in an order such that a minimum number of sub-regions can be combined to restore the complete copy of the data image.

摘要翻译： 一种对高速缓存存储子系统中的冗余数据副本的管理的方法，包括：提供多个目标存储设备，所述多个目标存储设备中的每一个能够存储数据图像的完整副本。 数据图像和多个目标存储设备被细分为多个子区域，多个子区域中的每一个被组织成使得目标存储设备上的多个子区域的最小累积总数等于数字 的子区域，以形成数据图像的完整副本。 从目标存储设备上的多个子区域的写高速缓冲存储器中的数据按照这样一个顺序被排除，使得可以组合最小数量的子区域以恢复数据图像的完整拷贝。

公开(公告)号：US20100251072A1

公开(公告)日：2010-09-30

申请号：US12813491

申请日：2010-06-10

申请人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Connor ,  Krishnakumar R Surugucchi

发明人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Connor ,  Krishnakumar R Surugucchi

IPC分类号： H03M13/05 ,  G06F11/10

CPC分类号： G06F11/1076 ,  G06F2211/1057

摘要： A RAID system is provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a “new data, unconfirmed” state. The state of written data is periodically checked, such as with a timer. If the data is in the “new data, unconfirmed” state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a “single dropped write confirmed” state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a “confirmed good” state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an “unquantifiable” state.

摘要翻译： 提供了一种RAID系统，用于检测和纠正存储系统中丢弃的写入。 将数据和校验和写入存储设备，如RAID阵列。 数据的状态被分类为“新数据，未确认”状态。 周期性地检查写入数据的状态，例如使用定时器。 如果数据处于“新数据，未确认”状态，则检查是否丢弃写入。 如果发生了丢弃的写入，则将数据的状态改变为“单个丢弃的写入确认”状态，并且优选地校正丢弃的写入错误。 如果没有检测到丢弃的写入，则状态被改变为“已确认的”状态。 如果在检查丢失的写入事件之前通过读取 - 修改 - 写入更新数据，则其状态将更改为“无法量化”状态。

公开(公告)号：US07793168B2

公开(公告)日：2010-09-07

申请号：US11843804

申请日：2007-08-23

申请人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Conner ,  Krishnakumar S. Rao

发明人： James L. Hafner ,  Carl E. Jones ,  David R. Kahler ,  Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen ,  James A. O'Conner ,  Krishnakumar S. Rao

IPC分类号： G06F11/00

CPC分类号： G06F11/1076 ,  G06F2211/1057

摘要： Method, system and computer program product are provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a “new data, unconfirmed” state. The state of written data is periodically checked, such as with a timer. If the data is in the “new data, unconfirmed” state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a “single dropped write confirmed” state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a “confirmed good” state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an “unquantifiable” state.

摘要翻译： 提供了方法，系统和计算机程序产品，用于检测和纠正存储系统中丢弃的写入。 将数据和校验和写入存储设备，如RAID阵列。 数据的状态被分类为“新数据，未确认”状态。 周期性地检查写入数据的状态，例如使用定时器。 如果数据处于“新数据，未确认”状态，则检查是否丢弃写入。 如果发生了丢弃的写入，则将数据的状态改变为“单个丢弃的写入确认”状态，并且优选地校正丢弃的写入错误。 如果没有检测到丢弃的写入，则状态被改变为“已确认的”状态。 如果在检查丢失的写入事件之前通过读取 - 修改 - 写入更新数据，则其状态将更改为“无法量化”状态。

公开(公告)号：US20090222621A1

公开(公告)日：2009-09-03

申请号：US12039911

申请日：2008-02-29

申请人： Kevin J. Ash ,  Robert A. Kubo ,  Alfred E. Sanchez

发明人： Kevin J. Ash ,  Robert A. Kubo ,  Alfred E. Sanchez

IPC分类号： G06F12/00

CPC分类号： G06F12/0866 ,  G06F3/0613 ,  G06F3/0656 ,  G06F3/0659 ,  G06F3/0689 ,  G06F9/5016

摘要： Systems and methods for allocating task control blocks in an information storage and retrieval system are disclosed. Task control blocks for new writes and reads are allocated by the cache code after a determination of the number of task control blocks already allocated for other tasks.

摘要翻译： 公开了在信息存储和检索系统中分配任务控制块的系统和方法。 在确定已经分配给其他任务的任务控制块的数量之后，新的写入和读取的任务控制块由高速缓存代码分配。

公开(公告)号：US07496796B2

公开(公告)日：2009-02-24

申请号：US11337391

申请日：2006-01-23

申请人： Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen

发明人： Robert A. Kubo ,  David F. Mannenbach ,  Karl A. Nielsen

IPC分类号： G06F11/00

CPC分类号： G06F11/008 ,  G06F11/3485 ,  G06F2201/88

摘要： An apparatus, system, and method are disclosed for predicting storage device failure. A technology descriptor module associates a technology descriptor with a storage device. A failure threshold module sets a predictive failure threshold for the storage device in response to the technology descriptor. In one embodiment, a workload management detection module detects workload management of the storage device. A threshold modification module may modify the predictive failure threshold in response to detecting the workload management of the storage device. A performance detection module detects a storage device error that exceeds the modified predictive failure threshold as a storage device predictive failure.

摘要翻译： 公开了用于预测存储设备故障的装置，系统和方法。 技术描述符模块将技术描述符与存储设备相关联。 故障阈值模块响应于技术描述符设置存储设备的预测故障阈值。 在一个实施例中，工作负载管理检测模块检测存储设备的工作负载管理。 响应于检测到存储设备的工作负载管理，阈值修改模块可以修改预测故障阈值。 性能检测模块检测超过修改的预测故障阈值的存储设备错误作为存储设备预测故障。

公开(公告)号：US20080155190A1

公开(公告)日：2008-06-26

申请号：US11613636

申请日：2006-12-20

申请人： Kevin J. Ash ,  Robert A. Kubo

发明人： Kevin J. Ash ,  Robert A. Kubo

IPC分类号： G06F12/02

CPC分类号： G06F12/0871 ,  G06F12/0804 ,  G06F2212/222 ,  G06F2212/262

摘要： A method of dynamic allocation of non-volatile memory in a cache storage subsystem includes constraining a redundant array of independent disks (RAID) 6 array in an unavailable state, the RAID 6 array having been newly created in the subsystem, destaging data from a write cache to a storage medium to free memory space, reserving the freed memory space for allocation to data-persistent storage, allocating the memory space to an adapter in the subsystem as a target for data in the event of a power loss, and changing the RAID 6 array to an available state. A system for dynamic allocation of non-volatile memory in a cache storage subsystem includes a computer processing device which constrains a newly-created redundant array of independent disks (RAID) 6 array in an unavailable state to execute a reconfiguration method.

摘要翻译： 在高速缓存存储子系统中动态分配非易失性存储器的方法包括限制在不可用状态下的独立磁盘（RAID）6阵列的冗余阵列，已经在子系统中新创建的RAID 6阵列，从写入 缓存到存储介质以释放存储器空间，保留释放的存储器空间用于分配到数据永久存储器，将存储器空间分配给子系统中的适配器作为电源丢失的数据的目标，并且改变RAID 6阵列到可用状态。 用于在高速缓存存储子系统中动态分配非易失性存储器的系统包括限制在不可用状态下新创建的独立磁盘冗余阵列（RAID）6阵列以执行重新配置方法的计算机处理设备。