公开(公告)号：US06760309B1

公开(公告)日：2004-07-06

申请号：US09535769

申请日：2000-03-28

申请人： Haim Rochberger ,  Gad Azriel ,  Sarit Shani Natanson

发明人： Haim Rochberger ,  Gad Azriel ,  Sarit Shani Natanson

IPC分类号： H04J314

CPC分类号： H04L65/1009 ,  H04L29/06027 ,  H04L47/10 ,  H04L47/11 ,  H04L47/2416 ,  H04L47/2441 ,  H04L47/2458 ,  H04L47/286 ,  H04L65/80

摘要： An apparatus for and a method of dynamically prioritizing packets over a packet based network. Packets are dynamically prioritized on the basis of their ‘time to live’ in the network as they travel from one network entity to another. Packets are assigned a priority in accordance with how ‘old’ or ‘young’ they are. Packets with a relatively long time left to live are assigned lower priority then those with relatively little time left to live. A time to live (TTL) field is added to the packet as it travels from one network entity to another. The contents of the time to live (TTL) field represents how ‘young’ or ‘old’ the packet is and conveys the time left before the packet is no longer of any use. Each network entity that receives the packet with a TTL field, subtracts from it the time the packet spends passing through that entity. The field decreases as it hops from network entity to entity until it reaches its destination or is discarded.

摘要翻译： 一种用于基于分组的网络动态地优先分组分组的装置和方法。 基于他们从一个网络实体到另一个网络实体的网络的“居住时间”，动态优先分组。 按照“旧”或“年轻”分配优先级。 具有相对较长时间的分组被分配到较低的优先级，那些相对较少的时间剩下的生活。 当数据包从一个网络实体传播到另一个网络实体时，生存时间（TTL）字段被添加到数据包中。 生活时间（TTL）字段的内容表示数据包的“年轻”或“旧”，并传达了数据包不再有任何用途之前的时间。 接收TTL字段的数据包的每个网络实体，在数据包通过该实体的时间内从其中减去。 该字段从网络实体跳到实体，直到到达目的地或被丢弃为止。

公开(公告)号：US06577653B1

公开(公告)日：2003-06-10

申请号：US09300884

申请日：1999-04-28

申请人： Haim Rochberger ,  Sarit Shani Natanson ,  Michael Gorokhovsky

发明人： Haim Rochberger ,  Sarit Shani Natanson ,  Michael Gorokhovsky

IPC分类号： H04J322

CPC分类号： H04L45/24 ,  H04L12/5601 ,  H04L2012/5624

摘要： An apparatus for and a method of establishing a connection between a source and destination node using multiple parallel paths. To establish a connection, bandwidth capacity on one or more parallel paths are aggregated and combined to form a route having a bandwidth larger than any single path. A technique similar to inverse multiplexing is used at the source end to split the cell stream into multiple parallel paths. At the destination end, the parallel multiple cell streams are inverse demultiplexed and the multiple cell streams combined into a single cell stream. The parallel routes are established by building parallel VCCs for the same call via the network. The ATM signaling process at the source and destination nodes is modified to accommodate multiple VCCs. In addition, the hardware at each source and destination node is configured to perform the inverse and inverse demultiplexing. Further, the control software at the source and destination nodes is modified to implement the call handling required of such a multiple VCC call, including setup, tear down, error handling, etc.

摘要翻译： 一种用于使用多个并行路径在源节点和目的地节点之间建立连接的装置和方法。 为了建立连接，在一个或多个并行路径上的带宽容量被聚合和组合以形成带宽大于任何单个路径的路由。 在源端使用类似于逆多路复用的技术将单元流分解成多个并行路径。 在目的地，并行多个单元流被反向解复用，并且多个单元流被组合成单个单元流。 并行路由通过构建用于通过网络的相同呼叫的并行VCC来建立。 源节点和目标节点上的ATM信令进程被修改以适应多个VCC。 此外，每个源节点和目的地节点处的硬件被配置为执行反向和反向解复用。 此外，修改源和目的地节点处的控制软件以实现这种多个VCC呼叫所需的呼叫处理，包括建立，拆除，错误处理等。

公开(公告)号：US06493345B1

公开(公告)日：2002-12-10

申请号：US09392096

申请日：1999-09-08

申请人： David Margulis ,  Meir Morgenstern ,  Haim Rochberger

发明人： David Margulis ,  Meir Morgenstern ,  Haim Rochberger

IPC分类号： H04L1228

CPC分类号： H04L12/1886 ,  H04L49/201 ,  H04L49/25 ,  H04L2012/5617 ,  H04L2012/5642

摘要： A system comprising a single sender SMS for forwarding multicast traffic is collocated in the same device as the LEC thus creating an optimal distribution path for multicast traffic. This reduces the latency of transmission of multicast traffic from sender to receivers by eliminating the need to transmit cells from the LEC to the SMS via the network and reassemble cells into frames and segment frames into cells. The SMS functions as a private SMS for exclusive use by the LEC alone. The SMS of the present invention enables a LEC to distribute multicast traffic in a very efficient manner. The LEC communicates with the SMS directly via the operating system internal messaging subsystem or equivalent. The multicast traffic is sent on a P2M connection from the sending LEC via the SMS to all the receiving LECs interested in listening to a given MMAC. In a first embodiment, the LEC internally has knowledge of the existence of the private SMS and is inhibited from issuing a LE_ARP_REQ to find the SMS. The LES is prevented from assigning an LEC to that SMS. In a second embodiment, address and LEC information is predefined for the SMS and propagated to all the SMSs. The LES is directed to permit only a single LEC to be assigned to a SMS preventing the LES from assigning a LEC other than the registered LEC to that SMS.

摘要翻译： 包括用于转发多播流量的单个发送者SMS的系统与LEC并置在同一设备中，从而为多播流量创建最佳分发路径。 这减少了从发送器到接收器的组播流量传输的延迟，消除了通过网络将小区从LEC发送到SMS，并将小区重新组合成帧和分段帧到小区中。 SMS作为私人短信功能，仅供LEC独家使用。 本发明的SMS使LEC能够以非常有效的方式分发多播业务。 LEC通过操作系统内部消息传递子系统或等效物体与SMS直接通信。 多播流量在P2M连接上从发送LEC经由SMS发送到所有接收到LEC，用于收听给定的MMAC。 在第一实施例中，LEC内部具有私人SMS的存在的知识，并且被禁止发出LE_ARP_REQ以找到SMS。 LES被阻止向该SMS分配LEC。 在第二实施例中，为SMS定义地址和LEC信息，并传播到所有的SMS。 LES被指示仅允许将单个LEC分配给SMS，以防止LES将除了注册的LEC之外的LEC分配给该SMS。

公开(公告)号：US06456600B1

公开(公告)日：2002-09-24

申请号：US09300882

申请日：1999-04-28

申请人： Haim Rochberger ,  Meir Morgenstern ,  David Margolis

发明人： Haim Rochberger ,  Meir Morgenstern ,  David Margolis

IPC分类号： H04L1228

CPC分类号： H04L45/06 ,  H04L45/02 ,  H04L45/026 ,  H04L45/04 ,  H04L45/46 ,  H04L2012/5621 ,  H04L2012/5623 ,  H04Q11/0478

摘要： A method of calculating a complex node representation for logical nodes in a hierarchical peer group in a PNNI based ATM network. The summary information for default spokes, exceptions and bypasses is determined in a way that closely represents the underlying metrics associated with the peer group. A list of all border nodes in the peer group is generated and maintained and a matrix of tables is generated, one table for each metric per each class of service. The table is populated by the best value associated with the corresponding metric for a particular pair of border nodes within a particular class of service. Once the table is populated, each element is placed in a group or bin. The bins are generated by dividing the full range of the metric into a plurality of smaller subgroups or bins. The bin having a count greater than a threshold count is selected and 80% of the worst value for that group is used as the default spoke. If no bin exists having at least a threshold count, the number of bins is reduced by one and a new set of subranges is calculated. The elements of the table are placed in the newly divided bins and a scan is performed once again.

摘要翻译： 一种基于PNNI的ATM网络中的分层对等体组中的逻辑节点的复数节点表示的计算方法。 以默认轮辐，异常和旁路的摘要信息的确定方式是密切地表示与对等组相关联的基础度量。 生成和维护对等体组中所有边界节点的列表，并生成一个表格矩阵，每个服务类别的每个度量标准都有一个表格。 该表由与特定服务类别内的特定边界节点对相应的度量相关联的最佳值填充。 一旦表格被填充，每个元素被放置在组或bin中。 通过将度量的全部范围划分成多个较小的子组或箱来生成箱。 选择具有大于阈值计数的计数器的存储区，并将该组的最坏值的80％用作默认辐条。 如果不存在具有至少阈值计数的箱，则将箱的数量减少1，并且计算一组新的子范围。 表格中的元素被放置在新分开的存储区中，并再次执行扫描。

公开(公告)号：US06272107B1

公开(公告)日：2001-08-07

申请号：US09076686

申请日：1998-05-12

申请人： Haim Rochberger ,  David Keynan

发明人： Haim Rochberger ,  David Keynan

IPC分类号： G01R3108

CPC分类号： H04L49/309 ,  H04L12/5601 ,  H04L2012/5627 ,  H04L2012/563 ,  H04L2012/5631

摘要： A method of path restoration in an ATM network which attempts to minimize the loss of cells in the event a node or a link fails. The method is suitable for failures in point to point Switched Virtual Circuits (SVCs) in ATM networks and makes use of the bi-directional properties of point to point SVC calls. The method permits the construction of an ATM network which includes call path redundancy whereby if a node or link fails, another path meeting the requirements of the call (if one exists) will be used on an automatic basis. The alternative or redundant path is placed into service in such a way that the flow of data is substantially constant. The switching of the path occurs with almost no data loss and without the overhead time associated with Q.SAAL based detection and signaling.

摘要翻译： 在ATM网络中的路径恢复方法，其尝试在节点或链路发生故障的情况下最小化小区的丢失。 该方法适用于ATM网络中点对点交换虚电路（SVC）的故障，并利用点对点SVC呼叫的双向属性。 该方法允许构建包括呼叫路径冗余的ATM网络，由此如果节点或链路出现故障，则将自动地使用满足呼叫要求的另一路径（如果存在）。 替代或冗余路径以数据流大致恒定的方式投入使用。 路径的切换几乎没有数据丢失，没有与基于Q.SAAL的检测和信令相关的开销时间。

公开(公告)号：US06212188B1

公开(公告)日：2001-04-03

申请号：US09071065

申请日：1998-05-01

申请人： Haim Rochberger ,  Alexander Or

发明人： Haim Rochberger ,  Alexander Or

IPC分类号： H04L1228

CPC分类号： H04L45/10 ,  H04L12/5602 ,  H04L45/26 ,  H04L45/34 ,  H04L47/122

摘要： A method of performing source routing when a node is in the overload state. A node in the overload state cannot perform routing functions in response to a SETUP message for destinations other than its directly connected neighbors. Instead, the node forwards the SETUP request to one of its directly connected neighbors. The neighboring node chosen must be able to support the overload routing feature, the link between the two nodes must be in the FULL state and the node chosen must not itself be in the overload state. The neighboring node chosen to calculate the source route receives a ROUTE REQUEST message encapsulating the information received from the source user. The neighboring node chosen then calculates the source route as if it was the node connected to the source user, i.e., the node in the overload state. The route, once calculated, is sent to the requesting node in the overload state.

摘要翻译： 当节点处于过载状态时执行源路由的方法。 处于过载状态的节点不能响应于除直接连接的邻居之外的目的地的SETUP消息执行路由功能。 相反，节点将SETUP请求转发到其直接连接的邻居之一。 选择的相邻节点必须能够支持过载路由功能，两个节点之间的链路必须处于FULL状态，并且所选择的节点本身不能处于过载状态。 选择用于计算源路由的相邻节点接收封装从源用户接收的信息的ROUTE REQUEST消息。 所选择的相邻节点然后计算源路由，就好像它是连接到源用户的节点，即处于过载状态的节点。 计算出的路由在过载状态下发送到请求节点。

公开(公告)号：US06178172B1

公开(公告)日：2001-01-23

申请号：US09047068

申请日：1998-03-24

申请人： Haim Rochberger

发明人： Haim Rochberger

IPC分类号： H04L1228

CPC分类号： H04L45/02 ,  H04L45/10

摘要： A method for performing topology database synchronization between two nodes wherein one of the nodes is in an overload state. The node in the overload state looks for a neighboring node directly connected to it that is not in the overload state. After such a node is found, the link to the neighboring node is intentionally dropped and subsequently re-established. This causes the neighboring node to transfer its entire knowledge of the network to the node in the overload state. The node in the overload state than functions to forward the topology database information to its other neighboring node. In this fashion, the portions of the network one either side of the link learn about the other side notwithstanding the fact that one of the nodes is in an overload state. Once the database synchronization process is complete, calls can then be initiated from either side of the network to the other side.

摘要翻译： 一种用于在其中一个节点处于过载状态的两个节点之间执行拓扑数据库同步的方法。 处于过载状态的节点查找直接连接到它的不处于过载状态的邻居节点。 在找到这样的节点之后，到相邻节点的链接被有意地丢弃并随后被重新建立。 这导致相邻节点将其整个网络的知识传递到处于过载状态的节点。 过载状态下的节点比拓扑数据库信息转发到其他邻居节点。 以这种方式，尽管有一个节点处于过载状态，链路的任一侧的网络的一部分了解另一侧。 一旦数据库同步过程完成，就可以从网络的任一侧向对方发起呼叫。

公开(公告)号：US6147971A

公开(公告)日：2000-11-14

申请号：US195342

申请日：1998-11-18

申请人： Haim Rochberger ,  Alexander Or

发明人： Haim Rochberger ,  Alexander Or

IPC分类号： H04L12/56 ,  H04L12/28

CPC分类号： H04L45/02 ,  H04L45/10

摘要： Two solutions are provided for optimizing the algorithm used by the local node in calculating the route to a destination node. A first method optimizes the route calculation by associating a port/link information with the PTSEs received by the node. The association is made using the first PTSE to arrive from an originating node. The routing algorithm is then steered to begin routing to a destination using the associated port/link. A second method optimizes the route calculation by utilizing a special PTSE termed a ROUTE.sub.-- PTSE into which is placed the route the PTSE takes upon being flooded from the creating node to the local node. When routing, the local node traverses the DTL already existing in the ROUTE.sub.-- PTSE received from a particular originating node that is the current destination node.

摘要翻译： 提供了两种解决方案，用于优化本地节点在计算到目标节点的路由时使用的算法。 第一种方法通过将端口/链路信息与节点接收的PTSE相关联来优化路由计算。 使用第一个PTSE从始发节点到达关联。 然后引导路由算法，开始使用关联的端口/链路路由到目的地。 第二种方法通过利用称为ROUTE-PTSE的特殊PTSE优化路由计算，其中将PTSE从创建节点淹没到路由中放置到本地节点。 当路由时，本地节点遍历从作为当前目的地节点的特定始发节点接收的ROUTE-PTSE中已经存在的DTL。