-
公开(公告)号：US06594268B1
公开(公告)日：2003-07-15
申请号：US09266622
申请日：1999-03-11
IPC分类号： H04L1228
CPC分类号： H04L47/10 , H04L45/04 , H04L45/302 , H04L47/2408
摘要： A packet network employs routers that determine network routing based on quality of service (QoS) provisioning parameters and network topology information. QoS provisioning parameters are provided to each router from a network management database, and the network topology information is determined from a link state database of the router. The link state database may include network topology information collected by the router in accordance with the open shortest path protocol (OSPF). A network link, router, or other node failure initiates a new path-selection process. First, a temporary set of provisioning entries may be determined with a shortest path first (SPF) routing method. Then, the network packet flows may be classified into packet flows, real-time and non-real-time, and then as packet flows that require reserved bandwidth or that may be multiplexed. A multicommodity flow (MCF) routing method is then employed to determine an optimized set of candidate provisioning entries for the packet flows that may be multiplexed. The MCF routing method determines new routing for the packet flows based on QoS provisioning commitments as parameters. The MCF routing method determines the new routing based on an optimization criterion, such as maximized revenue. Once the new routing is determined, routing of network traffic is enabled by converting the provisioning entries into filter rules, which are then loaded into the packet classifier of the router.
摘要翻译： 分组网络使用基于服务质量（QoS）提供参数和网络拓扑信息来确定网络路由的路由器。 从网络管理数据库向每个路由器提供QoS配置参数，并从路由器的链路状态数据库确定网络拓扑信息。 链路状态数据库可以包括根据开放最短路径协议（OSPF）由路由器收集的网络拓扑信息。 网络链路，路由器或其他节点故障启动新的路径选择过程。 首先，可以用最短路径优先（SPF）路由方法来确定一组临时配置条目。 然后，网络分组流可以被分类为分组流，实时和非实时，然后作为需要预留带宽或可以被多路复用的分组流。 然后使用多业务流（MCF）路由方法来确定可以被复用的分组流的优化的候选供应条目集合。 MCF路由方法基于QoS配置承诺来确定分组流的新路由作为参数。 MCF路由方法基于优化标准（如最大收入）来确定新路由。 一旦确定了新路由，就可以通过将配置条目转换为过滤规则来启用网络通信的路由，然后将其过滤到规则中，然后将其加载到路由器的分组分类器中。
-
2.发明授权
公开(公告)号：US6092115A
公开(公告)日：2000-07-18
申请号：US961122
申请日：1997-10-30
CPC分类号： H04L47/522 , H04L12/5693 , H04L47/52 , H04L47/525 , H04L47/621 , H04L47/623 , H04L47/6255 , H04L49/90 , H04L49/9036 , H04L49/9084 , H04L69/163 , H04L47/10 , H04L69/16
摘要： A per-flow queuing method and apparatus for IP networks carrying traffic from feedback controlled TCP connections enables flow of information packets from one or more sources to a destination through a link and comprises a buffer of predetermined size partitioned into a plurality of queues, each queue being allocated an occupancy b.sub.i for receiving and temporarily storing packets of information; a scheduler for removing packets from each buffer according to a predetermined rate and transmitting the packets over a network; and a control device for determining availablilty of queues in the buffer capable of receiving the packet and inputting the packet into a queue if the queue is available, the control device further selecting a queue and releasing a packet from the selected queue to accommodate input of the received packet when the queue is not available. Increased fairness and packet throughput through the link is achieved when the queue for dropping a packet is selected in accordance with a longest queue first or random drop scheme and, when a drop from front strategy for ACK packets is employed.
摘要翻译： 用于从反馈控制的TCP连接承载流量的IP网络的每流排队方法和装置使得能够通过链路从一个或多个源到目的地的信息分组流动，并且包括划分成多个队列的预定大小的缓冲器，每个队列 被分配用于接收和临时存储信息包的占用; 调度器，用于根据预定速率从每个缓冲器去除分组，并通过网络发送分组; 以及控制装置，用于确定能够接收分组的缓冲器中的队列的可用性，并且如果队列可用则将分组输入到队列中，则控制装置进一步选择队列并从所选择的队列中释放分组以适应 当队列不可用时收到数据包。 当根据最长队列第一或随机丢弃方案选择丢弃分组的队列时，并且当采用来自用于ACK分组的前策略的丢弃时，通过链路增加公平性和分组吞吐量。
-
公开(公告)号：US06289013B1
公开(公告)日：2001-09-11
申请号：US09145433
申请日：1998-09-02
IPC分类号： H04L900
CPC分类号： H04L49/602 , H04L49/205
摘要： A packet filter method and apparatus for a router employs an algorithm that decomposes a set of n filter rules of a k-dimensional space into sets of rule segments associated with non-overlapping intervals in each dimension. Such packet filter may be employed for layer four switching applications. Bit-parallel processing may be employed to compare each interval with corresponding fields of a packet received by the router. Bitmaps defined by the sets of rule segments, and so related to the corresponding filter rules are associated with the intervals. The interval bitmaps are combined to form a filter rule bitmap that identifies and associates one or more filter rules with the packet. For a case storing complete bitmaps for all intervals, the packet filter employs k*n2+O(n) bits of memory for each dimension, [log(2n)]+1 comparisons per dimension which may be performed in parallel, and [n/w] memory accesses for a pairwise combining operation, where w is a width of a bitmap used to identify the filter rule. Incremental memory read operations are employed to reduce memory space requirements of this packet filter case, allowing the packet-filter operation to be optimized in accordance with time complexity and memory space. Since a dominant contributing factor of execution time is off-chip memory accesses, availability of on-chip memory and the use of modified bitmap storage using interval bitmap pointers for incremental memory read operations significantly increases the number of filter rules that may be searched and applied within a given time constraint. For this algorithm employing incremental memory read operations, memory requirements may be reduced to O(n log n) bits while increasing the execution time by only a constant value, when log n≦w.
摘要翻译： 用于路由器的分组过滤方法和装置采用将k维空间的一组n个滤波器规则分解为与每个维度中的非重叠间隔相关联的规则分段的集合的算法。 这种分组过滤器可以用于第四层交换应用。 可以采用位并行处理来将每个间隔与由路由器接收的分组的相应字段进行比较。 由规则段集合定义的位图以及与相应的过滤规则相关的位图与间隔相关联。 间隔位图被组合以形成一个过滤器规则位图，用于标识和关联一个或多个过滤器规则与数据包。 对于存储所有间隔的完整位图的情况，分组过滤器对于可以并行执行的每个维度的每个维度[log（2n）] + 1比较采用k * n2 + O（n）个比特的存储器，并且[n / w]成对组合操作的存储器访问，其中w是用于标识过滤规则的位图的宽度。 采用增量存储器读取操作来减少该分组过滤器情况的存储空间需求，允许根据时间复杂度和存储器空间优化分组过滤器操作。 由于执行时间的主要贡献因素是片外存储器访问，片上存储器的可用性以及使用间隔位图指针进行增量存储器读取操作的修改位图存储的使用显着增加了可以搜索和应用的过滤器规则的数量 在给定的时间限制内。 对于采用增量存储器读取操作的该算法，当log n <= w时，存储器需求可以减少到O（nlog n）位，同时将执行时间仅增加一个恒定值。
-
公开(公告)号：US09619292B2
公开(公告)日：2017-04-11
申请号：US13459430
申请日：2012-04-30
申请人： Murali Kodialam , Fang Hao , T. V. Lakshman , Sarit Mukherjee
发明人： Murali Kodialam , Fang Hao , T. V. Lakshman , Sarit Mukherjee
IPC分类号： G06F15/173 , G06F9/50
CPC分类号： G06F9/5072 , G06F2209/502 , G06F2209/506
摘要： Various exemplary embodiments relate to a method and related network node including one or more of the following: receiving, at the cloud controller, a request message including a first request for a first cloud resource; identifying a set of potential devices for providing the first cloud resource; calculating a plurality of weight values corresponding to at least a portion of the set of potential devices for providing the first cloud resource, wherein the plurality of weight values are calculated based on a plurality of delta values associated with the at least a portion of the set of potential devices; selecting a device based on the plurality of weight values to provide the first cloud resource; and updating the delta value associated with the selected device.
-
5.发明授权Apparatus and method for matching offers and requests for sharing of resources 有权用于匹配资源和资源共享的请求的装置和方法公开(公告)号：US09535748B2
公开(公告)日：2017-01-03
申请号：US13370443
申请日：2012-02-10
IPC分类号： G06F15/173 , G06F9/50
CPC分类号： G06F9/5005 , G06F9/50 , H04L67/10
摘要： A resource assignment capability is presented. A resource specification associated with a plurality of elements is received. The resource specification includes, for each of the elements, a resource request including an indication of a quantity of resources requested by the element and a resource offer including an indication of a quantity of resources offered by the element for use by one or more other elements. A resource assignment, including an indication of an association between the resources requests and the resource offers, is determined using a resource assignment process. The resource assignment process may be a greedy assignment process or a maximum flow resource assignment process. The maximum flow resource assignment process includes constructing a maximum flow resource graph based on the one or more resource specifications and applying a maximum flow process to the maximum flow resource graph to determine thereby the resource assignment.
摘要翻译： 提出资源分配能力。 接收与多个元素相关联的资源规范。 所述资源规范对于每个元素包括资源请求，其包括由所述元素请求的资源的数量的指示和资源提供，所述资源提供包括由所述元素提供的资源的数量的指示以供一个或多个其他元素使用 。 使用资源分配处理来确定包括资源请求与资源提供之间的关联的指示的资源分配。 资源分配过程可以是贪心分配过程或最大流资源分配过程。 最大流资源分配过程包括基于一个或多个资源规范构建最大流资源图，并将最大流程应用于最大流资源图，从而确定资源分配。
-
公开(公告)号：US09306840B2
公开(公告)日：2016-04-05
申请号：US13627003
申请日：2012-09-26
IPC分类号： G06F15/173 , H04L12/717 , H04L29/06 , H04L12/715 , H04L12/743
CPC分类号： H04L47/80 , H04L12/4641 , H04L45/42 , H04L45/50 , H04L45/64 , H04L45/7457 , H04L63/1458 , H04L63/20
摘要： A flow deflection capability is provided for deflecting data flows within a Software Defined Network (SDN) in order to provide security for the SDN. A flow forwarding rule is generated for a first network element of the SDN based on detection of a condition (e.g., TCAM utilization condition, CPU utilization condition, or the like) associated with the first network element. The flow forwarding rule is generated by a control element of the SDN or the first network element of the SDN. The flow forwarding rule is indicative that at least a portion of new flow requests received at the first network element are to be forwarded from the first network element to a second network element of the SDN. The flow forwarding rule may specify full flow deflection or selective flow deflection.
摘要翻译： 提供流量偏转能力用于偏转软件定义网络（SDN）内的数据流，以便为SDN提供安全性。 基于与第一网元相关联的条件（例如，TCAM利用条件，CPU利用条件等）的检测，为SDN的第一网元生成流转发规则。 流转发规则由SDN的控制元素或SDN的第一网元生成。 流转发规则指示在第一网元接收到的新流请求的至少一部分将从第一网元转发到SDN的第二网元。 流量转移规则可以指定全流量偏转或选择性流量偏转。
-
7.发明授权Routing with service level guarantees between ingress-egress points in a packet network 有权在分组网络中的入口到出口点之间进行业务级保证的路由公开(公告)号：US06584071B1
公开(公告)日：2003-06-24
申请号：US09366620
申请日：1999-08-03
IPC分类号： H04L1256
CPC分类号： H04L45/50 , H04L45/02 , H04L45/10 , H04L45/12 , H04L45/122 , H04L45/124 , H04L45/302 , H04L45/308
摘要： A packet network of interconnected nodes employs a method of routing with service level guarantees to determine a path through the network for a requested label-switched path (LSP). Each of the nodes includes one or more routers that forward packets based on a forwarding table constructed from paths determined in accordance with the method of routing with service level guarantees. The method of routing with service level guarantees determines the path of the requested LSP based on the effect that routing those packets of the requested LSP may have on current and/or future demands on the capacity of network nodes for currently provisioned LSPs. Such method of routing with service level guarantees may not necessarily route packets of a requested LSP along the shortest path, or minimum number of hops, through the network. Given the packet network and LSP request, a linear programming system may be defined by a set of linear programming equations for a non-split demand case. The linear programming system is based on the network topology, the values of the ingress-egress point pair o and t and demand bd of the LSP request, and the total maxflow values of the existing ingress-egress point pair for currently provisioned LSPs. To estimate the solution for the linear programming system, a subnetwork is formed using link weights and links removed that cannot support the requested demand. Link weights are calculated based on the critical links of a pseudo-network in which increased maximum flow along existing paths between ingress-egress point pairs is maintained. A shortest path routing algorithm may then be employed to generate a path, if available, for the LSP request using the subnetwork with the calculated link weights.
摘要翻译： 互连节点的分组网络采用使用服务级别保证进行路由的方法，以针对所请求的标签交换路径（LSP）确定通过网络的路径。 每个节点包括一个或多个路由器，其基于由根据具有服务级别保证的路由方法确定的路径构建的转发表来转发分组。 基于服务级保证路由的方法基于所请求的LSP的这些分组的路由可能对当前和/或将来对当前配置的LSP的网络节点的容量的需求的影响来确定所请求的LSP的路径。 这种使用服务级别保证的路由方法可能不一定通过网络沿着最短路径或最小跳数来路由所请求的LSP的分组。 给定分组网络和LSP请求，可以通过用于非分裂需求情况的一组线性规划方程来定义线性规划系统。 线性规划系统基于网络拓扑，入口出口点对o和t的值以及LSP请求的需求bd以及当前配置的LSP的现有入口出口对的总最大流量值。 为了估计线性规划系统的解决方案，使用不能支持请求的需求的链路权重和链路去除形成子网络。 基于伪网络的关键链路来计算链路权重，其中保持了入口到出站对之间沿现有路径的最大流量增加。 然后可以采用最短路径路由算法来生成使用具有计算的链路权重的子网络的LSP请求的路径（如果可用）。
-
8.发明授权公开(公告)号：US06538991B1
公开(公告)日：2003-03-25
申请号：US09366619
申请日：1999-08-03
IPC分类号： H04L1256
CPC分类号： H04L47/825 , H04L41/5003 , H04L41/5025 , H04L45/00 , H04L45/14 , H04L45/302 , H04L45/308 , H04L45/50 , H04L47/15 , H04L47/70 , H04L47/724 , H04L47/781 , H04L47/783 , H04L47/822 , H04L47/823 , H04L47/829
摘要： A packet network of interconnected nodes employs a constraint-based routing method to determine a path through the network for a requested label-switched path (LSP). Each of the nodes includes one or more routers that forward packets based on a forwarding table constructed from paths determined in accordance with the constraint-based routing method. The constraint-based method determines the path of the requested LSP based on the effect that routing those packets of the requested LSP may have on current and/or future demands on the capacity of network nodes for currently provisioned LSPs. Such constraint-based routing method may not necessarily route packets of a requested LSP along the shortest path, or minimum number of hops, through the network. Given the packet network and LSP request, a linear programming system is defined by a set of linear programming equations. The linear programming system is based on the network topology, the values of the ingress-egress point pair o and t and demand bd of the LSP request, and the total maxflow values of the existing ingress-egress point pair for currently provisioned LSPs. The solution is estimated for a linear programming system of either split demand, non-split demand, or batch demand implementations for routing packets of the LSP. The constraint-based routing method may solve the linear programming system using common linear programming techniques.
摘要翻译： 互连节点的分组网络采用基于约束的路由方法来确定针对所请求的标签交换路径（LSP）的通过网络的路径。 每个节点包括一个或多个路由器，其基于由根据基于约束的路由方法确定的路径构建的转发表来转发分组。 基于约束的方法基于所请求的LSP的这些分组的路由可能对当前和/或将来对当前配置的LSP的网络节点的容量的需求的影响来确定所请求的LSP的路径。 这种基于约束的路由方法可能不一定通过网络沿着最短路径或最小跳数来路由请求的LSP的分组。 给定分组网络和LSP请求，线性规划系统由一组线性规划方程定义。 线性规划系统基于网络拓扑，入口出口点对o和t的值以及LSP请求的需求bd以及当前配置的LSP的现有入口出口对的总最大流量值。 该解决方案针对用于路由LSP的分组的分组需求，非分裂需求或批量需求实现的线性规划系统进行估计。 基于约束的路由方法可以使用公共线性规划技术来解决线性规划系统。
-
公开(公告)号：US20210092132A1
公开(公告)日：2021-03-25
申请号：US16578535
申请日：2019-09-23
申请人： Randeep Bhatia , Bhawna Gupta , Steven Benno , Jairo Esteban , T. V. Lakshman
发明人： Randeep Bhatia , Bhawna Gupta , Steven Benno , Jairo Esteban , T. V. Lakshman
摘要： Network-based, unsupervised classifiers are provided. The classifiers identify both known and unknown attacks aimed at industrial networks without the need to have a priori knowledge of known malicious attack patterns.
-
公开(公告)号：US10374939B2
公开(公告)日：2019-08-06
申请号：US15605790
申请日：2017-05-25
申请人： Fang Hao , Murali Kodialam , T. V. Lakshman
发明人： Fang Hao , Murali Kodialam , T. V. Lakshman
IPC分类号： H04L12/26 , H04L12/721 , H04L12/733
摘要： A method of determining a maximum flow on a network path using segment routing, the method including establishing a segment graph, establishing underlying dual weights on the segment graph, computing the dual weights from the segment graph, finding a minimum dual weight path not having more than a predetermined number of hops, augmenting a flow on the dual weight path, and updating the dual weights on the underlying segment graph.
-
-
-
-
-
-
-
-
-
搜索结果
18 条记录 (耗时 0.022 秒)
