公开(公告)号：US08862930B2

公开(公告)日：2014-10-14

申请号：US12953972

申请日：2010-11-24

申请人： Neil R. T. Horman ,  Vivek Goyal

发明人： Neil R. T. Horman ,  Vivek Goyal

IPC分类号： G06F11/00 ,  G06F11/14 ,  G06F11/34

CPC分类号： G06F11/1415 ,  G06F11/1417 ,  G06F11/3409

摘要： A computing system stores actual memory usage data in a user memory space. The actual memory usage data represents memory usage of a plurality of device drivers that are loaded by a first kernel. The computing system generates an estimate of memory space to be reserved for a second kernel based on the actual memory usage data for the plurality of device drivers that are loaded by the first kernel and reserves memory space for the second kernel using the estimate.

摘要翻译： 计算系统将实际的存储器使用数据存储在用户存储器空间中。 实际的存储器使用数据表示由第一内核加载的多个设备驱动程序的存储器使用。 计算系统基于由第一内核加载的多个设备驱动程序的实际存储器使用数据生成将为第二内核保留的存储器空间的估计，并且使用估计为第二内核预留存储器空间。

公开(公告)号：US08219716B2

公开(公告)日：2012-07-10

申请号：US12702136

申请日：2010-02-08

申请人： Vivek Goyal

发明人： Vivek Goyal

IPC分类号： G06F3/00 ,  G06F13/00

CPC分类号： G06F3/0611 ,  G06F3/0659 ,  G06F3/0676

摘要： An input and output (IO) scheduler dispatches a first IO request from a first IO queue during a first dispatch cycle and dispatches a second IO request from a second IO queue during a second dispatch cycle to one or more disks, where the first IO request is the last remaining IO request in the first queue and the second IO request is the first overall IO request in the second queue. The first and second IO requests are selected from multiple IO requests received from one or more clients for accessing the one or more disks. A seek time monitor measures disk seek time (IO queue switch time) incurred between the first IO request and the second IO request, and determines whether the first IO request and the second IO request belong to the same group. The disk seek time is attributed to a group associated with the first IO request and the second IO request, if the first IO request and the second IO request belong to the same group. If not, then this seek time incurred during queue switch is attributed to a common ancestor of two IO queues in the IO hierarchy.

摘要翻译： 输入和输出（IO）调度器在第一调度周期期间从第一IO队列调度第一IO请求，并且在第二调度周期期间将第二IO请求从第二IO队列调度到一个或多个磁盘，其中第一IO请求 是第一个队列中最后剩余的IO请求，第二个IO请求是第二个队列中的第一个整体IO请求。 从用于访问一个或多个磁盘的一个或多个客户端接收到的多个IO请求中选择第一和第二IO请求。 寻道时间监视器测量在第一个IO请求和第二个IO请求之间产生的磁盘查找时间（IO队列切换时间），并确定第一个IO请求和第二个IO请求是否属于同一个组。 如果第一个IO请求和第二个IO请求属于同一个组，则磁盘查找时间归因于与第一个IO请求和第二个IO请求相关联的组。 如果没有，则在队列切换期间产生的此查找时间归因于IO层次结构中两个IO队列的共同祖先。

公开(公告)号：US08148985B2

公开(公告)日：2012-04-03

申请号：US12580076

申请日：2009-10-15

申请人： Adam C Zelinski ,  Kawin Setsompop ,  Elfar Adalsteinsson ,  Vivek Goyal

发明人： Adam C Zelinski ,  Kawin Setsompop ,  Elfar Adalsteinsson ,  Vivek Goyal

IPC分类号： G01V3/00

CPC分类号： G01R33/583 ,  G01R33/288 ,  G01R33/5612

摘要： A method for reducing maximum local specific absorption rate (“SAR”) in a magnetic resonance imaging (“MRI”) system is disclosed. More specifically, a plurality of candidate radio frequency (“RF”) pulses are designed and the manner in which they are applied to a subject is determined such that the maximum local SAR is substantially reduced relative to applying the candidate RF pulse that produces the lowest maximum local SAR alone. Put another way, this “time-multiplexing” of a set of RF pulses that each produce approximately the same excitation pattern yields a lower maximum local SAR than does transmitting the individual RF pulse having the lowest local SAR over many repetition times (“TRs”). A convex optimization method is utilized to determine the manner in which the RF pulses are multiplexed in time such that a substantially lower maximum local SAR is achieved.

摘要翻译： 公开了一种在磁共振成像（“MRI”）系统中降低最大局部比吸收率（“SAR”）的方法。 更具体地，设计多个候选射频（“RF”）脉冲，并且确定它们被施加到被摄体的方式，使得最大局部SAR相对于施加产生最低的候选射频脉冲而显着减小 最大的地方特区。 换句话说，每个产生大致相同的激励模式的一组RF脉冲的这种“时间复用”产生比在许多重复时间（“TR”）发送具有最低局部SAR的单个RF脉冲的最小局部SAR更低的最小局部SAR， ）。 使用凸优化方法来确定RF脉冲在时间上多路复用的方式，使得实现基本上较低的最大局部SAR。

公开(公告)号：US20110196996A1

公开(公告)日：2011-08-11

申请号：US12702136

申请日：2010-02-08

申请人： Vivek Goyal

发明人： Vivek Goyal

IPC分类号： G06F3/00 ,  G06F9/46 ,  G06F12/08

CPC分类号： G06F3/0611 ,  G06F3/0659 ,  G06F3/0676

摘要： An input and output (IO) scheduler dispatches a first IO request from a first IO queue during a first dispatch cycle and dispatches a second IO request from a second IO queue during a second dispatch cycle to one or more disks, where the first IO request is the last remaining IO request in the first queue and the second IO request is the first overall IO request in the second queue. The first and second IO requests are selected from multiple IO requests received from one or more clients for accessing the one or more disks. A seek time monitor measures disk seek time (IO queue switch time) incurred between the first IO request and the second IO request, and determines whether the first IO request and the second IO request belong to the same group. The disk seek time is attributed to a group associated with the first IO request and the second IO request, if the first IO request and the second IO request belong to the same group. If not, then this seek time incurred during queue switch is attributed to a common ancestor of two IO queues in the IO hierarchy.

摘要翻译： 输入和输出（IO）调度器在第一调度周期期间从第一IO队列调度第一IO请求，并且在第二调度周期期间将第二IO请求从第二IO队列调度到一个或多个磁盘，其中第一IO请求 是第一个队列中最后剩余的IO请求，第二个IO请求是第二个队列中的第一个整体IO请求。 从用于访问一个或多个磁盘的一个或多个客户端接收到的多个IO请求中选择第一和第二IO请求。 寻道时间监视器测量在第一个IO请求和第二个IO请求之间产生的磁盘查找时间（IO队列切换时间），并确定第一个IO请求和第二个IO请求是否属于同一个组。 如果第一个IO请求和第二个IO请求属于同一个组，则磁盘查找时间归因于与第一个IO请求和第二个IO请求相关联的组。 如果没有，则在队列切换期间产生的此查找时间归因于IO层次结构中两个IO队列的共同祖先。

公开(公告)号：US20100134105A1

公开(公告)日：2010-06-03

申请号：US12580076

申请日：2009-10-15

申请人： Adam C. Zelinski ,  Kawin Setsompop ,  Elfar Adalsteinsson ,  Vivek Goyal

发明人： Adam C. Zelinski ,  Kawin Setsompop ,  Elfar Adalsteinsson ,  Vivek Goyal

IPC分类号： G01R33/48

CPC分类号： G01R33/583 ,  G01R33/288 ,  G01R33/5612

摘要： A method for reducing maximum local specific absorption rate (“SAR”) in a magnetic resonance imaging (“MRI”) system is disclosed. More specifically, a plurality of candidate radio frequency (“RF”) pulses are designed and the manner in which they are applied to a subject is determined such that the maximum local SAR is substantially reduced relative to applying the candidate RF pulse that produces the lowest maximum local SAR alone. Put another way, this “time-multiplexing” of a set of RF pulses that each produce approximately the same excitation pattern yields a lower maximum local SAR than does transmitting the individual RF pulse having the lowest local SAR over many repetition times (“TRs”). A convex optimization method is utilized to determine the manner in which the RF pulses are multiplexed in time such that a substantially lower maximum local SAR is achieved.

摘要翻译： 公开了一种在磁共振成像（“MRI”）系统中降低最大局部比吸收率（“SAR”）的方法。 更具体地，设计多个候选射频（“RF”）脉冲，并且确定它们被施加到被摄体的方式，使得最大局部SAR相对于施加产生最低的候选射频脉冲而显着减小 最大的地方特区。 换句话说，每个产生大致相同的激励模式的一组RF脉冲的这种“时间复用”产生比在许多重复时间（“TR”）发送具有最低局部SAR的单个RF脉冲的最小局部SAR更低的最小局部SAR， ）。 使用凸优化方法来确定RF脉冲在时间上多路复用的方式，使得实现基本上较低的最大局部SAR。

公开(公告)号：US07249291B2

公开(公告)日：2007-07-24

申请号：US10367573

申请日：2003-02-14

申请人： Jens Rasmussen ,  Amin Shokrollahi ,  Soren Lassen ,  Gavin Horn ,  Vivek Goyal ,  Barry Dobyns ,  Michael Luby

发明人： Jens Rasmussen ,  Amin Shokrollahi ,  Soren Lassen ,  Gavin Horn ,  Vivek Goyal ,  Barry Dobyns ,  Michael Luby

IPC分类号： H03M13/05

CPC分类号： H04N21/4382 ,  H04L1/004 ,  H04L1/0041 ,  H04L1/0083 ,  H04L2001/0093 ,  H04L2001/0096 ,  H04N21/234327 ,  H04N21/2383 ,  H04N21/2385 ,  H04N21/2389 ,  H04N21/2662 ,  H04N21/4385 ,  H04N21/631 ,  H04N21/64792 ,  H04N21/8456

摘要： A method for communicating the content of a live data stream to a receiver using a plurality of channels comprising two encoder channels used to encode the live data content prior to transmission. Initially a plurality of segments of a live data stream are received, wherein each segment contains segment data. A forward error correction algorithm is applied to each segment's data, thereby producing FEC-encoded segment data. The FEC-encoded segment data is contained within an FEC-encoded block, resulting in a corresponding plurality of FEC-encoded blocks being generated. Each FEC-encoded block is copied to a sub-channel on both a first encoder channel and a second encoder channel, resulting in a plurality of FEC-encoder blocks residing on the first and second encoder channels. The first and second encoder channels differ in the number of sub-channels they contain (interleaving depth), and accordingly the arrangement of the FEC-encoded blocks in the first and second encoder channels are different. A first cross-section of the FEC-encoded segment data contained within the FEC-encoded blocks resident in the first encoder channel is added to a first transmit block T0. Similarly, A first cross-section of the FEC-encoded segment data contained within the FEC-encoded blocks resident in the second encoder channel is added to a second transmit block T1. The first and second transmit blocks are then communicated to the receiver.

摘要翻译： 一种用于使用包括用于在传输之前对实时数据内容进行编码的两个编码器信道的多个信道将实时数据流的内容传送到接收机的方法。 最初，接收实时数据流的多个段，其中每个段包含段数据。 向每个段的数据应用前向纠错算法，从而产生FEC编码的段数据。 FEC编码的段数据被包含在FEC编码的块内，导致生成相应的多个FEC编码的块。 每个FEC编码的块被复制到第一编码器信道和第二编码器信道上的子信道，导致驻留在第一和第二编码器信道上的多个FEC编码器块。 第一和第二编码器通道在它们包含的子通道数量（交织深度）上不同，因此第一和第二编码器通道中的FEC编码块的排列是不同的。 包含在驻留在第一编码器信道中的FEC编码块内的FEC编码段数据的第一横截面被添加到第一传输块T 0 0。 类似地，包含在驻留在第二编码器信道中的FEC编码块中的FEC编码段数据的第一横截面被添加到第二传输块T 1。 然后将第一和第二发送块传送到接收器。

公开(公告)号：US20120131380A1

公开(公告)日：2012-05-24

申请号：US12953972

申请日：2010-11-24

申请人： Neil R.T. Horman ,  Vivek Goyal

发明人： Neil R.T. Horman ,  Vivek Goyal

IPC分类号： G06F11/16 ,  G06F12/02 ,  G06F12/00

CPC分类号： G06F11/1415 ,  G06F11/1417 ,  G06F11/3409

摘要： A computing system stores actual memory usage data in a user memory space. The actual memory usage data represents memory usage of a plurality of device drivers that are loaded by a first kernel. The computing system generates an estimate of memory space to be reserved for a second kernel based on the actual memory usage data for the plurality of device drivers that are loaded by the first kernel and reserves memory space for the second kernel using the estimate.

摘要翻译： 计算系统将实际的存储器使用数据存储在用户存储器空间中。 实际的存储器使用数据表示由第一内核加载的多个设备驱动程序的存储器使用。 计算系统基于由第一内核加载的多个设备驱动程序的实际存储器使用数据生成将为第二内核保留的存储器空间的估计，并且使用估计为第二内核预留存储器空间。