Network Appliance, Inc.

Boston University

Netapp, Inc.

Mark Crovella

Azer Bestavros

Vitaly Revsin

Vijay R Basani

Domenic Joseph LaRosa

Leroy R Karge

Lynne M Murach

Krishna Mangipudi

Krishna Mangiapudi

Anukool Lakhina

To improve network reliability and management in today's high-speed communication networks, we propose an intelligent system using adaptive statistical approaches. The system learns the normal behavior of the network. Deviations from the norm are detected and the information is combined in the probabilistic framework of a Bayesian network. The proposed system is thereby able to detect unknown or unseen faults. As demonstrated on real network data, this method can detect abnormal behavior before a fault actually occurs, giving the network management system (human or automated) the ability to avoid a potentially serious problem.

Method and apparatus for whole-network anomaly diagnosis and method to detect and classify network anomalies using traffic feature distributions

A method for distributing information to a plurality of servers is described. The information is transmitted from a control manager server to a first group leader server. The first group leader server transmits the information to a second group leader server. The second group leader server requests instructions from the control manager server. The second group leader server then transmits, in response to the instructions, the information to a further group leader server. The first group leader server transmits the information to back end servers belonging to a group of the first group leader server. The second group leader server transmits the information to back end servers belonging to a group of the second group leader server. The further group leader server transmits the information to back end servers belonging to a group of the further group leader server.

Method and apparatus for dynamic resource discovery and information distribution in a data network

A system and apparatus for control and distribution of data files in large-scale distributed networks. Members of a group of servers in a multicast network elect a group leader whenever a new group leader is required, as detected by absence of a periodic message published by the leader. Election is carried out by a system of voting by each candidate whereby each candidate has a priority calculated from its configuration, and the server with the highest priority is configured to claim the leadership faster than the other candidates. As part of the claim, each candidate multicasts its priority. Each candidate that receives a multicast claim for leadership from another candidate compares its own priority against the claimant and only votes for itself if its own priority is higher. After a preconfigured period of hearing no other claimants with higher priority, the candidate with the highest priority becomes the new leader.

Method and apparatus for election of group leaders in a distributed network

The present invention provides a system and apparatus for efficient and reliable, control and distribution of data files or portions of files, applications, or other data objects in large-scale distributed networks. A unique content-management front-end provides efficient controls for triggering distribution of digitized data content to selected groups of a large number of remote computer servers. Network distribution messages are dispatched according to a sorted list of servers, based on factors such as nearness, server processor speed, reliability, and CPU Usage. For large numbers of servers, a store-and-forward approach becomes much more efficient. A first selected server receives the message from a content control manager (CCM). The first server requests instructions for the next server listed on an ordered list in the CCM and forwards a copy of that message to the next identified server. Each server reports its completion and requests further instructions from the CCM. This mechanism permits highly efficient and robust distribution of assignments and data content from the CCM to each required GL using a store-and-forward tree structure.

System for creating and distributing prioritized list of computer nodes selected as participants in a distribution job

The present invention provides a system and apparatus for efficient and reliable, control and distribution of data files in large-scale distributed networks. The members of a group of servers in a multicast network elect a group leader whenever a new group leader is required, as when the prior group leader become unavailable, as detected by absence of a periodic heartbeat message published by the leader. The election is carried out by a system of voting by each candidate whereby each candidate has a priority calculated from its configuration, and the server with the highest priority is configured to claim the leadership faster than the other candidates. As part of the claim, each candidate multicasts its priority. Each candidate that receives a multicast claim for leadership from another candidate compares its own priority against the claimant and only votes for itself if its own priority is higher. After a preconfigured period of hearing no other claimants with higher priority, the candidate with the highest priority becomes the new leader.

Method and apparatus for scalable distribution of information in a distributed network

The present invention provides a system and apparatus for efficient and reliable, control and distribution of data files or portions of files, applications, or other data objects in large-scale distributed networks. A unique content-management front-end provides efficient controls for triggering distribution of digitized data content to selected groups of a large number of remote computer servers. Transport-layer protocols interact with distribution controllers to automatically determine an optimized tree-like distribution sequence to group leaders selected by network devices at each remote site. Reliable store-and-forward transfer to clusters is accomplished using a unicast protocol in the ordered tree sequence. Once command messages and content arrive at all participating group leaders, local hybrid multicast protocols efficiently and reliably distribute them to the back-end nodes for interpretation and execution. Positive acknowledgement is then sent back to the content manager from each group leader, and the-updated content in each remote device autonomously goes &ldquo;live&rdquo; when the content change is locally completed.

Method and apparatus for reliable and scalable distribution of data files in distributed networks

In a computer network with a plurality of client computers and a group of host computers, each host computer in the group is capable of both servicing requests from the client computers and rerouting requests to other host computers in the group. Distribution of servicing and routing functions in accordance with predetermined criteria promotes fault tolerance, transparency and load sharing within the group. The predetermined criteria can be either stateless or utilize states. In the case of an Internet Web server, distribution of Layer routing obviates the need for a connection router to achieve load balancing. Distribution of Layer 3 routing reduces reliance on routing devices in general.

Growing community of inventors

Scituate, MA, United States of America

Mark Crovella