公开(公告)号：US20050086390A1

公开(公告)日：2005-04-21

申请号：US10687235

申请日：2003-10-16

申请人： Dwip Banerjee ,  Kavitha Baratakke ,  Vinit Jain ,  Venkat Venkatsubra

发明人： Dwip Banerjee ,  Kavitha Baratakke ,  Vinit Jain ,  Venkat Venkatsubra

IPC分类号： G06F13/42 ,  G06F15/16 ,  H04L29/06

CPC分类号： H04L69/16 ,  H04L69/161 ,  H04L69/22

摘要： A method, system, and program for efficient packet desegmentation on a network adapter are provided. Multiple data packet segments received at a network adapter from a single connection are buffered at the network adapter. The single connection is identified by addresses and ports extracted from the header of each data packet segment. Responsive to detecting a buffering release condition, the data packet segments are released from the network adapter as a desegmented group to a network stack, such that the data packets segments received for the single connection are efficiently passed to the network stack together. In particular, the single connection is a TCP connection identified by a four-tuple of source and destination addresses and ports extracted from each TCP header of each of said plurality of data packet segments.

摘要翻译： 提供了一种用于在网络适配器上实现有效的分组分段的方法，系统和程序。 在网络适配器处从单个连接接收的多个数据分组段被缓冲。 单个连接由从每个数据分组段的报头提取的地址和端口来标识。 响应于检测缓冲释放条件，将数据分组片段作为被分组的组从网络适配器释放到网络堆栈，使得为单个连接接收的数据分组片段被有效地传递到网络堆栈。 特别地，单个连接是由从所述多个数据分组段中的每一个的每个TCP报头提取的源和目的地地址和端口的四元组标识的TCP连接。

公开(公告)号：US08422501B2

公开(公告)日：2013-04-16

申请号：US11619676

申请日：2007-01-04

申请人： Ranadip Das ,  Vinit Jain ,  Uma Krishnan ,  Venkat Venkatsubra

发明人： Ranadip Das ,  Vinit Jain ,  Uma Krishnan ,  Venkat Venkatsubra

IPC分类号： H04L12/56 ,  H04L12/28 ,  G06F11/00

CPC分类号： H04L47/10 ,  H04L47/36

摘要： A method, computer program product, and data processing system for efficiently discovering and storing path MTU information in a sending host are disclosed. In a preferred embodiment, two path MTU tables are maintained. One path MTU table contains MTU values corresponding to the first-hop routers associated with the sending host. The other path MTU table contains MTU values corresponding to individual destination hosts. When the sending host needs to send information to a destination, it first consults the MTU table associated with individual destination hosts. If an entry for that destination host is found in the table, the sending host uses that MTU value. If not, the sending host consults the MTU table for the first-hop router on the path to the destination host and uses that MTU value. If that MTU value is too high, a new entry is made in the host-specific MTU table for the destination host.

摘要翻译： 公开了一种用于在发送主机中有效发现和存储路径MTU信息的方法，计算机程序产品和数据处理系统。 在优选实施例中，保持两个路径MTU表。 一个路径MTU表包含与发送主机关联的第一跳路由器对应的MTU值。 另一路径MTU表包含对应于各个目标主机的MTU值。 当发送主机需要向目的地发送信息时，首先查询与各个目的主机关联的MTU表。 如果表中找到该目标主机的条目，则发送主机将使用该MTU值。 如果不是，则发送主机在目的地主机的路径上查询第一跳路由器的MTU表，并使用该MTU值。 如果该MTU值太高，则在目标主机的主机特定MTU表中创建一个新条目。

公开(公告)号：US07913106B2

公开(公告)日：2011-03-22

申请号：US11958548

申请日：2007-12-18

申请人： Vinit Jain ,  Mallesh Lepakshaiah ,  Elizabeth J. Murray ,  Venkat Venkatsubra

发明人： Vinit Jain ,  Mallesh Lepakshaiah ,  Elizabeth J. Murray ,  Venkat Venkatsubra

IPC分类号： G06F11/00

CPC分类号： H04L29/12028 ,  H04L61/103 ,  H04L69/18 ,  H04L69/40

摘要： A host enables any adapter of multiple adapters of the host to concurrently support any VIPA of the multiple VIPAs assigned to the host. Responsive to a failure of at least one particular adapter from among the multiple adapters, the host triggers the remaining, functioning adapters to broadcast a separate hardware address update for each VIPA over the network, such that for a failover in the host supporting the multiple VIPAs the host directs at least one other host accessible via the network to address any new packets for the multiple VIPAs to one of the separate hardware addresses of one of the remaining adapters.

摘要翻译： 主机允许主机的多个适配器的任何适配器同时支持分配给主机的多个VIPA的任何VIPA。 响应于多个适配器中的至少一个特定适配器的故障，主机触发剩余的功能适配器以通过网络为每个VIPA广播单独的硬件地址更新，使得对于支持多个VIPA的主机中的故障转移 主机引导至少一个可通过网络访问的其他主机，以将多个VIPA的任何新数据包解析为其余适配器之一的单独硬件地址之一。

公开(公告)号：US07519724B2

公开(公告)日：2009-04-14

申请号：US12101750

申请日：2008-04-11

申请人： Venkat Venkatsubra ,  Vinit Jain

发明人： Venkat Venkatsubra ,  Vinit Jain

IPC分类号： G06F15/16

CPC分类号： H04L65/4092 ,  H04L29/06027 ,  H04L65/608

摘要： A graphic user interface of a network client (106) includes a stock ticker from a stock server (110) and a news sidebar from a news server (112) over the IP addresses advertised to the client (106) in its list of multi-homed addresses from the video server (104) specified under Stream Control Transmission Protocol (SCTP). The client accepts real-time data from the stock exchange server and the news agency server on the multi-homed IP addresses designated in the association with the video server (104) without knowing that the data is coming from a different source than the video server (104). The real-time data feeds from the video, stock and news servers are aggregated on the client (106) with enhanced speed because the feeds come directly to the client and not via the video server. The operating systems of the home server and remote servers utilize SCTP and specialized commands to implement the enhanced speed of real-time data aggregation being streamed to network clients without requiring modifications to existing client systems.

摘要翻译： 网络客户端（106）的图形用户界面包括来自库存服务器（110）的股票行情和来自新闻服务器（112）的新闻侧栏，通过在其多媒体列表中通告给客户端（106）的IP地址， 来自在流控制传输协议（SCTP）下指定的视频服务器（104）的归属地址。 客户端接收来自证券交易所服务器和通讯社服务器的与视频服务器（104）相关联指定的多归属IP地址的实时数据，而不知道数据来自与视频服务器不同的源 （104）。 来自视频，股票和新闻服务器的实时数据馈送以高速度聚合在客户端（106）上，因为馈送直接进入客户端而不是经由视频服务器。 家庭服务器和远程服务器的操作系统使用SCTP和专用命令来实现流式传输到网络客户端的实时数据聚合的增强速度，而不需要修改现有的客户端系统。

公开(公告)号：US07483990B2

公开(公告)日：2009-01-27

申请号：US10443667

申请日：2003-05-22

申请人： Kavitha Vittal Murthy Baratakke ,  Vinit Jain ,  Ketan Priyakant Pancholi ,  Venkat Venkatsubra

发明人： Kavitha Vittal Murthy Baratakke ,  Vinit Jain ,  Ketan Priyakant Pancholi ,  Venkat Venkatsubra

IPC分类号： G06F15/16

CPC分类号： H04L47/263 ,  H04L47/10 ,  H04L47/193 ,  H04L47/30 ,  H04L67/14 ,  H04L69/16 ,  H04L69/163 ,  H04L69/329

摘要： The server side Transfer Control Protocol is modified so that the server returns a SYNACK message with the window size equal to zero if the server is busy. When a client sends a TCP connection request and receives a synchronization acknowledgement message with the window size equal to zero, the client knows that the server received the connection request and that the server is busy. The client may then send an acknowledgement message to complete the three-way synchronization handshake, thus successfully completing the connection. Thereafter, the client side TCP may probe the server side TCP until a window update message is received from the server. When the server sends a window update message to set the window size to a non-zero size, the client knows that the server is no longer busy and the client application may then use the TCP connection.

摘要翻译： 修改服务器端传输控制协议，以便服务器返回SYNACK消息，如果服务器正忙，则窗口大小等于零。 当客户端发送TCP连接请求并接收到窗口大小等于零的同步确认消息时，客户端知道服务器接收到连接请求，并且服务器正忙。 然后，客户端可以发送确认消息以完成三次同步握手，从而成功地完成连接。 此后，客户端TCP可以探测服务器端TCP，直到从服务器接收到窗口更新消息。 当服务器发送窗口更新消息以将窗口大小设置为非零大小时，客户端知道服务器不再忙，客户端应用程序可能会使用TCP连接。

公开(公告)号：US20080195701A1

公开(公告)日：2008-08-14

申请号：US12101750

申请日：2008-04-11

申请人： Venkat Venkatsubra ,  Vinit Jain

发明人： Venkat Venkatsubra ,  Vinit Jain

IPC分类号： G06F15/16

CPC分类号： H04L65/4092 ,  H04L29/06027 ,  H04L65/608

摘要： A graphic user interface of a network client (106) includes a stock ticker from a stock server (110) and a news sidebar from a news server (112) over the IP addresses advertised to the client (106) in its list of multi-homed addresses from the video server (104) specified under Stream Control Transmission Protocol (SCTP). The client accepts real-time data from the stock exchange server and the news agency server on the multi-homed IP addresses designated in the association with the video server (104) without knowing that the data is coming from a different source than the video server (104). The real-time data feeds from the video, stock and news servers are aggregated on the client (106) with enhanced speed because the feeds come directly to the client and not via the video server. The operating systems of the home server and remote servers utilize SCTP and specialized commands to implement the enhanced speed of real-time data aggregation being streamed to network clients without requiring modifications to existing client systems.

摘要翻译： 网络客户端（106）的图形用户界面包括来自库存服务器（110）的股票行情和来自新闻服务器（112）的新闻侧栏，通过在其多媒体列表中通告给客户端（106）的IP地址， 来自在流控制传输协议（SCTP）下指定的视频服务器（104）的归属地址。 客户端接收来自证券交易所服务器和通讯社服务器的与视频服务器（104）相关联指定的多归属IP地址的实时数据，而不知道数据来自与视频服务器不同的源 （104）。 来自视频，股票和新闻服务器的实时数据馈送以高速度聚合在客户端（106）上，因为馈送直接进入客户端而不是经由视频服务器。 家庭服务器和远程服务器的操作系统使用SCTP和专用命令来实现流式传输到网络客户端的实时数据聚合的增强速度，而不需要修改现有的客户端系统。

公开(公告)号：US20080056263A1

公开(公告)日：2008-03-06

申请号：US11470455

申请日：2006-09-06

申请人： Vinit Jain ,  Alexander Medvedev ,  Ketan P. Pancholi ,  Venkat Venkatsubra

发明人： Vinit Jain ,  Alexander Medvedev ,  Ketan P. Pancholi ,  Venkat Venkatsubra

IPC分类号： H04L12/56

CPC分类号： H04L69/16 ,  H04L69/161 ,  H04L69/162

摘要： A method for improving the efficiency of Internet Protocol Control Block (INPCB) lookup operations at a TCP receiver and TCP sender. The TCP/IP stack is modified such that every data packet generated includes both data and INPCB address information. Specifically, the TCP/IP stacks of a sending device and a receiving device are modified to provide a contemporaneous transmission of INPCB addresses along with each data packet. The TCP/IP stack of the receiver is also modified such that the receiver echoes this INPCB address along with each TCP packet sent back to the sender to eliminate the necessity of an INPCB address lookup operation at the sending device. Contemporaneous transmission of data and INPCB addresses facilitates direct delivery of data packets to the corresponding sockets.

摘要翻译： 一种用于提高TCP接收器和TCP发送器上的因特网协议控制块（INPCB）查找操作的效率的方法。 修改TCP / IP堆栈，使得每个生成的数据包都包括数据和INPCB地址信息。 具体地说，发送设备和接收设备的TCP / IP堆栈被修改以提供与每个数据分组一起的INPCB地址的同时传输。 接收机的TCP / IP堆栈也被修改，使得接收机回送这个INPCB地址以及发回到发送器的每个TCP分组，以消除在发送设备处的INPCB地址查找操作的必要性。 数据和INPCB地址的同时传输有助于将数据包直接传递到相应的插座。

公开(公告)号：US07283527B2

公开(公告)日：2007-10-16

申请号：US10087939

申请日：2002-02-27

申请人： Deanna Lynn Quigg Brown ,  Vinit Jain ,  Jeffrey Paul Messing ,  Venkat Venkatsubra

发明人： Deanna Lynn Quigg Brown ,  Vinit Jain ,  Jeffrey Paul Messing ,  Venkat Venkatsubra

IPC分类号： H04L12/56 ,  H04J3/24 ,  G06F15/16

CPC分类号： H04L69/16 ,  H04L69/161 ,  H04L69/166 ,  H04L69/168

摘要： A method, system and apparatus for maintaining a two-byte identification field of IP headers when a Gigabit Ethernet is in use are provided. Since the IP identification field is of two bytes, there is a limited number of identification numbers that may be used. Due to the high data transfer rate of the Gigabit Ethernet, this number may be reached within one second. Hence, there may be a possibility that two or more packets may have the same identification number while in transit. In a representative embodiment, one identification number is used for packets that may not be fragmented; thus, freeing the rest of the numbers for packets that may be fragmented.

摘要翻译： 提供了当使用千兆以太网时用于维护IP报头的两字节识别字段的方法，系统和装置。 由于IP识别字段是两个字节，因此可以使用有限数量的标识号。 由于千兆以太网的高数据传输速率，这个数字可能会在一秒内达到。 因此，在运输中可能存在两个或更多个分组可能具有相同的识别号码的可能性。 在代表性实施例中，一个识别号用于可能不被分段的分组; 因此，释放剩余的数字可能碎片的数据包。

公开(公告)号：US20060262738A1

公开(公告)日：2006-11-23

申请号：US11130694

申请日：2005-05-17

申请人： Lilian Fernandes ,  Vinit Jain ,  Ketan Pancholi ,  Venkat Venkatsubra

发明人： Lilian Fernandes ,  Vinit Jain ,  Ketan Pancholi ,  Venkat Venkatsubra

IPC分类号： H04B1/44 ,  H04L12/56

CPC分类号： H04L69/16 ,  H04L1/1832 ,  H04L1/1854 ,  H04L47/10 ,  H04L47/27 ,  H04L47/283 ,  H04L47/323 ,  H04L69/163

摘要： Administering acknowledgment messages (‘ACKs’) in the Transmission Control Protocol (“TCP”) that include establishing a TCP connection between a sender and a receiver, receiving by the receiver one or more TCP messages from the sender, and transmitting an ACK from the receiver to the sender only when necessary. Administering ACKs in TCP according to embodiments of the present invention may include measuring by the sender a round trip time for transmission of TCP messages between the sender and the receiver and transmitting to the receiver a data-bearing TCP message once per round trip time. Establishing a TCP connection between a sender and a receiver may include measuring by the receiver a round trip time for transmission of TCP messages between the receiver and the sender. Transmitting an ACK only when necessary may include transmitting an ACK when no message from the sender arrives at the receiver for two round trip times.

摘要翻译： 管理传输控制协议（“TCP”）中的确认消息（“ACK”），包括建立发送方和接收方之间的TCP连接，由接收方接收来自发送方的一个或多个TCP消息，并从 收件人只有在必要时才发送给发件人。 根据本发明的实施例，在TCP中管理ACK可以包括由发送者测量在发送者和接收者之间传输TCP消息的往返时间，并且每次往返时间向接收者发送一次数据携带TCP消息。 建立发送方和接收方之间的TCP连接可能包括测量接收方和发送方之间传输TCP消息的往返时间。 仅在必要时发送ACK可以包括当来自发送者的消息到达接收机两次往返时间时发送ACK。

公开(公告)号：US07065581B2

公开(公告)日：2006-06-20

申请号：US09892968

申请日：2001-06-27

申请人： Herman Dietrich Dierks, Jr. ,  Vinit Jain ,  Agustin Mena, III ,  Venkat Venkatsubra

发明人： Herman Dietrich Dierks, Jr. ,  Vinit Jain ,  Agustin Mena, III ,  Venkat Venkatsubra

IPC分类号： G06F15/16 ,  G06F3/06

CPC分类号： H04L49/90

摘要： An apparatus and method for an improved bulk read socket call are provided. With the apparatus and method, a new field, so_rcvlen, is added to the socket structure that identifies the bulk read size requested by the user. The kernel of the prior art recv( ) function is also modified so that it sets the so_rcvlen to the size requested by the user prior to the recv( ) function going to sleep and waiting for the full data size requested by the user. A new flag, SP_MSGWAITALL, is also provided in the socket structure. In the TCP input processing, when data is received for a particular socket, the current setting of the SP_MSGWAITALL is checked. If the SP_MSGWAITALL flag is set, it is determined whether the amount of data stored in the socket receive buffer is less than the value of so_rcvlen. If not, the TCP input processing does not wake up the recv( ) thread. However, for every alternate segment, the TCP input processing sends back an acknowledgment (ACK). In the TCP output processing, when the SP_MSGWAITALL flag is set and the amount of data in the socket receive buffer is less than so_rcvlen, the full window is advertised. Once the TCP input processing determines that there is at least an amount of data in the socket receive buffer equal to the value of so_rcvlen, the TCP input processing will wake up the recv( ) thread and the SP_MSGWAITALL flag is reset.

摘要翻译： 提供了一种改进的大容量读取插座呼叫的装置和方法。 使用设备和方法，将新字段so_rcvlen添加到标识用户请求的批量读取大小的套接字结构中。 现有技术recv（）函数的内核也被修改，以便在recv（）函数进入休眠状态并等待用户请求的完整数据大小之前，将so_rcvlen设置为用户请求的大小。 插座结构中还提供了一个新标志SP_MSGWAITALL。 在TCP输入处理中，当为特定套接字收到数据时，将检查SP_MSGWAITALL的当前设置。 如果设置了SP_MSGWAITALL标志，则确定存储在套接字接收缓冲区中的数据量是否小于so_rcvlen的值。 如果没有，TCP输入处理不会唤醒recv（）线程。 然而，对于每个备用段，TCP输入处理发回确认（ACK）。 在TCP输出处理中，当设置了SP_MSGWAITALL标志并且套接字接收缓冲区中的数据量小于so_rcvlen时，将通告完整窗口。 一旦TCP输入处理确定套接字接收缓冲区中至少有一定数量的数据等于so_rcvlen的值，则TCP输入处理会唤醒recv（）线程，并重置SP_MSGWAITALL标志。