Computer networks generally comprise various interconnected computing devices that can exchange data. Computing devices in a computer network can be in direct communication with one or more other computing devices. Each direct communication connection between computing devices in a computer network is generally referred to as a network link or link. While a computer network is generally made up of a number of links, computing devices in a computer network do not typically include links to every other computing device in a computer network. Rather, data to be exchanged between computing devices can be subdivided into packets and propagated via the computer network to eventually reach an intended recipient, regardless of whether there is a direct link between the sender and recipient.
More specifically, packets of data are typically transmitted from an origin computing device to an identified destination computing device. If a packet of data is received by a computing device that is not the identified destination computing device, the receiving computing device becomes an intermediate in the communication path between the origin computing device and the destination computing device by forwarding the packet to another computing device in the computer network. Accordingly, each packet of data is transmitted through a series of intermediate links in the computer network until the packet reaches its destination computing device. The series of links for delivery of a packet of data between an origin computing device and a destination computing device is generally referred to as a network path or path.
At each computing device in a communication network, an independent decision may be made regarding the path to the identified destination computing device for each received data packet. Each computing device can use several factors for making the decision regarding the path to the identified decision. For example, in some networks, portions of the destination address included in the data packet may be used to compare to a lookup table on the computing device. Based on the independent decision, a receiving computing device transmits a received data packet on the next intermediate link in the path.
Indications of total traffic on any one link in the network may be obtained by measuring packets transmitted and/or received on the two computing devices connected by that link. However, as networks become increasingly complex, network operators may desire to obtain information regarding the performance of paths in the network, rather than indications of total traffic on individual links. The performance of paths in the network may include a view of the interconnection between all the computing devices in the network. Such a view is a closer approximation of what is experienced by a network user in terms of network performance. Performance of the paths may also include indications of dropped or lost packets, of service degradation, or even of a network halt due to excessive traffic.
Additionally, although there exist tools to detect when network outages occur, it may be difficult to find where the cause of failure lies in order to fix network related issues. Network path information allows network operators to isolate any network issues down to devices in the networks in order to take remedial action. Finally, in conventional systems, one or more hosts are dedicated to monitoring hundreds and even thousands of hosts. As the network becomes more complex, increasingly more hosts need to be dedicated to monitoring. Additionally, this type of monitoring simply provides an indication of responsive or non-responsive behavior from a host.
The foregoing aspects and many of the attendant advantages will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:
Generally described, aspects of the present disclosure relate to systems and methods for monitoring and detecting causes of failures of network paths. As discussed above, existing systems may enable a determination of total traffic on any one link in the network by measuring packets transmitted or received by two computing devices corresponding to a network link. However, network operators would like to obtain information regarding the performance of paths in the network, rather than indications of total traffic on individual links. Once equipped with network path information, network operators may also desire to isolate any network issues down to devices and links in the networks in order to take remedial action. Aspects of the present disclosure enable continuous monitoring of network paths to determine anomalies in the various nodes and links in order to find causes of failures and remedy them as appropriate.
Specifically, in one aspect, the network path monitoring and cause of failure detection system collects performance information from a plurality of nodes and links in a network, aggregates the collected performance information across paths in the network, processes the aggregated performance information for detecting failures on the paths, analyzes each of the detected failures to determine at least one root cause for each of the failures, and remedies the at least one root cause determined. In some aspects, processing the aggregated information may include solving a set of equations for the performance indications on each of a plurality of paths in the network. In another aspect, the system may also include an interface which makes available for display one or more of the network topologies, the collected and aggregated performance information, and indications of the detected failures in the one or more topologies.
Although various aspects of the disclosure will be described with regard to illustrative examples and embodiments, one skilled in the art will appreciate that the disclosed embodiments and examples should not be construed as limiting.
Illustratively, the network path monitoring and cause of failure detection system 102 may control and collect information from various nodes in network 130 through the control and collection service 110. The network path monitoring and cause of failure detection system 102 may also aggregate the information collected through the aggregation service 112. In various embodiments, the information collected may include paths taken between nodes in the network, as well as performance indices along the various paths. The performance indices may include latency, dropped packets, bandwidth of links, and the like. Using the information collected and aggregated, the data processing component 108 may create a network topology which may be made available to be displayed on a client computing device 120 through the interface component 104. The data processing component 108 may also use information received from one or more external data source 140. The data processing component 108 may also store information collected and aggregated into the data store 106 for later retrieval and use.
Illustratively, the origin node does not specify the path in which a packet may or must travel. Rather, if a receiving node is not the destination node (e.g., an intermediary node), the receiving node obtains a packet and transmits the packet to another node via a selected link. Accordingly, the results of each intermediary node forwarding a packet defines the path which the packet takes. As such, the same intermediary node may forward successive packets along different links, which would result in the successive packets being forwarded to the destination node along different paths based on the selection of the link the intermediary node. With reference to
One skilled in the relevant art will appreciate that networks monitored by the network path monitoring and cause of failure detection system 102 may include several more nodes than the illustrative network shown in
At block 302, the topology of the network is gathered and verified, in order to be used for allocation decisions as well as in failure detection, as described further in connection with the routines 400 and 500 illustrated in
At block 304, health checks are allocated across the links in the network. It is desired to not overload links in the network with health checks. However, a minimum number of health checks across the network are necessary for adequate monitoring of the network. The frequency of health checks may be set and adjusted in various ways. The frequency may be static, it may be manually adjusted, or it may also be dynamically adjusted based on business logic. The frequency of health checks may also be adjusted based on topology changes observed in block 302, based on frequency of such topology changes, as well as based on whether faults are detected at block 406 described below. As described above, a path includes source and destination nodes, and a series of intermediate nodes and links between the nodes. Packets arriving at a node may await transmission to the next node according to the packet's protocol as handled by the node. If the memory of the node is full when a packet arrives, the packet may be discarded. Otherwise, the packet is held until it reaches the front of the queue and is then forwarded to the next node on the way to its destination. This waiting mechanism may be responsible for observed packet losses and for packet latencies. Other reasons may also contribute to packet losses and/or latencies.
The ping utility may be used to check if a remote computing device is operating and to determine network connectivity. The source computing device may send an Internet Control Message Protocol (ICMP) packet to the remote computing device's IP address. If the destination computing device is up and the network connections are fine, the source computing device may receive a return an ICMP packet. Thus, one can collect data on roundtrip times and delays using the ping utility. Using other packet protocols, including for example TCP, UDP, and the like, may have different advantages and may be used in various embodiments. In some embodiments, transmitting a message with UDP packets instead of ICMP packets provides the added advantage of being able to manipulate paths between two endpoints. The manipulation of paths between the two endpoints may be achieved by manipulating port numbers. For example, the manipulation of paths may be achieved in accordance with flow preserving next-hop packet forwarding protocols such as Equal Cost Multi-Path (ECMP). With ECMP, and similar flow preserving packet forwarding strategies, at each node in the network, the decision on which path to take to send a packet to the destination computing device is done independently, and is deterministically dependent on the source port number, the destination port number, the source IP address and the destination IP address. The use of UDP packets by the transmitters of the network path monitoring and cause of failure detection system 102 allows the packets to be re-routed as necessary to a path for which data needs to be gathered. The re-routing is enabled by manipulation of port numbers. Each node learns and takes a default flow through the nodes in the network to arrive at a given destination. By manipulating the destination port through the use of UDP packets, the intermediate packet forwarding devices can be forced into taking a different, desired path. Therefore, in the network path monitoring and cause of failure detection system 102, each link is covered by a sufficient number of paths in order to identify a failing link from a set of failing paths. The various paths covering a link may be achieved by using one or more of the transmitters. Block 304 is described in further detail in connection with the routine 600 illustrated in
At block 306, the communication attributes across the network are measured. The communication attributes may be measured on one-way or on round-trip paths. Since the different paths of the network are discovered during topology gathering at block 302, the route followed by a data packet is known based on the combination of the source IP and port and destination IP and port used in the packet. The time taken to send and receive the packet is recorded by the data control and collection service 110.
In an illustrative embodiment, paths are discovered by sending UDP packets from the source to a destination, and received back at the source from the destination. The time-to-live (TTL) field of a packet may be exploited to determine the path that the packet takes to its destination. IP packets typically have a TTL field that can take on values between 0 and 255. When a node receives an IP packet, it may decrement this TTL field and forward the packet to its destination according to the node's routing table. If, however, the TTL field was already 0, the node may send back a packet, indicating TTL exceeded, to the source.
The UDP packets may be sent at increasing values of TTL, starting with 1, until the destination is actually reached. The source may send a UDP packet to some invalid port at the destination. When the destination receives a packet destined for an invalid port, a packet indicating “Port unreachable” may be sent back to the source to indicate the error. The source then knows the destination was reached. All the previous packets failed to reach the destination because the TTL was too small and the source received a TTL exceeded message from each of the intervening nodes between the source and the destination, in the order in which they appear. The rate of frequency for discovering and verifying paths is centrally scheduled by the data control and collection service 110. The frequency may be the same for all nodes in the network, or it may be adjusted according to the load on each node. At block 308, the communication attributes measured and recorded are transmitted to the aggregation service 112 and/or the data store 106 and the interface component 104 of the network path monitoring and cause of failure detection system 102.
At block 402, the communication attributes collected by the control and collection service 110 are received by the aggregation service 112. The communication attributes collected by each of the selected nodes are aggregated. Aggregation of the communication attributes enables reliable detection of failing paths. Data collected across several paths crossing the same node through different links and/or through packets sent from different transmitter nodes are aggregated.
At block 403, the communication attributes collected are used to determine whether the allocation strategy adopted is appropriate. The allocation strategy aims to provide adequate coverage of all the paths in the network. The communication attributes collected may indicate a need to adjust the allocation strategy in order to collect more path information. The health check frequency may thus be increased in some scenarios. In some scenarios, new paths may be allocated to one more different agents on the networks. Additionally, a determination of inadequate coverage of network paths may trigger an alarm at block 410. Block 302 of the control and collection service routine 300 may also be repeated if the allocation strategy is deemed to not be appropriate at block 403.
At block 404, using the communication attributes aggregated, the aggregation service calculates performance characteristics of the nodes and links in the network, using the network topology gathered at block 302 of the control and collection service routine 300. Performance characteristics may include indications of packet loss, latency, throughput, jitter and the like. The aggregation service 112 may store the information collected and aggregated in a data store such as data store 106 illustrated in
At block 406, the aggregation service 112 performs failure detection. Using the network topology gathered at block 302 of the control and collection service routine 300, the aggregation service 112 may iterate through all the links in the network topology in order to compute a percentage of links and nodes which indicate a failure. The links and nodes may be sorted by failure percentage in order to isolate a link or node experiencing performance issues such as packet latency or loss, jitter, or the like. A search may be performed on the network topology for nodes around a failing link. Connected failures may be found to be due to common failing nodes. Block 302 of the control and collection service routine 300 may also be repeated if failures are detected at block 406.
At block 408, the aggregation service performs statistical analysis on the aggregated data to determine the root cause of the detected failures. Candidate failures may be fully determined by the inference process leading to a root cause candidate, or otherwise the aggregation service may perform additional queries or analysis to isolate root causes of failures. In some aspects, the aggregation service 112 may also use additional information from the network, including the external data source(s) 140. For example, information collected by the devices, such as, for example Simple Network Management Protocol (SNMP) data, information from syslogs on the nodes, information from Terminal Access Controller Access-Control System (TACACS), and Change Management (CM) logs may be used to perform root cause analysis. Change Management logs may be used in managing and tracking changes to physical and logical configurations of network devices, links, and other services running in a data center. In various scenarios, the root cause of a failure may be isolated to be related to a power device failure, or to a network device, cabling, or transceiver failure. In some embodiments, the cause may be attributable to multiple simultaneous events.
In some embodiments, root cause analysis may be performed by developing a set of equations given a performance indication across a path in order to solve for the performance indication for each link and node in the path. For example, one indication of performance may be latency. Latency includes the latency across a link connecting two nodes, as well as the latency of processing the network packets on each of the endpoint nodes. The total latency across a path may be equated to the sum of the latencies of each node on the path of a packet, and the latencies of each link on that path. Each value of latency may be an integer, or it may be represented by a statistical distribution aggregated from several sample measurements. By using the latencies across all paths for which data is aggregated, the latency of each node and each link may be solved for by solving the set of equations. Once an indication of the latency at each node and each link is known, it is possible to determine the root cause of failure by isolating the faulty link and/or node.
In order to perform efficient root cause analysis, data for enough different paths needs to be collected, as indicated above. As the size of the network grows, the set of equations to be solved for becomes increasingly more complex. As information for more paths is collected, it becomes easier to isolate a link or a node in the network associated with a failure. Some other indications of performance may be packet loss, jitter, available bandwidth, and the like. In some embodiments, packet loss may be determined to be a threshold value of latency. For example, latencies over 100 ms may be considered to be packet losses. In other embodiments, latencies over different values may be considered to be packet losses. In some embodiments, the triangulation may include assigning a confidence level to the finding of a faulty node or link.
After the candidate cause analysis, there may optionally be potential scenarios of failure identified, along with an indication of the likelihood of each scenario (not shown in
At block 410, the aggregation service develops remediation and alarming strategies. Remediation and alarming may take on several different forms in various embodiments. Before remediation and alarming is performed, there may be validation of the identified failure in order to ensure that it is a failure for which a remediation and/or alarming strategy is known, and/or still required. In some embodiments, remediation may include shutting down a node which is known, with a high level of confidence, to be faulty. In some embodiments, the remediation may include ensuring that packets are not routed through the faulty link or node. In some embodiments, remediation may include disabling ports on a given node. In some embodiments, alarming may include sending an engineer or technician to have a look at the node or link identified to be faulty. In some embodiments, power to a node may be cycled. The power cycle of the node may be done after packets have been routed away from the device, or it may also be done as a fallback mechanism.
At block 502, the control and collection service 110 gathers a network topology. A network may include a set of nodes and links, which are interconnections between pairs of nodes. Packets traverse through the network along paths. A path may include a sequence of links which indicate the transmission route when a packet travels between nodes within the network.
Data transferred between the nodes may be controlled by several different protocols, with each protocol fulfilling a particular need. For example, the Transmission Control Protocol (TCP) may guarantee reliable and in-order delivery of data from a sender to receiver. Open Shortest Path First (OSPF) is an adaptive routing protocol for Internet Protocol (IP) networks which uses a link state routing algorithm. The link-state protocol is performed by every switching node in the network (i.e. nodes that are prepared to forward packets). With link-state routing, each node constructs a map of the connectivity to the network, in the form of a graph, showing which nodes are connected to which other nodes. Each node then independently calculates the next best logical path from it to every possible destination in the network. The collection of best paths will then form the node's routing table. The control and collection service 110 may thus use information on each node's routing table in order to create the network topology.
At block 504, the control and collection service 110 verifies to see whether the topology created based on the nodes' routing tables is the topology expected. The topology expected may be based on diagrams provided by network technicians. The diagrams provided may also be associated with various confidence levels. The expected topology may also be based on a knowledge of the workflow of the build process for the network. For example, it may be known that the network was initially designed with a 100 nodes, and there was a planned expansion of a doubling of nodes in a given timeframe within a given geographic area. The topology may also be inferred from a combination of external sources, such as configuration files, technicians' information, automated switch building, subnet analysis, SNMP query information regarding run-time configuration states of devices, or other monitoring services. The topology of the network is gathered and stored, and it may be made available for display. If the topology gathered at block 502 is as expected, the routine then moves to block 508. However, if the topology is not as expected, the control and collection service 110 uses the other information available, in order to reconcile the topology at block 506.
In some embodiments, it may be most desirable to track network topologies at Layer 3 of the Open Systems Interconnection (OSI) standard, such as the Internet Protocol (IP). However, topologies of other OSI layers, for example Layer 2 such as Ethernet, as well as other data center topologies, such as for power to racks, power to hosts, power to network devices, and others, may also be desired to be gathered at block 508. A topology may also be created for a physical location within a thermal zone, a room, a data center, or another physical locality. The topologies thus created may be used to correlate with failures detected at block 406. If other topologies are gathered, the control and collection service reconciles information gathered from the external sources with those other layers and topologies at block 510. Various embodiments may be widely applicable to any networking protocol suite or layer and are not restricted to Layer 2 or 3, IP, Ethernet, or the Internet Protocol Suite.
As described above, transmitting a message with UDP packets instead of ICMP packets provides the added advantage of being able to manipulate paths between two endpoints by manipulating port numbers. It may be desired to throttle the frequency of health checks to manage the load generated on network links. However, a minimum number of health checks are necessary for adequate coverage and monitoring of the network. In order to accurately measure packet drops on links to nodes, each node is tested for reachability at an ideal frequency designed to keep the amount of data generated by the transmission of the messages to a workable level while accurately measuring packet loss. In some embodiments, a health check may be initiated every 100 milliseconds, or every 500 milliseconds, or every 5 seconds, or every 5 minutes, or other any other suitable period of time according to business and/or other requirements of the network supported service.
At block 604, using the network topology previously gathered, each link in the network is iterated through in order to ensure that at least one path traverses the link. If a path is successfully allocated to a given link, a counter for all links on a path may be incremented by a certain value. If however if a path is not allocated to a link yet, then at block 606, the health check allocation may be adjusted to achieve a desired path until all links achieve a target number of paths per link. In some embodiments, a warning may be issued by the control and collection service to indicate links which are not covered.
It will be appreciated by those skilled in the art and others that all of the functions described in this disclosure may be embodied in software executed by one or more processors of the disclosed components and mobile communication devices. The software may be persistently stored in any type of non-volatile storage.
Conditional language, such as, among others, “can,” “could,” “might,” or “may,” unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements, and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular embodiment.
Any process descriptions, elements, or blocks in the flow diagrams described herein and/or depicted in the attached figures should be understood as potentially representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process. Alternate implementations are included within the scope of the embodiments described herein in which elements or functions may be deleted, executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those skilled in the art. It will further be appreciated that the data and/or components described above may be stored on a computer-readable medium and loaded into memory of the computing device using a drive mechanism associated with a computer readable storing the computer executable components such as a CD-ROM, DVD-ROM, or network interface further, the component and/or data can be included in a single device or distributed in any manner. Accordingly, general purpose computing devices may be configured to implement the processes, algorithms, and methodology of the present disclosure with the processing and/or execution of the various data and/or components described above.
It should be emphasized that many variations and modifications may be made to the above-described embodiments, the elements of which are to be understood as being among other acceptable examples. All such modifications and variations are intended to be included herein within the scope of this disclosure and protected by the following claims.
This application is a continuation of U.S. patent application Ser. No. 17/024,558, entitled MONITORING AND DETECTING CAUSES OF FAILURES OF NETWORK PATHS, which is a continuation application of U.S. patent application Ser. No. 15/200,398, entitled MONITORING AND DETECTING CAUSES OF FAILURES OF NETWORK PATHS, and filed on Jul. 1, 2016. U.S. patent application Ser. No. 15/200,398 is a continuation application of U.S. patent application Ser. No. 13/077,589, entitled MONITORING AND DETECTING CAUSES OF FAILURES OF NETWORK PATHS, and filed on Mar. 31, 2011. The entireties of these applications are incorporated by reference herein.
Number | Name | Date | Kind |
---|---|---|---|
4853927 | Wenzel | Aug 1989 | A |
5477531 | McKee et al. | Dec 1995 | A |
5504921 | Dev et al. | Apr 1996 | A |
5832225 | Hacherl et al. | Nov 1998 | A |
5864662 | Brownmiller et al. | Jan 1999 | A |
6185612 | Jensen et al. | Feb 2001 | B1 |
6377987 | Kracht | Apr 2002 | B1 |
6614762 | Iliadis | Sep 2003 | B1 |
6654914 | Kaffine et al. | Nov 2003 | B1 |
6671818 | Mikurak | Dec 2003 | B1 |
6678250 | Grabelsky et al. | Jan 2004 | B1 |
6694455 | Scrandis | Feb 2004 | B1 |
6738933 | Fraenkel et al. | May 2004 | B2 |
6747991 | Hemy et al. | Jun 2004 | B1 |
6823479 | McElhaney et al. | Nov 2004 | B1 |
6901530 | Cerami et al. | May 2005 | B2 |
6909741 | Smith et al. | Jun 2005 | B1 |
6937579 | Bantz et al. | Aug 2005 | B2 |
6978302 | Chisholm et al. | Dec 2005 | B1 |
6981039 | Cerami et al. | Dec 2005 | B2 |
7016313 | Harper | Mar 2006 | B1 |
7134135 | Cerami et al. | Nov 2006 | B2 |
7251055 | Sawada et al. | Jul 2007 | B2 |
7260060 | Abaye et al. | Aug 2007 | B1 |
7337206 | Wen | Feb 2008 | B1 |
7385924 | Riddle | Jun 2008 | B1 |
7423995 | Elliott et al. | Sep 2008 | B1 |
7441154 | Klotz et al. | Oct 2008 | B2 |
7457868 | Guo | Nov 2008 | B1 |
7525922 | Cidon et al. | Apr 2009 | B2 |
7546609 | Florissi et al. | Jun 2009 | B2 |
7609650 | Roskowski et al. | Oct 2009 | B2 |
7639705 | Watanabe et al. | Dec 2009 | B2 |
7706373 | Xu et al. | Apr 2010 | B2 |
7751350 | Pabst | Jul 2010 | B1 |
7788536 | Qureshi et al. | Aug 2010 | B1 |
7844730 | Kawaguchi | Nov 2010 | B2 |
7936694 | Choudhury | May 2011 | B2 |
7949739 | Florissi et al. | May 2011 | B2 |
7953020 | Breslau et al. | May 2011 | B2 |
7990946 | Deguchi | Aug 2011 | B2 |
8018844 | Bender et al. | Sep 2011 | B2 |
8050182 | Yamamoto | Nov 2011 | B2 |
8072901 | Blair | Dec 2011 | B1 |
8098583 | Cahn | Jan 2012 | B2 |
8196199 | Hrastar et al. | Jun 2012 | B2 |
8204980 | Sandstrom et al. | Jun 2012 | B1 |
8223655 | Heinz et al. | Jul 2012 | B2 |
8254272 | Vasseur | Aug 2012 | B1 |
8300554 | Vijendra et al. | Oct 2012 | B1 |
8375244 | Bobak et al. | Feb 2013 | B2 |
8433894 | Reznik et al. | Apr 2013 | B2 |
8520556 | Karuppiah | Aug 2013 | B2 |
8661295 | Khanna | Feb 2014 | B1 |
8675502 | Blair | Mar 2014 | B2 |
8937870 | Callaghan | Jan 2015 | B1 |
9001667 | Khanna et al. | Apr 2015 | B1 |
9077633 | Meirosu | Jul 2015 | B2 |
9104543 | Cavanagh et al. | Aug 2015 | B1 |
9197495 | Rauser et al. | Nov 2015 | B1 |
9210038 | Rauser et al. | Dec 2015 | B1 |
9385917 | Khanna et al. | Jul 2016 | B1 |
9712290 | Callaghan | Jul 2017 | B2 |
9742638 | McCabe et al. | Aug 2017 | B1 |
9882984 | Chiu et al. | Jan 2018 | B2 |
10103851 | Callaghan | Oct 2018 | B2 |
10411972 | McCabe et al. | Sep 2019 | B2 |
10785093 | Khanna et al. | Sep 2020 | B2 |
10868708 | Jayaraman | Dec 2020 | B2 |
20020010735 | McMillen et al. | Jan 2002 | A1 |
20020016856 | Tallegas et al. | Feb 2002 | A1 |
20020107980 | Kawaguchi | Aug 2002 | A1 |
20020165957 | Devoe et al. | Nov 2002 | A1 |
20030076840 | Rajagopal et al. | Apr 2003 | A1 |
20030086425 | Bearden et al. | May 2003 | A1 |
20030088671 | Klinker et al. | May 2003 | A1 |
20030156541 | Haihong | Aug 2003 | A1 |
20040034614 | Asher et al. | Feb 2004 | A1 |
20040044764 | Padmanabhan et al. | Mar 2004 | A1 |
20040044765 | Meek et al. | Mar 2004 | A1 |
20040208128 | Lu et al. | Oct 2004 | A1 |
20040218582 | Kennedy et al. | Nov 2004 | A1 |
20040252700 | Anandakumar et al. | Dec 2004 | A1 |
20050018623 | Bantz et al. | Jan 2005 | A1 |
20050041593 | Kikuchi | Feb 2005 | A1 |
20050053009 | Denby et al. | Mar 2005 | A1 |
20050091361 | Bernstein et al. | Apr 2005 | A1 |
20050122996 | Azenkot et al. | Jun 2005 | A1 |
20050169185 | Qiu et al. | Aug 2005 | A1 |
20050256961 | Alon et al. | Nov 2005 | A1 |
20060107086 | Walker et al. | May 2006 | A1 |
20060164975 | Filsfils | Jul 2006 | A1 |
20060215577 | Guichard et al. | Sep 2006 | A1 |
20060218447 | Garcia et al. | Sep 2006 | A1 |
20060259984 | Juneau | Nov 2006 | A1 |
20070047453 | Bender et al. | Mar 2007 | A1 |
20070053283 | Bidwell et al. | Mar 2007 | A1 |
20070064715 | Lloyd et al. | Mar 2007 | A1 |
20070086335 | McCanne et al. | Apr 2007 | A1 |
20070091811 | Thubert et al. | Apr 2007 | A1 |
20070201380 | Ma et al. | Aug 2007 | A1 |
20070263540 | Charzinski et al. | Nov 2007 | A1 |
20080089235 | Kotrla et al. | Apr 2008 | A1 |
20080089236 | Kotrla et al. | Apr 2008 | A1 |
20080148099 | Bhat et al. | Jun 2008 | A1 |
20080186866 | Morinaga et al. | Aug 2008 | A1 |
20080205263 | Cooley et al. | Aug 2008 | A1 |
20080225733 | Hua et al. | Sep 2008 | A1 |
20080253295 | Yumoto et al. | Oct 2008 | A1 |
20080298271 | Morinaga et al. | Dec 2008 | A1 |
20090037771 | Morse | Feb 2009 | A1 |
20090067483 | Casas et al. | Mar 2009 | A1 |
20090086642 | Davis et al. | Apr 2009 | A1 |
20090086643 | Kotrla et al. | Apr 2009 | A1 |
20090116404 | Mahop et al. | May 2009 | A1 |
20090122697 | Madhyasha et al. | May 2009 | A1 |
20090138618 | Li et al. | May 2009 | A1 |
20090245115 | Krishnaswamy et al. | Oct 2009 | A1 |
20090271513 | Liu et al. | Oct 2009 | A1 |
20090285101 | Lu | Nov 2009 | A1 |
20090290497 | Gibbings | Nov 2009 | A1 |
20100002722 | Porat et al. | Jan 2010 | A1 |
20100027415 | So et al. | Feb 2010 | A1 |
20100034098 | Wang | Feb 2010 | A1 |
20100067396 | Cui et al. | Mar 2010 | A1 |
20100085948 | Yu et al. | Apr 2010 | A1 |
20100121910 | Kim | May 2010 | A1 |
20100157516 | Doorhy et al. | Jun 2010 | A1 |
20100161852 | Veni et al. | Jun 2010 | A1 |
20100165849 | Eisenberg et al. | Jul 2010 | A1 |
20100246408 | Kerber et al. | Sep 2010 | A1 |
20100278049 | Meloche et al. | Nov 2010 | A1 |
20100278056 | Meloche et al. | Nov 2010 | A1 |
20100316373 | Chang et al. | Dec 2010 | A1 |
20110007629 | Atlas et al. | Jan 2011 | A1 |
20110063979 | Matthews et al. | Mar 2011 | A1 |
20110063986 | Denecheau et al. | Mar 2011 | A1 |
20110078291 | Bickson et al. | Mar 2011 | A1 |
20110096675 | Li et al. | Apr 2011 | A1 |
20110134791 | So et al. | Jun 2011 | A1 |
20110164502 | Mohan et al. | Jul 2011 | A1 |
20110199911 | Ikada | Aug 2011 | A1 |
20110205898 | Ichiki et al. | Aug 2011 | A1 |
20110292813 | Dunbar et al. | Dec 2011 | A1 |
20110317580 | Kozisek et al. | Dec 2011 | A1 |
20120093004 | Nishi | Apr 2012 | A1 |
20120106561 | Horio | May 2012 | A1 |
20120109600 | Saeed et al. | May 2012 | A1 |
20120163163 | Kim et al. | Jun 2012 | A1 |
20120182864 | Heinz et al. | Jul 2012 | A1 |
20120213224 | Chen | Aug 2012 | A1 |
20120239256 | Hammerschmidt et al. | Sep 2012 | A1 |
20120278477 | Terrell et al. | Nov 2012 | A1 |
20120320784 | Edwards et al. | Dec 2012 | A1 |
20120327765 | Gibbings | Dec 2012 | A1 |
20130064075 | Pu | Mar 2013 | A1 |
20130070612 | Timus et al. | Mar 2013 | A1 |
20130117272 | Barga et al. | May 2013 | A1 |
20130308474 | Krzanowski et al. | Nov 2013 | A1 |
20140098685 | Shattil | Apr 2014 | A1 |
20140280884 | Searle et al. | Sep 2014 | A1 |
20150142970 | Callaghan | May 2015 | A1 |
20150223095 | Centonza | Aug 2015 | A1 |
20170155544 | Khanna | Jun 2017 | A1 |
20170310432 | Callaghan | Oct 2017 | A1 |
20170331704 | McCabe et al. | Nov 2017 | A1 |
20200112905 | Railkar et al. | Apr 2020 | A1 |
Entry |
---|
Batsakis, A., Practical Passive Lossy Link Inference, Proc. of PAM, 2005. |
Bu, T., et al., Network Tomography on General Topologies, Proc. of the ACM Sigmetrics, Jun. 2002. |
Cavanagh, et al., Determining Locations of Network Failures, U.S. Appl. No. 13/441,179, filed Apr. 6, 2012. |
Coates, M., et al., Network Inference From Passive Unicast Measurements, Rice University, ECE Department, Technical Report, Jan. 21, 2000. |
Khanna, R., et al., Monitoring and Detecting Causes of Failures of Network Paths, U.S. Appl. No. 13/077,589, filed Mar. 31, 2011. |
Czepiel, S., Maximum Likelihood Estimation of Logistic Regression Models: Theory and Implementation, http://czep.net/stat/mlelr.pdf, available as of Feb. 5, 2005. |
Kniaz, K., Non-Gradient Optimization Techniques (Nelder-Mead and Rosenbrock), http://www.kniaz.net/software/rosnm.aspx, May 2009. |
Liakopoulos, et al., Applying distributed monitoring techniques in autonomic networks, 2010 IEEE Globecom Workshops, 2010, pp. 498-502, doi: 10.1109/GLOCOMW.2010.5700369. (Year 2010). |
McCabe, K.A., et al., Determining Impact of Network Failures, U.S. Appl. No. 13/959,566, filed Aug. 5, 2013. |
Salakhutdinov, R., et al., Optimization With EM and Expectation-Conjugate-Gradient, Proceedings of the Twentieth International Conference on Machine Learning, Washington, D.C., 2003. |
Sauro, J., What's a Z-score and Why Use it in Usability Testing?, http://www.measuringusability.com/z.htm, Sep. 17, 2004. |
Sharman, K.C., Maximum Likelihood Parameters Estimation by Simulated Annealing, International Conference on Acoustics, Speech, and Signal Processing, pp. 2741-2744, Apr. 1988. |
Tachibana, A., et al., Empirical Study on Locating Congested Segments Over the Internet Based on Multiple End-to-End Path Measurements, Proc. IEEE/IPSG International Symposium on Applications and the Internet, Jan. 2005. |
Tsang, Y., et al., Passive Network Tomography Using EM Algorithms, IEEE International Conference on Acoustics, Speech, and Signal Processing, Proceedings, vol. VI, May 2001. |
Number | Date | Country | |
---|---|---|---|
20230171148 A1 | Jun 2023 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 17024558 | Sep 2020 | US |
Child | 18161819 | US | |
Parent | 15200398 | Jul 2016 | US |
Child | 17024558 | US | |
Parent | 13077589 | Mar 2011 | US |
Child | 15200398 | US |