公开(公告)号：US20120079492A1

公开(公告)日：2012-03-29

申请号：US12889741

申请日：2010-09-24

申请人： David D. Chambliss ,  Prashant Pandey

发明人： David D. Chambliss ,  Prashant Pandey

IPC分类号： G06F9/50

CPC分类号： G06F9/5083

摘要： Embodiments are provided for managing the system performance of resources performing tasks in response to task requests from tenants. In one aspect, a system that comprises at least one resource configured to perform at least one admitted task with an impact under the control of a computer system. The computer system provides services to more than one tenant. The computer system comprises a strategist configured to assess the impact of the admitted task to create a cost function vector containing multiple cost function specifications and a budget policy vector containing multiple budget policies and an actuator. The actuator receives the cost function vector and the budget policy vector from the strategist, receives a task request one of the more than one tenants, and calculates cost functions based upon the cost function vector to predict the impact of the task request on the resources for each of the task requests. The actuator throttles the task requests based upon the budget policies for the impact on the resources to create at least one of the admitted task performed by the resource and a delayed task request.

摘要翻译： 提供实施例以响应于来自租户的任务请求来管理执行任务的资源的系统性能。 在一个方面，一种系统，其包括被配置为在计算机系统的控制下执行具有冲击的至少一个允许的任务的至少一个资源。 计算机系统为多个租户提供服务。 计算机系统包括策略师，其被配置为评估所允许的任务的影响以创建包含多个成本函数规范的成本函数向量和包含多个预算策略的预算策略向量和执行器。 执行器从策略者接收成本函数向量和预算策略向量，接收多个租户中的一个租户的任务请求，并根据成本函数向量计算成本函数，以预测任务请求对资源的影响 每个任务请求。 执行器基于预算政策来阻碍任务请求，以对资源产生影响，以创建资源执行的至少一个被允许的任务和延迟的任务请求。

公开(公告)号：US08473960B2

公开(公告)日：2013-06-25

申请号：US12889741

申请日：2010-09-24

申请人： David D. Chambliss ,  Prashant Pandey

发明人： David D. Chambliss ,  Prashant Pandey

IPC分类号： G06F9/46 ,  G06F15/173 ,  G06F17/00

CPC分类号： G06F9/5083

摘要： Embodiments are provided for managing the system performance of resources performing tasks in response to task requests from tenants. In one aspect, a system that comprises at least one resource configured to perform at least one admitted task with an impact under the control of a computer system. The computer system provides services to more than one tenant. The computer system comprises a strategist configured to assess the impact of the admitted task to create a cost function vector containing multiple cost function specifications and a budget policy vector containing multiple budget policies and an actuator. The actuator receives the cost function vector and the budget policy vector from the strategist, receives a task request one of the more than one tenants, and calculates cost functions based upon the cost function vector to predict the impact of the task request on the resources for each of the task requests. The actuator throttles the task requests based upon the budget policies for the impact on the resources to create at least one of the admitted task performed by the resource and a delayed task request.

摘要翻译： 提供实施例以响应于来自租户的任务请求来管理执行任务的资源的系统性能。 在一个方面，一种系统，其包括被配置为在计算机系统的控制下执行具有冲击的至少一个允许的任务的至少一个资源。 计算机系统为多个租户提供服务。 计算机系统包括策略师，其被配置为评估所允许的任务的影响以创建包含多个成本函数规范的成本函数向量和包含多个预算策略的预算策略向量和执行器。 执行器从策略者接收成本函数向量和预算策略向量，接收多个租户中的一个租户的任务请求，并根据成本函数向量计算成本函数，以预测任务请求对资源的影响 每个任务请求。 执行器基于预算政策来阻碍任务请求，以对资源产生影响，以创建资源执行的至少一个被允许的任务和延迟的任务请求。

公开(公告)号：US08713572B2

公开(公告)日：2014-04-29

申请号：US13233162

申请日：2011-09-15

申请人： David D. Chambliss ,  Joseph S. Hyde, II ,  Prashant Pandey ,  William G. Sherman ,  Rui Zhang

发明人： David D. Chambliss ,  Joseph S. Hyde, II ,  Prashant Pandey ,  William G. Sherman ,  Rui Zhang

IPC分类号： G06F9/46

CPC分类号： G06F9/4881

摘要： Methods, systems, and physical computer-readable storage medium for processing a plurality of IO request jobs are provided. The method includes determining whether one or more request jobs are not meeting a QoS target, each job of the one or more request jobs having a corresponding priority, selecting a highest priority job from the one or more request jobs, if one or more request jobs are not meeting the QoS target, determining whether the highest priority job has a corresponding effective rate limit imposed thereon, if so, relaxing the corresponding effective rate limit, and if not, selecting one or more lower priority jobs from the one or more request jobs and tightening a corresponding effective limit on the one or more lower priority jobs from the one or more request jobs in accordance with a delay factor limit.

摘要翻译： 提供了用于处理多个IO请求作业的方法，系统和物理计算机可读存储介质。 该方法包括确定一个或多个请求作业是否不满足QoS目标，一个或多个请求作业的每个作业具有相应的优先级，从一个或多个请求作业中选择最高优先级的作业，如果一个或多个请求作业 不满足QoS目标，确定最高优先级作业是否具有对其施加的对应的有效速率限制，如果是，放宽对应的有效速率限制，如果不是，则从一个或多个请求作业中选择一个或多个较低优先级的作业 以及根据延迟因子限制从所述一个或多个请求作业收紧对所述一个或多个较低优先级作业的对应的有效限制。

公开(公告)号：US07882202B2

公开(公告)日：2011-02-01

申请号：US12060501

申请日：2008-04-01

申请人： Guillermo Alejandro Alvarez ,  David Darden Chambliss ,  Prashant Pandey ,  Vijayshankar Raman ,  James Alan Ruddy ,  Garret Frederick Swart ,  Harry Morris Yudenfriend

发明人： Guillermo Alejandro Alvarez ,  David Darden Chambliss ,  Prashant Pandey ,  Vijayshankar Raman ,  James Alan Ruddy ,  Garret Frederick Swart ,  Harry Morris Yudenfriend

IPC分类号： G06F15/16

CPC分类号： H04L63/102 ,  H04L67/1097

摘要： A method is disclosed that places data-intensive subprocesses in close physical and logical proximity to the facility responsible for storing the data, so that high efficiencies at reduced cost are achieved. In one specific example, new computer programs, termed adjuncts, are added and placed in a logical partition on a storage facility so that they can be invoked using appropriate commands issued on the I/O channel. Further, programs or changes are added to existing programs on the host machine, wherein such programs or changes discover the function extensions and invoke them to perform data processing.

摘要翻译： 公开了一种将数据密集型子处理器紧密地物理和逻辑地接近负责存储数据的设施的方法，从而实现了降低成本的高效率。 在一个具体示例中，称为附件的新计算机程序被添加并放置在存储设备上的逻辑分区中，使得可以使用在I / O通道上发出的适当命令来调用它们。 此外，将程序或更改添加到主机上的现有程序，其中这些程序或改变发现功能扩展并调用它们来执行数据处理。

公开(公告)号：US20090248699A1

公开(公告)日：2009-10-01

申请号：US12060501

申请日：2008-04-01

申请人： Guillermo Alejandro Alvarez ,  David Darden Chambliss ,  Prashant Pandey ,  Vijayshankar Raman ,  James Alan Ruddy ,  Garret Frederick Swart ,  Harry Morris Yudenfriend

发明人： Guillermo Alejandro Alvarez ,  David Darden Chambliss ,  Prashant Pandey ,  Vijayshankar Raman ,  James Alan Ruddy ,  Garret Frederick Swart ,  Harry Morris Yudenfriend

IPC分类号： G06F17/30

CPC分类号： H04L63/102 ,  H04L67/1097

摘要： A method is disclosed that places data-intensive subprocesses in close physical and logical proximity to the facility responsible for storing the data, so that high efficiencies at reduced cost are achieved. In one specific example, new computer programs, termed adjuncts, are added and placed in a logical partition on a storage facility so that they can be invoked using appropriate commands issued on the I/O channel. Further, programs or changes are added to existing programs on the host machine, wherein such programs or changes discover the function extensions and invoke them to perform data processing.

摘要翻译： 公开了一种将数据密集型子处理器紧密地物理和逻辑地接近负责存储数据的设施的方法，从而实现了降低成本的高效率。 在一个具体示例中，称为附件的新计算机程序被添加并放置在存储设备上的逻辑分区中，使得可以使用在I / O通道上发出的适当命令来调用它们。 此外，将程序或更改添加到主机上的现有程序，其中这些程序或改变发现功能扩展并调用它们来执行数据处理。

公开(公告)号：US09792363B2

公开(公告)日：2017-10-17

申请号：US13019210

申请日：2011-02-01

申请人： Ryan Patrick McConville ,  Bhupendra Singh ,  Prashant Pandey ,  Chhavi Upadhyay ,  Srikanth Kakani

发明人： Ryan Patrick McConville ,  Bhupendra Singh ,  Prashant Pandey ,  Chhavi Upadhyay ,  Srikanth Kakani

IPC分类号： H04N7/173 ,  G06F17/30 ,  H04N21/2343 ,  H04N21/414 ,  H04N21/43 ,  H04N21/431 ,  H04N21/4782 ,  H04N21/61 ,  H04N21/81 ,  H04N21/485

CPC分类号： G06F17/30846 ,  H04N21/234327 ,  H04N21/234381 ,  H04N21/41407 ,  H04N21/4307 ,  H04N21/4316 ,  H04N21/4782 ,  H04N21/4852 ,  H04N21/6125 ,  H04N21/812 ,  H04N21/8153

摘要： A method for video playback uses only resources universally supported by a browser (“inline playback”) operating in virtually all handheld media devices. In one case, the method first prepares a video sequence for display by a browser by (a) dividing the video sequence into a silent video stream and an audio stream; (b) extracting from the silent video stream a number of still images, the number of still images corresponding to at least one of a desired output frame rate and a desired output resolution; and (c) combining the still images into a composite image. In one embodiment, the composite image having a number of rows, with each row being formed by the still images created from a fixed duration of the silent video stream. Another method plays the still images of the composite image as a video sequence by (a) loading the composite image to be displayed through a viewport defined the size of one of the still images; (b) selecting one of the still images of the composite image; (c) setting the viewport to display the selected still image; and (d) setting a timer for a specified time period based on a frame rate, such that, upon expiration of the specified time period: (i) selecting a next one of the still images to be displayed in the viewport, unless all still images of the composite image have been selected; and (ii) return to step (c) if not all still images have been selected.

公开(公告)号：US08612689B2

公开(公告)日：2013-12-17

申请号：US13541430

申请日：2012-07-03

申请人： James Allen Larkby-Lahet ,  Prashant Pandey

发明人： James Allen Larkby-Lahet ,  Prashant Pandey

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

CPC分类号： G06F12/126 ,  G06F12/0888 ,  G06F12/122

摘要： A process for caching data in a cache memory includes upon detecting that a first page is in a first or second list, the first page is moved to a most recently used (MRU) position in the second list. Upon detecting that the first page is in a first history list, a first target size is updated to a second target size for the first and second lists, the first page is moved from the first history list to the MRU position in the second list, and the first page is fetched to the cache memory. Upon detecting that the first page is in a second history list, the second target size is updated to a third target size for the first and second lists, and the first page is moved from the second history list to the MRU position in the second list.

摘要翻译： 用于缓存高速缓冲存储器中的数据的过程包括在检测到第一页面处于第一或第二列表中时，第一页面被移动到第二列表中的最近使用（MRU）位置。 在检测到第一页面处于第一历史列表中时，第一目标尺寸被更新为第一和第二列表的第二目标尺寸，第一页从第一历史列表移动到第二列表中的MRU位置， 并且第一页被提取到高速缓冲存储器。 在检测到第一页面处于第二历史列表中时，第二目标尺寸被更新为第一和第二列表的第三目标尺寸，并且第一页面从第二历史列表移动到第二列表中的MRU位置 。

公开(公告)号：US20120272010A1

公开(公告)日：2012-10-25

申请号：US13541430

申请日：2012-07-03

申请人： James Allen Larkby-Lahet ,  Prashant Pandey

发明人： James Allen Larkby-Lahet ,  Prashant Pandey

IPC分类号： G06F12/12

CPC分类号： G06F12/126 ,  G06F12/0888 ,  G06F12/122

摘要： A process for caching data in a cache memory includes upon detecting that a first page is in a first or second list, the first page is moved to a most recently used (MRU) position in the second list. Upon detecting that the first page is in a first history list, a first target size is updated to a second target size for the first and second lists, the first page is moved from the first history list to the MRU position in the second list, and the first page is fetched to the cache memory. Upon detecting that the first page is in a second history list, the second target size is updated to a third target size for the first and second lists, and the first page is moved from the second history list to the MRU position in the second list.

摘要翻译： 用于缓存高速缓冲存储器中的数据的过程包括在检测到第一页面处于第一或第二列表中时，第一页面被移动到第二列表中的最近使用（MRU）位置。 在检测到第一页面处于第一历史列表中时，第一目标尺寸被更新为第一和第二列表的第二目标尺寸，第一页从第一历史列表移动到第二列表中的MRU位置， 并且第一页被提取到高速缓冲存储器。 在检测到第一页面处于第二历史列表中时，第二目标尺寸被更新为第一和第二列表的第三目标尺寸，并且第一页面从第二历史列表移动到第二列表中的MRU位置 。

公开(公告)号：US20120054429A1

公开(公告)日：2012-03-01

申请号：US12871970

申请日：2010-08-31

申请人： Karan Gupta ,  Reshu Jain ,  Prashant Pandey ,  Himabindu Pucha

发明人： Karan Gupta ,  Reshu Jain ,  Prashant Pandey ,  Himabindu Pucha

IPC分类号： G06F12/02 ,  G06F12/16 ,  G06F12/00

CPC分类号： G06F3/0631 ,  G06F3/0604 ,  G06F3/064 ,  G06F3/067

摘要： Embodiments of the invention relate to block layout and block allocation in a file system to support write transactions. Regions in a cluster file system are defined to support a block allocation, include both write affinity and wide striping region. An allocation map is maintained to define and support the block allocation regions. For each write transaction, a copy of data is written to both regions, and in the event there is a shortage of blocks in one of the regions, at least one block in the other region is dynamically converted to support the write transaction.

摘要翻译： 本发明的实施例涉及文件系统中的块布局和块分配以支持写入事务。 定义集群文件系统中的区域以支持块分配，包括写入亲和度和宽条带区域。 维护分配图以定义和支持块分配区域。 对于每个写入事务，数据的副本被写入两个区域，并且在其中一个区域中存在块的不足的情况下，另一个区域中的至少一个块被动态转换以支持写入事务。

公开(公告)号：US08103903B2

公开(公告)日：2012-01-24

申请号：US12710123

申请日：2010-02-22

申请人： James L. Hafner ,  Prashant Pandey ,  Tarun Thakur

发明人： James L. Hafner ,  Prashant Pandey ,  Tarun Thakur

IPC分类号： G06F11/00

CPC分类号： G06F11/1076 ,  G06F11/1474 ,  G06F2211/1028 ,  G06F2211/1057

摘要： Data storage reliability is maintained in a write-back distributed data storage system including multiple nodes, each node comprising a processor and an array of failure independent data storage devices. Information is stored as a set of stripes, each stripe including a collection of multiple data strips and associated parity strips, the stripes distributed across multiple corresponding primary data nodes and multiple corresponding parity nodes. A primary data node maintains the data strip holding a first copy of data, and each parity node maintains a parity strip holding a parity for the multiple data strips. A read-modify-write parity update protocol is performed for maintaining parity coherency, the primary data node driving parity coherency with its corresponding parity nodes, independently of other data nodes, in order to keep its relevant parity strips coherent.

摘要翻译： 在包括多个节点的回写分布式数据存储系统中维持数据存储可靠性，每个节点包括处理器和故障独立数据存储设备的阵列。 信息被存储为一组条带，每个条带包括多个数据条带和相关联的奇偶校验条的集合，条带分布在多个对应的主数据节点和多个对应的奇偶校验节点上。 主数据节点保持数据带保持数据的第一副本，并且每个奇偶校验节点维持保持多个数据条的奇偶校验的奇偶校验条。 执行读 - 修改 - 写奇偶校验更新协议以维持奇偶校验一致性，主数据节点与其对应的奇偶校验节点驱动奇偶校验一致性，独立于其他数据节点，以便保持其相关奇偶校验条相符。