Packet classification using multi-dimensional splitting

    公开(公告)号:US10257090B2

    公开(公告)日:2019-04-09

    申请号:US15438317

    申请日:2017-02-21

    摘要: A computer-implemented method classifies a packet received from a network. The method comprises receiving the packet having header information and a rule-set that indicates a plurality of actions to perform on the packet. A geometric representation of the rule-set is constructed having a plurality of dimensions and rule boundaries. At least one rule boundary and dimension of the geometric representation is split to form a search structure having a plurality of nodes. Each node in the plurality of nodes is constrained to a predetermined memory threshold. The search structure is searched using the header information to obtain a rule in the rule-set that indicates an action to take on the packet.

    Networking service level agreements for computer datacenters

    公开(公告)号:US10735279B2

    公开(公告)日:2020-08-04

    申请号:US15488334

    申请日:2017-04-14

    摘要: A service level agreement (SLA) system is provided for multi-tenant cloud computing to enable provisioning of computing instances for tenants with a networking SLA. A networking SLA may specify a bandwidth requirement, for example, between instances to be scheduled for a tenant in a datacenter. The SLA system evaluates nodes in the datacenter based on a remaining capacity of links associated with nodes implementing the networking SLA. A modular system is provided to enable integration of the SLA system into existing cloud computing infrastructures. The cloud management system may query the SLA system to determine nodes in the datacenter capable of meeting a networking SLA. The list may be ordered to indicate nodes having better capacity to fulfill the networking SLA. The cloud management system can utilize the ordered list of nodes in scheduling instance deployments for tenant requests.

    Network verification system
    6.
    发明授权

    公开(公告)号:US10778545B2

    公开(公告)日:2020-09-15

    申请号:US16040017

    申请日:2018-07-19

    发明人: Yan Sun Wei Xu Kun Tan

    IPC分类号: H04L12/26 H04L12/24

    摘要: A network state is verified in real-time by precomputing a data structure representing header spaces of data packets in respective equivalence classes (ECs) in a first snapshot of the network and in a second snapshot of the network at a later time. The data structure includes respective bitmaps indicating for each data packet forwarding rule for each node in the network a forwarding action to be taken for each EC in the network. The respective bitmaps store bits indicating for each data packet forwarding rule whether the data packet forwarding rule applies to the respective ECs in the network. The snapshots are compared to identify network paths that have changed between a time of the first snapshot and the second snapshot of the network and network verification operations are performed on the changed network paths as a logical function of a bitmap corresponding to each changed network path.

    NETWORK VERIFICATION SYSTEM
    7.
    发明申请

    公开(公告)号:US20200028762A1

    公开(公告)日:2020-01-23

    申请号:US16040017

    申请日:2018-07-19

    发明人: Yan Sun Wei Xu Kun Tan

    IPC分类号: H04L12/26 H04L12/24

    摘要: A network state is verified in real-time by precomputing a data structure representing header spaces of data packets in respective equivalence classes (ECs) in a first snapshot of the network and in a second snapshot of the network at a later time. The data structure includes respective bitmaps indicating for each data packet forwarding rule for each node in the network a forwarding action to be taken for each EC in the network. The respective bitmaps store bits indicating for each data packet forwarding rule whether the data packet forwarding rule applies to the respective ECs in the network. The snapshots are compared to identify network paths that have changed between a time of the first snapshot and the second snapshot of the network and network verification operations are performed on the changed network paths as a logical function of a bitmap corresponding to each changed network path.

    SYSTEM AND METHOD FOR PACKET CLASSIFICATION USING MULTIPLE SECURITY DATABASES

    公开(公告)号:US20180091556A1

    公开(公告)日:2018-03-29

    申请号:US15280881

    申请日:2016-09-29

    IPC分类号: H04L29/06 G06F17/30

    摘要: A packet classification system is provided, including a first security database and a second security database for use in connection with packet classification in accordance with an Internet security protocol. The packet classification system further includes processing circuitry in communication with the first security database and the second security database, with the processing circuitry configured to identify at least one aspect of at least one packet received by the processing circuitry, select either the first security database or the second security database as a selected security database, based on the at least one aspect of the at least one packet, select at least one of a plurality of algorithms to classify the at least one packet, wherein the selection of the at least one algorithm is based on a criteria related to the at least one packet, and classify the at least one packet, utilizing the selected security database.

    Memory addressing mechanism using a buffer of a hierarchy of collision free hash tables
    10.
    发明授权
    Memory addressing mechanism using a buffer of a hierarchy of collision free hash tables 有权
    使用无冲突哈希表层次的缓冲区的内存寻址机制

    公开(公告)号:US09361238B2

    公开(公告)日:2016-06-07

    申请号:US14532874

    申请日:2014-11-04

    发明人: Yan Sun Norbert Egi

    摘要: Methods and apparatuses for insertion, searching, deletion, and load balancing using a hierarchical series of hash tables are described herein. The techniques disclosed provide nearly collision free or deterministic hash functions using a bitmap as a pre-filter. The hash functions have different priorities and one hashing result will be used to perform main memory access. For the hash functions, two hash bitmaps are used to store valid data and collision information. There is no collision allowed in the hash tables except for the hash table with the lowest priority. The hash tables and bitmaps may be stored in one or more caches in (e.g., a cache of a CPU, Block RAMs in FPGAs, etc.) which perform much faster than main memory.

    摘要翻译: 这里描述了使用分级序列的散列表来插入,搜索,删除和负载平衡的方法和装置。 所公开的技术使用位图作为前置滤波器提供了几乎无冲突的或确定性的散列函数。 哈希函数具有不同的优先级,一个散列结果将用于执行主内存访问。 对于哈希函数,使用两个哈希位图来存储有效的数据和冲突信息。 散列表中不存在冲突,除了优先级最低的哈希表外。 散列表和位图可以存储在执行比主存储器快得多的(例如,CPU的高速缓存,FPGA中的块RAM等)中的一个或多个高速缓存中。