公开(公告)号：US20090031058A1

公开(公告)日：2009-01-29

申请号：US11966760

申请日：2007-12-28

Applicant: Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William T. Futral ,  Sujoy Sen ,  Gregory D. Cummings ,  Kenneth C. Creta ,  David C. Lee

Inventor： Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William T. Futral ,  Sujoy Sen ,  Gregory D. Cummings ,  Kenneth C. Creta ,  David C. Lee

IPC: G06F13/00 ,  G06F3/00

CPC classification number: G06F13/1631 ,  G06F12/0831 ,  G06F12/0879 ,  G06F12/0897 ,  H04L69/12

Abstract: Methods and apparatuses for flushing write-combined data from a buffer within a memory to an input/output (I/O) device.

Abstract translation: 将写入组合数据从存储器中的缓冲器冲洗到输入/输出（I / O）设备的方法和装置。

公开(公告)号：US07353301B2

公开(公告)日：2008-04-01

申请号：US10977235

申请日：2004-10-29

Applicant: Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William T. Futral ,  Sujoy Sen ,  Gregory D. Cummings ,  Kenneth C. Creta ,  David C. Lee

Inventor： Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William T. Futral ,  Sujoy Sen ,  Gregory D. Cummings ,  Kenneth C. Creta ,  David C. Lee

IPC: G06F3/00 ,  G06G3/00

CPC classification number: G06F13/1631 ,  G06F12/0831 ,  G06F12/0879 ,  G06F12/0897 ,  H04L69/12

Abstract: Write-combining in a computer system that uses a push model is set forth herein. In one embodiment, the method comprises creating one or more packets having a descriptor and the data associated with detected write transactions stored in the buffer assigned to a write-combinable range in response to a flush request to flush the buffer, and sending (pushing) these packets to the network I/O device.

Abstract translation: 本文介绍了使用推送模型的计算机系统中的写入组合。 在一个实施例中，该方法包括：响应于冲洗缓冲器的刷新请求，创建一个或多个具有描述符的数据包，以及存储在缓冲区中的检测到的写入事务的数据，该数据被分配给可写入可组合范围，并发送（推送） 这些数据包到网络I / O设备。

公开(公告)号：US20060136639A1

公开(公告)日：2006-06-22

申请号：US11020500

申请日：2004-12-21

Applicant: William Futral ,  Kenneth Creta ,  Sujoy Sen ,  Gregory Cummings ,  Sivakumar Radhakrishnan

Inventor： William Futral ,  Kenneth Creta ,  Sujoy Sen ,  Gregory Cummings ,  Sivakumar Radhakrishnan

IPC: G06F13/14

CPC classification number: G06F13/4291

Abstract: A system, method and apparatus for prioritizing transactions is disclosed. I/O devices may generate transactions with a stream identifier. A transaction classifier may assign a priority to a transaction based upon a stream identifier of the transaction. An arbiter may select a transaction for processing based upon priorities assigned to the transaction.

公开(公告)号：US20060095609A1

公开(公告)日：2006-05-04

申请号：US10977235

申请日：2004-10-29

Applicant: Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William Futral ,  Sujoy Sen ,  Gregory Cummings ,  Kenneth Creta ,  David Lee

Inventor： Sivakumar Radhakrishnan ,  Siva Balasubramanian ,  William Futral ,  Sujoy Sen ,  Gregory Cummings ,  Kenneth Creta ,  David Lee

IPC: G06F5/00

CPC classification number: G06F13/1631 ,  G06F12/0831 ,  G06F12/0879 ,  G06F12/0897 ,  H04L69/12

Abstract: Write-combining in a computer system that uses a push model is set forth herein. In one embodiment, the method comprises creating one or more packets having a descriptor and the data associated with detected write transactions stored in the buffer assigned to a write-combinable range in response to a flush request to flush the buffer, and sending (pushing) these packets to the network I/O device.

Abstract translation: 本文介绍了使用推送模型的计算机系统中的写入组合。 在一个实施例中，该方法包括：响应于冲洗缓冲器的刷新请求，创建一个或多个具有描述符的数据包，以及存储在缓冲区中的检测到的写入事务的数据，该数据被分配给可写入可组合范围，并发送（推送） 这些数据包到网络I / O设备。

公开(公告)号：US5751697A

公开(公告)日：1998-05-12

申请号：US568347

申请日：1995-12-06

Applicant: Sivakumar Radhakrishnan ,  Stephen J. Dabecki

Inventor： Sivakumar Radhakrishnan ,  Stephen J. Dabecki

IPC: G06F7/52 ,  G06F7/535 ,  H04L12/56 ,  H04Q11/04 ,  H04L12/24 ,  G06F1/02

CPC classification number: G06F7/535 ,  H04Q11/0478 ,  H04L2012/5635 ,  H04L2012/5684 ,  Y10S370/905

Abstract: A method of scheduling cell transmission over an asynchronous transfer mode communication channel. The channel has a characteristic transmission rate that is related to the system clock frequency .function. and an allowed cell rate ACR expressed as a floating point number having a mantissa m, and an exponent e, where 0.ltoreq.m.ltoreq.511, 0.ltoreq.e.ltoreq.31 and ACR=(1+m/512)*2.sup.e. If m.gtoreq.128 then the reciprocal of the mantissa portion (1+m/512) is evaluated by piece-wise linear approximation of the function: ##EQU1## Otherwise, if m

Abstract translation: 一种通过异步传输模式通信信道调度小区传输的方法。 信道具有与系统时钟频率f相关的特征传输速率，以及表示为具有尾数m的浮点数的允许信元速率ACR和指数e，其中0≤m≤511,0 / = 128，则尾数部分的倒数（1 + m / 512）通过函数的分段线性近似来估计：否则，如果m <128，则尾数部分通过分段估计 函数的明智的线性近似：然后，调度所选择的单元在相对于所选单元之前发送的单元的传输时间T0的时间T发送，其中T = T0 +（ACR-1）* f 。 本发明的硬件实施例仅需要一个9位二进制补码器，三个11位加法器和用于多路复用/控制的胶合逻辑就可以使用符合ATM论坛流量管理标准的VLSI电路来实现，而不需要硬件划分或 乘法。

公开(公告)号：US5706288A

公开(公告)日：1998-01-06

申请号：US622398

申请日：1996-03-27

Applicant: Sivakumar Radhakrishnan ,  Stephen J. Dabecki ,  David Walden Wong

Inventor： Sivakumar Radhakrishnan ,  Stephen J. Dabecki ,  David Walden Wong

IPC: H04L12/56 ,  H04Q11/04 ,  H04J3/22

CPC classification number: H04Q11/0478 ,  H04L2012/5635 ,  H04L2012/5679 ,  H04L2012/568

Abstract: An available bit rate scheduling method and apparatus for asynchronous transfer mode communication of a plurality of cells over a network characterized by a system clock frequency f and an allowed cell rate ACR. Each cell belongs to an assigned virtual circuit communication channel which is defined by a set of negotiated traffic parameters. The invention partitions the ACR's of the virtual circuits into a smaller subset of profiles/sub-profiles and conducts a deterministic search to service them. The scheduler incorporates a profile generator for iteratively generating a number p of the profiles by (i) outputting a k*modulo 2.sup.i th one of the profiles during each kth iteration of the profile generator, where 1.ltoreq.i.ltoreq.p and 1.ltoreq.k.ltoreq.p-1; (ii) outputting a null profile during each 2.sup.p th one of the iterations; and, (iii) dispatching the profiles from the profile generator to the profile queue such that a particular profile is dispatched at a time T=T.sub.0 +(1/ACR)*f, where T.sub.0 is the dispatch time of a profile dispatched immediately prior to the particular profile. A profile queue coupled to the profile generator receives and sequentially stores the generated profiles. A virtual circuit processor sequentially receives the profiles from the profile queue and, for each one of the received profiles, dispatches to an output queue all virtual circuits which are characterized by the one received profile.

Abstract translation: 一种用于通过网络异步传输模式通信的可用比特率调度方法和装置，其特征在于系统时钟频率f和允许的小区速率ACR。 每个小区属于由一组协商的业务参数定义的分配的虚拟电路通信信道。 本发明将虚拟电路的ACR分成简档/子简档的较小子集，并进行确定性搜索以对其进行维护。 调度器包括一个简档生成器，用于通过（i）在简档生成器的每个第k次迭代期间输出ak *模2个模型之一来迭代地生成数据p，其中1 i = p和1 < / = k