摘要:
A controller may fulfill hardware address requests that are sent by source end hosts in a network to discover hardware addresses of destination end hosts. The controller may use network topology information to determine how to process the hardware address requests. The controller may retrieve a requested hardware address from a database of end hosts. If the controller is able to retrieve the hardware address of a destination end host from the database of end hosts, the controller may provide the source end host with a reply packet that contains the requested hardware address. If the controller is unable to retrieve the requested hardware address, the controller may form request packets to discover the address of the second end host and/or to discover a packet forwarding path between the source end host and the destination end host.
摘要:
Network packets may be transmitted from packet sources to packet destinations through a network of switches. The switches may have corresponding flow tables that control how the packets are forwarded through the switches. A controller server may generate network switch forwarding paths for the network packets by modifying the flow tables with entries based on attributes of the network packets and network topology information. The controller server may forward selected packets directly to packet destinations instead of generating the network switch forwarding paths. To determine which packets to directly forward, the controller server may calculate cost metrics associated with the network switch forwarding paths and associated with forwarding network packets directly to packet destinations. The cost metrics may be calculated based on network topology information, network traffic conditions, controller server processing load, packet forwarding delays, an expected number of future network packets, or attributes of the network packets.
摘要:
A controller may help reduce network traffic that is associated with broadcasting of Dynamic Host Configuration Protocol (DHCP) packets by converting broadcast DHCP packets into unicast DHCP packets and forwarding the unicast DHCP packets to appropriate DHCP servers. The servers may be identified from a database of servers that is updated with DHCP server address information based on DHCP reply packets that are received by the controller from servers in the network. To convert DHCP request packets into unicast packets, the controller may modify address header fields of the packets such as Ethernet addresses and Internet Protocol (IP) addresses. The controller may forward the modified DHCP request packets to the server by providing packet forwarding rules such as flow table entries to the switches or by forwarding the modified DHCP request packets through the controller.
摘要:
A controller may help reduce network traffic that is associated with broadcasting of Dynamic Host Configuration Protocol (DHCP) packets by converting broadcast DHCP packets into unicast DHCP packets and forwarding the unicast DHCP packets to appropriate DHCP servers. The servers may be identified from a database of servers that is updated with DHCP server address information based on DHCP reply packets that are received by the controller from servers in the network. To convert DHCP request packets into unicast packets, the controller may modify address header fields of the packets such as Ethernet addresses and Internet Protocol (IP) addresses. The controller may forward the modified DHCP request packets to the server by providing packet forwarding rules such as flow table entries to the switches or by forwarding the modified DHCP request packets through the controller.
摘要:
A controller may be used to control client switches in a network that includes non-client switches. The controller may form client domains from groups of client switches that are separated by intervening non-client domains formed from non-client switches. The controller may determine a network domain topology from the client domains and non-client domains. The controller may determine a spanning tree that interconnects the nodes of the network domain topology. The controller may control client switches of the client domains to allow only network traffic between the client domains and the non-client domains along the spanning tree. The controller may use the network domain topology to generate inter-domain forwarding maps. The inter-domain forwarding maps may be used to determine network forwarding paths between end hosts in the network.
摘要:
A monitoring system gathers both site-to-site measurements, and agent-to-agent measurements, wherein one or more agents are distributed in one or more sites in a distributed network environment. Site-to-site measurements between two sites, such as site one and site two, are obtained by testing between any agent at site one and any agent at site two. A measurement rate between sites and between agents is determined that allows for the detection of events, such as user perceivable events, without overwhelming the set of agents that form the distributed measurement system. With, for example, a scheduling mechanism used to schedule measurement tests of finite duration, as opposed to continuous streams of measurement packets, for each pair of agents, the test can be used as an indication as to whether the measurement system is overwhelmed.