公开(公告)号：US20100290458A1

公开(公告)日：2010-11-18

申请号：US12772528

申请日：2010-05-03

申请人： Hamid Assarpour

发明人： Hamid Assarpour

IPC分类号： H04L12/56

CPC分类号： H04L45/00 ,  H04L12/437 ,  H04L45/245 ,  H04L45/28 ,  H04L49/557 ,  Y02D50/30

摘要： A method, apparatus and computer program product for providing fast reroute of a packet that may be forwarded on one of a plurality of Equal Cost Multi Path (ECMP) routes through a network is presented. A packet is received by a Forwarding Data Unit (FDU) in a data plane of a network element. The unicast packet is routed at L3, and ECMP is enabled for a next hop for the unicast packet. An ECMP route is selected for forwarding the packet to a destination port. A lookup is performed in a port state table maintained by the FDU to determine an available local port for said ECMP route that is in an UP state for the destination, and if no local port is UP, then a lookup is performed in the port state table to determine an available remote port that is in an UP state for the selected ECMP route.

摘要翻译： 提出了一种用于提供可以通过网络在多个等成本多路径（ECMP）路由之一上转发的分组的快速重新路由的方法，装置和计算机程序产品。 分组由网元的数据平面中的转发数据单元（FDU）接收。 单播数据包在L3路由，为单播数据包的下一跳使能ECMP。 选择ECMP路由将数据包转发到目标端口。 在由FDU维护的端口状态表中执行查找，以确定目的地处于UP状态的所述ECMP路由的可用本地端口，并且如果没有本地端口是UP，则在端口状态中执行查找 表以确定所选ECMP路由处于UP状态的可用远程端口。

公开(公告)号：US07751329B2

公开(公告)日：2010-07-06

申请号：US11906668

申请日：2007-10-03

申请人： Roger Lapuh ,  Mohnish Anumala ,  Robert Lariviere ,  Hamid Assarpour ,  Martin L. White

发明人： Roger Lapuh ,  Mohnish Anumala ,  Robert Lariviere ,  Hamid Assarpour ,  Martin L. White

IPC分类号： H04L12/28 ,  G01R31/08 ,  G06F11/00

CPC分类号： H04L49/552 ,  H04L49/351 ,  H04L49/555 ,  H04L49/70

摘要： In a communications network, a cluster switch is provided, where the cluster switch has plural individual switches. An abstraction layer is provided in the cluster switch, such that an interface having a set of ports is provided to upper layer logic in the cluster switch. The set of ports includes a collection of ports of the individual switches. Control traffic and data traffic are communicated over virtual tunnels between individual switches of the cluster switch, where each virtual tunnel has an active channel and at least one standby channel.

摘要翻译： 在通信网络中，提供集群交换机，其中集群交换机具有多个单独的交换机。 在集群交换机中提供抽象层，使得具有一组端口的接口被提供给集群交换机中的上层逻辑。 端口集包括各个交换机的端口集合。 在群集交换机的各个交换机之间的虚拟隧道上传送控制业务和数据业务，其中每个虚拟隧道具有活动信道和至少一个备用信道。

公开(公告)号：US20050249229A1

公开(公告)日：2005-11-10

申请号：US11069778

申请日：2005-03-01

申请人： Richard Wilkie ,  Michael Davis ,  Mike Popp ,  Hamid Assarpour

发明人： Richard Wilkie ,  Michael Davis ,  Mike Popp ,  Hamid Assarpour

IPC分类号： H04L12/28 ,  H04L12/46 ,  H04L12/56 ,  H04L12/66

CPC分类号： H04L49/351 ,  H04L45/60 ,  H04L49/30 ,  H04L2012/5665

摘要： The present invention provides a dynamically scalable edge router, wherein multiple data cards are connected to multiple input/output (I/O) cards via a dynamically configurable cross point. The cross point can be reconfigured to establish different connections between the data cards and the I/O cards, as well as change or establish new connections when additional data cards or I/O cards are added to the edge router. Accordingly, the cross point cooperates with the I/O cards to establish specific connections to the various data cards. The data cards may be connected to each other through a packet fabric, wherein the I/O cards, the cross point, the data cards, and the packet fabric form a significant portion of the forwarding plane of the edge router. The control plane is substantially provided by control cards, which are connected to the data cards through an internal network, which may be a switched network.

摘要翻译： 本发明提供了一种动态可扩展的边缘路由器，其中多个数据卡经由可动态配置的交叉点连接到多个输入/输出（I / O）卡。 可以重新配置交叉点，以在数据卡和I / O卡之间建立不同的连接，以及在向边缘路由器添加附加数据卡或I / O卡时更改或建立新连接。 因此，交叉点与I / O卡协作以建立到各种数据卡的特定连接。 数据卡可以通过分组结构彼此连接，其中I / O卡，交叉点，数据卡和分组结构形成边缘路由器的转发平面的重要部分。 控制平面基本由控制卡提供，控制卡通过可能是交换网络的内部网络连接到数据卡。

公开(公告)号：US09451056B2

公开(公告)日：2016-09-20

申请号：US13539276

申请日：2012-06-29

申请人： Hamid Assarpour ,  Marten Terpstra

发明人： Hamid Assarpour ,  Marten Terpstra

IPC分类号： H04L12/28 ,  H04L29/06 ,  H04L29/08 ,  H04L12/26

CPC分类号： H04L69/22 ,  H04L43/026 ,  H04L69/321 ,  H04L69/324

摘要： An operating system adds an application signature as a tag in a packet header. In one embodiment the tag is inserted as a Q-tag in an Ethernet header. When a network element receives the tagged packet, it uses the tag alone or in combination with one or more additional header fields to map the packet to a network virtualization identifier segregating the application traffic on the network. Services are applied to packets according to network virtualization identifier to enable distributed application of services without requiring network elements to maintain state associated with packet flows.

摘要翻译： 操作系统将应用签名作为标签添加到数据包头中。 在一个实施例中，标签作为Q标签插入到以太网报头中。 当网络元件接收到标记的分组时，它单独使用标签或与一个或多个附加报头字段组合来将分组映射到分离网络上的应用流量的网络虚拟化标识符。 根据网络虚拟化标识符将服务应用于数据包，以实现服务的分布式应用，而不需要网络元件来维护与数据包流相关联的状态。

公开(公告)号：US09164771B2

公开(公告)日：2015-10-20

申请号：US13539223

申请日：2012-06-29

申请人： Hamid Assarpour

发明人： Hamid Assarpour

IPC分类号： G06F9/38 ,  G06F9/46 ,  G06F9/30

CPC分类号： G06F9/38 ,  G06F9/30076 ,  G06F9/3838 ,  G06F9/3851 ,  G06F9/3877 ,  G06F9/46

摘要： Key lookup operations are broken into two instructions: a Key Dispatch Instruction (KDI), and a Return Result Instruction (RRI). The thread uses KDI to dispatch key information to a selected coprocessor to initiate a key lookup operation. Upon dispatch of the key value to the coprocessor, the KDI is retired to enable the thread to continue to dispatch and retire addition instructions in the pipeline and does not go idle. Subsequently, the thread will issue a RRI to obtain the key lookup result from the coprocessor. While a thread is executing, it maintains, as part of its context, a busy flag per coprocessor in a scoreboard register and a return result register per coprocessor. KDI causes the corresponding busy flag in the scoreboard register to be set. When the key lookup operation is complete, the busy flag is cleared and the result is stored in the return result register.

摘要翻译： 密钥查找操作分为两个指令：密钥分派指令（KDI）和返回结果指令（RRI）。 该线程使用KDI将密钥信息分配到所选择的协处理器以启动密钥查找操作。 在将密钥值发送到协处理器时，KDI退休以使线程能够继续调度并且删除流水线中的加法指令，并且不会空闲。 随后，线程将发出RRI以从协处理器获取密钥查找结果。 当线程正在执行时，它作为其上下文的一部分维护每个协调器在记分板寄存器中的忙标志和每个协处理器的返回结果寄存器。 KDI会使记分板寄存器中的相应忙标志置位。 当键查找操作完成时，忙标志被清除，结果存储在返回结果寄存器中。

公开(公告)号：US09152423B2

公开(公告)日：2015-10-06

申请号：US13071730

申请日：2011-03-25

申请人： Hamid Assarpour ,  Mike Craren ,  Rich Modelski

发明人： Hamid Assarpour ,  Mike Craren ,  Rich Modelski

IPC分类号： G06F15/00 ,  G06F7/38 ,  G06F9/00 ,  G06F9/44 ,  G06F9/32 ,  G06F9/30

CPC分类号： G06F9/325 ,  G06F9/30018 ,  G06F9/30061 ,  G06F9/30072 ,  G06F9/3013 ,  G06F9/327

摘要： A method, apparatus and computer program product for performing efficient loop instruction execution using bit vector scanning is presented. A bit vector is scanned, each bit in the bit vector representing at least one of a feature and a conditional status. The presence of a bit of said bit vector set to a first state is detected. The bit is set to a second state. An instruction address for a routine corresponding to said bit set to a first state is looked up using a bit position of said bit that was set to a first state. The routine is executed. The scanning, said detecting, said setting and said using are repeated until there are no remaining bits of said bit vector set to said first state.

摘要翻译： 提出了一种使用位向量扫描执行高效循环指令执行的方法，装置和计算机程序产品。 扫描位向量，位向量中的每个位表示特征和条件状态中的至少一个。 检测到设置为第一状态的位向量的位的存在。 该位被设置为第二个状态。 使用被设置为第一状态的所述比特的比特位置来查找与设置为第一状态的所述比特相对应的例程的指令地址。 执行例程。 重复扫描，所述检测，所述设置和所述使用，直到所述位向量的剩余位被设置为所述第一状态为止。

公开(公告)号：US08804507B2

公开(公告)日：2014-08-12

申请号：US13077577

申请日：2011-03-31

申请人： Hamid Assarpour

发明人： Hamid Assarpour

IPC分类号： G08C15/00 ,  H04L12/28 ,  G06F15/16 ,  G06F15/173 ,  G06F3/00 ,  G06F9/34 ,  H04L12/56 ,  H04L12/54 ,  G06F17/30 ,  G06F5/06

CPC分类号： H04L47/10 ,  G06F5/06 ,  G06F17/30949 ,  H04L47/39 ,  H04L47/6215 ,  H04L47/70 ,  H04L49/90

摘要： A method, apparatus and computer program product for temporal-based flow distribution across multiple packet processors is presented. A packet is received and a hash identifier (ID) is computed for the packet. The hash ID is used to index into a State Table and to retrieve a corresponding record. When a time credit field of the record is zero then the time credit field is set to a to a new value; a Packet Processing Engine (PE) whose First-In-First-Out buffer (FIFO) has the lowest fill level is selected; and a PE number field in the state table record is updated with the selected PE number. When the time credit field of the record is non-zero then the packet is sent to a PE based on the value stored in the record; and the time credit field in the record is decremented if the time credit field is greater than zero.

摘要翻译： 提出了一种用于跨多个分组处理器的基于时间的流分布的方法，装置和计算机程序产品。 接收到分组，并为分组计算散列标识符（ID）。 散列ID用于索引到状态表并检索相应的记录。 当记录的时间信用字段为零时，信用字段的时间设置为新值; 选择其先进先出缓冲器（FIFO）具有最低填充电平的分组处理引擎（PE）; 并且用所选择的PE号更新状态表记录中的PE号字段。 当记录的时间信用字段不为零时，根据记录中存储的值将数据包发送到PE; 如果时间信用字段大于零，则记录中的时间信用字段递减。

公开(公告)号：US20140003434A1

公开(公告)日：2014-01-02

申请号：US13539276

申请日：2012-06-29

申请人： Hamid Assarpour ,  Marten Terpstra

发明人： Hamid Assarpour ,  Marten Terpstra

IPC分类号： H04L12/56

CPC分类号： H04L69/22 ,  H04L43/026 ,  H04L69/321 ,  H04L69/324

摘要： An operating system adds an application signature as a tag in a packet header. In one embodiment the tag is inserted as a Q-tag in an Ethernet header. When a network element receives the tagged packet, it uses the tag alone or in combination with one or more additional header fields to map the packet to a network virtualization identifier segregating the application traffic on the network. Services are applied to packets according to network virtualization identifier to enable distributed application of services without requiring network elements to maintain state associated with packet flows.

摘要翻译： 操作系统将应用签名作为标签添加到数据包头中。 在一个实施例中，标签作为Q标签插入到以太网报头中。 当网络元件接收到标记的分组时，它单独使用标签或与一个或多个附加报头字段组合来将分组映射到分离网络上的应用流量的网络虚拟化标识符。 根据网络虚拟化标识符将服务应用于数据包，以实现服务的分布式应用，而不需要网络元件来维护与数据包流相关联的状态。

公开(公告)号：US20130250762A1

公开(公告)日：2013-09-26

申请号：US13426902

申请日：2012-03-22

申请人： Hamid Assarpour

发明人： Hamid Assarpour

IPC分类号： H04L12/24 ,  H04L12/26

CPC分类号： H04L49/506 ,  H04L47/266 ,  H04L47/30 ,  H04L49/103 ,  H04L49/9047

摘要： Packets are colored and stored in a shared packet buffer without assigning fixed page allocations per port. The packet buffer is divided into three areas—an unrestricted area, an enforced area, and a headroom area. Regardless of the fullness level, when a packet is received it will be stored in the packet buffer. If the fullness level is in the unrestricted area, no flow control messages are generated. If the fullness level is in the enforced region, a probabilistic flow control generation process is used determine if a flow control messages will be generated. If the fullness level is in the headroom area, flow control is automatically generated. Quanta timers are used to control regeneration of flow control messages.

摘要翻译： 数据包被着色并存储在共享数据包缓冲区中，而不会为每个端口分配固定的页面分配。 分组缓冲区分为三个区域 - 无限制区域，强制区域和净空区域。 不管丰满程度如何，当收到数据包时，它将被存储在数据包缓冲区中。 如果饱和度级别在非限制区域内，则不会生成流量控制消息。 如果饱和度级别在强制区域中，则使用概率流控制生成过程来确定是否将生成流控制消息。 如果饱和度处于净空区域，则自动生成流量控制。 Quanta定时器用于控制流量控制消息的再生。

公开(公告)号：US08489849B2

公开(公告)日：2013-07-16

申请号：US12953952

申请日：2010-11-24

申请人： Hamid Assarpour ,  Andrew Hull

发明人： Hamid Assarpour ,  Andrew Hull

IPC分类号： G06F12/00

CPC分类号： H04L12/4625 ,  G06F9/32 ,  G06F9/322 ,  H04L45/54

摘要： A method and apparatus for providing TCAM functionality in a custom integrated circuit (IC) is presented. An incoming key is broken into a predefined number of sub-keys. Each sub-key is sued to address a Random Access Memory (RAM), one RAM for each sub-key. An output of the RAM is collected for each sub-key, each output comprising a Partial Match Vector (PMV). The PMVs are bitwise ANDed to obtain a value which is provided to a priority encoder to obtain an index. The index is used to access a result RAM to return a result value for the key.

摘要翻译： 提出了一种在定制集成电路（IC）中提供TCAM功能的方法和装置。 传入的密钥被分成预定义数量的子密钥。 每个子密钥都被要求寻址随机存取存储器（RAM），每个子密钥都有一个RAM。 为每个子键收集RAM的输出，每个输出包括部分匹配向量（PMV）。 PMV被按位AND以获得提供给优先编码器以获得索引的值。 该索引用于访问结果RAM以返回键的结果值。