1844 results about "Cluster systems" patented technology

In one embodiment, a method comprises executing a plurality of virtual machines on a plurality of nodes of a cluster computing system, wherein at least one application is executed within each of the plurality of virtual machines, generating data that is related to performance of applications in the virtual machines, analyzing, by a management process, the data in view of parameters that encode desired performance levels of applications, and migrating, by the management process, a virtual machine on a first node to a second node of the plurality of nodes in response to the analyzing.

A firewall clustering system connects two or more firewalls between an internal network and an external network. The plurality of two or more firewalls are combined to supply high-availability and scaling of processing capacity. Firewalls maintain client-server state information. Flow controllers are connected to the firewalls and placed on both the internal “trusted” side and the external “untrusted” side of the firewalls. Flow controllers are placed on both sides of the firewalls to ensure that traffic for a given client-server session flows through the same firewall in both inbound and outbound directions. The firewalls perform filtering operations and / or network address translation (NAT) services. In both cases, the flow controllers supply high availability, scalability, and traffic distribution for the firewalls in the firewall cluster.

A path control device that controls first and second paths for accessing a peripheral subsystem, includes a command substituting unit that substitutes a first reserve command that allows an access through the first path, with a second reserve command that allows both of accesses through both of the first path and the second path.

In an embodiment, the systems, methods, and devices disclosed herein comprise a computer resource monitoring and allocation system. In an embodiment, the resource monitoring and allocation system can be configured to allocate computer resources that are available on various nodes of a cluster to specific jobs and / or sub-jobs and / or tasks and / or processes.

A method, apparatus and program storage device for providing failover for continuous or near-continuous availability in an N-way logical shared-nothing cluster system is disclosed. Cluster application data space partitions are assigned to each node in the cluster and each node's or server software's internal architecture is partitioned in accordance with the application data partitions assigned to the node. Cluster-integrity protection is performed. A failover and recovery protocol is performed based upon the assigned partitions and the partitioned and bound internal architecture. Containment of the impact of failure is provided such that most of the application data space partitions are not impacted. Affected partition sets are failed over fast and in constant time and so actual load on the surviving nodes does not affect failover duration. When shared storage is not provided, synchronous log replication may be used to facilitate failover and log-based recovery.

A system and method for using continuous failure predictions for proactive failure management in distributed cluster systems includes a sampling subsystem configured to continuously monitor and collect operation states of different system components. An analysis subsystem is configured to build classification models to perform on-line failure predictions. A failure prevention subsystem is configured to take preventive actions on failing components based on failure warnings generated by the analysis subsystem.

Systems, methods, apparatus and software can make use of coordinator resources and SCSI-3 persistent reservation commands to determine which nodes of a cluster should be ejected from the cluster, thereby preventing them from corrupting data on a shared data resource. Fencing software operating on the cluster nodes monitors the cluster for a cluster partition (split-brain) event. When such an event occurs, software on at least two of the nodes attempts to unregister other nodes from a majority of coordinator resources. The node that succeeds in gaining control of the majority of coordinator resources survives. Nodes failing to gain control of a majority of coordinator resources remove themselves from the cluster. The winning node can also proceed to unregister ejected nodes from shared data resources. These operations can be performed in parallel to decrease failover time. The software can continue to execute on all nodes to prevent additional problems should a node erroneously attempt to reenter the cluster.

A cluster system and method accesses from an internet network, a network server within one or a plurality of clusters, each cluster being identified by a single cluster public Internet Protocol (IP) address. The cluster system has a plurality of network servers organized in one of a plurality of clusters and a network load balancer system for selecting a destination network server in a cluster. Each cluster has one or a plurality of identical network servers, the network load balancer system being connected on one hand to an access routing device and on another hand to the plurality of network servers through a private network server subnet. The method includes the steps of at initialization time, on each network server defining, as a non-advertising alias, in an interface table, the public IP address of each cluster to which the network server belongs, and upon reception, by the network load balancing system, of a datagram having an IP header including a destination IP address field and a medium access control (MAC) header including a destination MAC address field, selecting a destination network server within the cluster corresponding to the cluster public IP address identified in the destination IP address field of the datagram IP header, replacing the destination medium access control (MAC) address field of the datagram MAC header by the MAC address of the selected destination network server, and sending the datagram through the private network server subnet, using the MAC address of the selected destination network server. Upon reception, by the destination network server, of the datagram sent by the network load balancing system, the MAC address in the destination MAC address field of the datagram MAC header is identified as being the MAC address of the selected destination network server, and the IP datagram is processed if the identified cluster public IP address in the destination IP address field of the datagram IP header, is defined as a non-advertising alias in the interface table of the destination network server.

A recording medium which is readable by a computer stores a performance analyzing program for searching for a node that is peculiar in performance in a cluster system, as well as unknown problems. The performance analyzing program enables the computer to function as various functional units. A performance data analyzing unit collects performance data of nodes which make up the cluster system from performance data storage unit for storing a plurality of types of performance data of the nodes, and analyzes performance values of the nodes based on the collected performance data. A classifying unit classifies the nodes into a plurality of groups by statistically processing the performance data collected by the performance data analyzing unit according to a predetermined classifying condition. A group performance value calculating unit statistically processes the performance data of the respective groups based on the performance data of the nodes classified into the groups, and calculates statistic values for the respective types of the performance data of the groups. A performance data comparison display unit displays the statistic values of the groups for the respective types of the performance data for comparison between the groups.

A system and method for using continuous failure predictions for proactive failure management in distributed cluster systems includes a sampling subsystem configured to continuously monitor and collect operation states of different system components. An analysis subsystem is configured to build classification models to perform on-line failure predictions. A failure prevention subsystem is configured to take preventive actions on failing components based on failure warnings generated by the analysis subsystem.

Various embodiments of the invention provide solutions to allow more efficient management of clustered environments. Merely by way of example, some embodiments provide a framework for the relatively automated management of clusters, including the addition, deletion and / or modification of cluster nodes. In certain embodiments, this framework may be incorporated within a software application that can assist in the configuration of clusters. Other embodiments provide methods of managing clusters and / or their nodes, including, inter alia, methods of establishing and / or implementing such frameworks. Still other embodiments provide systems, including clustered systems, that provide relatively automated cluster management capabilities.

Systems and methods for grouping mesh networks. Mesh networks enable energy data to be transferred from an energy sensing device to a destination device using various networks and reduces the need to install metering stations. Embodiments of the invention join or group mesh networks that otherwise are not able to communicate. The connections between the grouped mesh networks can be passive using directional antennas or passive antennas or active using higher power RF devices, solar repeaters, mobile mesh devices, etc. The energy data is routed in a smart manner, by performing load balancing at gateways, by monitoring the signal to noise ratio of available communication paths, etc. Mesh clusters can also be created to define low power clusters, address blocking clusters, and frequency based clusters. Grouping clusters facilitates the efficient transfer of energy data from an energy sensing device to a energy management station.

The invention discloses a Hadoop cluster system and Hadoop cluster rapid establishment method based on a container technology and relates to the technical field of big data and cloud computing. The method comprises the steps of configuring that containers between different container operation nodes communicate through a VXLAN, and configuring that virtual network bridges of the container operation nodes are not located in the same network segment; installing SSH and JAVA operating environments, producing packaging files of a DNS service mirror image, a Hadoop master service mirror image and hadoop slave service mirror images, packaging the mirror images, and transmitting the packaged mirror images to a mirror image warehouse for storage; and submitting requests for deploying three pieces of services to a container management node in sequence by a user through a client, scheduling the corresponding requests to the container operating nodes for deployment by the container management node, and registering the corresponding service. According to the system and the method, a Hadoop cluster can be deployed rapidly, the condition that the number of servers is insufficient is solved, and the availability of a Hadoop container cluster deployed on a Docker platform is improved.

A load distribution method is adopted by a client-server system comprising a plurality of clients 100 and a server cluster 1100, which includes a plurality of servers 800 each used for processing requests made by the clients 100 and allows the number of the servers 800 to be changed dynamically. Each of the clients 100 detects the number of servers 800 composing the server cluster 1100, sets an allocation of requests transmissible out to a newly added server 900 at a value small in comparison with that set for each of the other servers 800 right after detecting an increase in server count and then transmits out requests to the servers on the basis of the set allocation. It is thus possible to provide the clients 100 and the server-cluster system 1100 with the load distribution method suitable for a cluster reconfiguration technology for changing the number of servers composing the server-cluster system 1100 in accordance with an increase and a decrease in demand for a service.

A system and method for utilizing virtual ports associated with a physical port in a Fibre Channel adapter for use in a clustered storage system is disclosed. A plurality of virtual ports are created and associated with a physical port on a storage system. One of the virtual ports is utilized as a standby port to be used to assume the identity of a port associated with a failed storage system in the cluster. In the event of a failover operation, the standby port is activated and assumes the identity of the failed storage systems port. Data access requests previously directed to the failed storage system are received by the activated virtual port and processed by the surviving storage system.

The application discloses a file processing method and system based on cloud storage, and a server cluster system. The method comprises the following steps that: a WEB server receives a file uploading request sent by a client; the WEB server returns a file ID (Identity) which is generated for a new file to be uploaded by a server cluster; the client splits the new file and generates a unique slice ID for each slice file; the client uploads a slice file data packet to the WEB server, the slice file data packet comprises the file ID, the slice ID of each slice file and slice data of each slice file; and the WEB server sends the slice data of each slice file in the slice file data packet to corresponding service terminals in the server cluster for storage. By virtue of the embodiment of the application, the file is subjected to splitting processing, and the slice files are stored in the server cluster in a distributed manner, so that the cloud storage performance based on the server cluster is relatively optimal.

Provided is a consistency maintenance system and methods for distributed-type data in a distributed-type cluster system. The consistency maintenance system and methods for distributed-type data is composed of a cluster management node, a global lock management node and a plurality of data maintenance nodes which are scattered in the system, wherein the data maintenance nodes are the isomorphic nodes suitable for storing one or more data replicas. The data maintenance nodes are provided with three modules of monitoring events, maintaining data consistency and sending events. The global lock management node is responsible for storing and managing update locks of all the data in a system, and storing all information of the data maintenance nodes. The global lock management node is provided with three modules of receiving and sending massage, the update locks requesting queue, and the information management for data maintenance nodes. The cluster management node is responsible for managing the latest information of all the nodes in the system, and testing the state of each node in the system periodically. The consistency maintenance system and methods for distributed-type data have the advantages of being reliable and flexible in work methods. Besides, a client can develop a data update towards a plurality of targets at the same time. The consistency maintenance system and methods for distributed-type data have a simple operation, a less communication overhead, a short updating time delay, and a promising application prospects.

Provided is a failover method for a cluster system for realizing smooth failover of the guest OS's, even when there are many guest OS's, while reducing consumption of computer resources of a server. Smooth failover is realized by preventing competition during failover even when the number of guest OS's is increased. In a cluster configuration in which a slave / master cluster program is operated in a guest OS / host OS, the master cluster program (510) collects and transmits heartbeats of the slave cluster program, thereby realizing failure monitoring through the certain amount of heartbeats without depending on the number of guest OS's. Further, when the master cluster program monitors failures of the slave cluster program of its own computer to find a normal operation of the guest OS, the amount of communication through heartbeats is reduced by eliminating the necessity of communication to a standby system slave cluster program.

A method and an apparatus of allocating available resources in a cluster system with learning models and tuning methods are provided. The learning model may be trained from historic performance data of previously executed jobs and used to project a suggested amount of resources for execution of a job. The tuning process may suggest a configuration for the projected amount of resources in the cluster system for an optimal operating point. An optimization may be performed with respect to a set of objective functions to improve resource utilization and system performance while suggesting the configuration. Through many executions and job characterization, the learning / tuning process for suggesting the configuration for the projected amount of resources may be improved by understanding correlations of historic data and the objective functions.

In a cluster that is composed of two computer nodes and has no common storage, mutual aliveness is monitored over networks. However, this is insufficient because a party node may be wrongly determined as inactive. If failover is performed according to wrong determination, the counterpart may be restored to a normal condition after the failover, so that both the two computers may operate as master. The two nodes to constitute the cluster and other computers to communicate with the cluster are connected by switches that can disable ports to which the computers are connected. A network control program that controls the switches changes the legality of use of ports to which the nodes are connected, synchronously with node failover.

The embodiment of the invention provides a method and equipment for migrating virtual machines based on a network data flow direction, and a cluster system. Each network flow monitoring module of the cluster system is used for interacting with a virtual machine monitor (VMM) to acquire network flow information of each virtual machine on the current service node; a virtual machine deployment adjustment module is used for acquiring network flow information of the virtual machines from network flow monitoring modules on a plurality of service nodes in the cluster system, determining a data flow direction relation among the virtual machines in the cluster system according to the network flow information, and transmitting a virtual machine migration command or outputting a corresponding virtual machine migration suggestion to the VMM on which the determined virtual machines to be migrated are operated according to a virtual machine migration strategy and the data flow direction relation among the virtual machines in the cluster system; and a VMM unit is used for receiving the virtual machine migration command transmitted by the virtual machine deployment adjustment module and migrating the virtual machines to be migrated to a target service node. By the method and the equipment, communication flow among the virtual machines in the cluster system can be effectively reduced, and network burden is reduced.

The invention discloses a method and system for distributed computing and querying massive data in online analysis and processing. The method uses a cluster system to perform distributed pre-calculation and querying on data cubes. Based on the MapReduce framework, the present invention divides large-capacity data sets into blocks and distributes them to each node through MapReduce, then the Map task on the node calculates a corresponding local closed cube for each data block, and finally starts the Map task on different nodes Parallel query is performed on each local closed cube, and the Reduce task merges the metric values ​​obtained from the query. The present invention can simply and effectively perform pre-calculation and query on-line analysis and processing of large-capacity data, greatly compress the storage space of the data cube, and can quickly respond to user queries.

The invention provides an online log analysis method, which comprises the following steps of: acquiring a log file from each node of a cluster system on line; defining an event sequence and an event correlation rule according to the log file, counting unitary event sequences, searching out a frequent unitary event sequence set, and obtaining a binary frequent event sequence candidate set C (2); performing statistics on support degree counting and posterior degree counting of the candidate binary frequent event sequence in the C (2), and calculating the degree of confidence and posterior degree; generating a binary frequent event sequence set F (2) and a binary rule set R (2); repeating the process until F(k) and R(k) are obtained, performing join operation on k-element frequent event sequences in the F(k), and generating an F(k+1)-element frequent event sequence set; performing approximate calculation on attribute values of elements which are not in R(k+1) but in F(k+1), until all frequent event sequences are generated; finding the event correlation rule according to the frequent event sequences; and predicting events in the log according to the event correlation rule on line.

The present invention provides a cluster load balancing method transparent to an operation system. Main functional modules comprise a load balancing module, a processor migrating module and a communication module. The method is characterized by comprising the following steps: 1, driving a virtual processor to migrate; 2, driving balance migration; 3, sending a migrating request to a target node and negotiating; 4, storing and restoring a state of the virtual processor; and 5, communicating. The method better solves the problem of low resource utilization rate of a cluster system. Along with the development of more popularization of the cluster system and the continuous development of hardware virtualization technology in the future, the method can be a good solution for the low resource utilization rate of the cluster system and has good application prospect.

A system and method for clustering user contexts in a customer self service system that performs resource search and selection. The customer self service system includes a context attribute database comprising types of user contexts and one or more context attributes associated with each user context for processing by the system, and context attribute function database comprising functions for computing values for each context attribute. The context clustering system receives user interaction data from among a database of user interaction records and a distance metric for associating closeness of the user interaction data and clusters the user interaction data according to the distance metric to determine new user contexts and associated attributes for use in subsequent resource searches initiated by users in the system. Improved user query definition and resource lookup results from the new determined user context attributes. The user interaction data comprises past and present user queries, system responses to user queries, raw context information including: one or more of static, historical context, transient context, organizational context, community context, and environment context, and, other raw context associated with the user and dependent upon that user's interaction state and customer query domain.