Abstract: Embodiments for optimizing a backup operation for a data protection system, by connecting with network devices in a network implementing the backup operation, the backup operation implementing a policy that specifies a plurality of service requirements and that is initiated at a time within a plurality of time windows, and determining a size of dataset to be backed up by the backup operation. The method further monitors a flow of data between a source and a destination of data in the network, compiles historical data of operational parameters for data transfers between the source and destination, and determines an optimum time within the plurality of time windows based on the dataset size and the historical data.

Abstract: Aspects of the subject disclosure may include, for example, determining, an application requirement for a server of a data center geographically separated from a central location. A configuration file adapted for the server according to the application requirement is obtained at the central location and a host processor located at the data center is selected according to the application requirement. A static IP address preassigned to an administrative portion of the selected host processor is identified. The selected host processor is initialized remotely over a wide area network via the administrative port without utilizing any software resources pre-existing at the data center. The initializing configures the selected host processor according to the configuration file to service the application requirement. Other embodiments are disclosed.

Abstract: A method routes packets from a source to a destination across an IP network having a plurality of nodes (including the source and destination), and a plurality of network segments interconnecting the plurality of nodes. The source and destination are configured to use a given service. To those ends, the method receives information relating to the given service, and forms a path between the source and the destination. The path includes a) at least one intermediate node between the source and the destination and b) a plurality of specific network segments extending from the source to the destination. The plurality of specific network segments are a sub-set of the plurality of network segments. To form the path, the method assigns the plurality of specific network segments to the network path between the source and the destination as a function of the information relating to the given service.

Abstract: A method for automatically establishing an address-port mapping table of a switching device in an interconnect fabric uses hardware link-up and link-down processes to build and update the lowest cost (e.g., shortest path) port entries in the mapping table. Traffic loops are precluded by comparing cost values based on the source addresses of the devices in the interconnect fabric, without blocking any particular port.

Abstract: A method to broker events of event-driven application components, within a distributed computing environment and using a mesh broker, is described. The mesh broker is instantiated as several mesh agents, the mesh agents being provisioned to support mediation activities relating to a plurality of computational nodes within the distributed computing environment. The mesh agents are further deployed as a mesh network among the computational nodes of the distributed computing environment. A connectivity catalog stores cost data associated with transmission of an event notification between each of multiple pairs of computational nodes of the computational nodes. Routes across the mesh network are automatically selected, by the mesh agents and using the cost data to determine low-cost routes across the mesh network.

Abstract: Systems and methods are disclosed for determining shared risk link group (SRLG) disjoint paths in a communications network. An original graph representing the communications network can be obtained. The original graph can include vertices and edges corresponding to nodes and communication links in the communication network. The vertices can include a source vertex and a target vertex and each edge can be associated with a set of SRLGs A reduced graph can be generated from the original graph. Generation of the reduced graph can include identification of a first edge of the edges as a dominating edge and removal of the first edge. Two SRLG-disjoint paths can then be identified on the reduced graph.

Abstract: Methods, systems, and apparatus, including computer-readable storage media, optimizing interior gateway protocol (IGP) metrics using reinforcement learning (RL) for a network domain. The system can receive a topology (G) of a network domain, a set of flows (F), and an objective function. The system can optimize, using reinforcement learning, the objective function based on the received topology and the one or more flows F. The system can determine updated IGP metrics based on the optimization of the objective function. The IGP metrics for the metric domain may be updated with the updated IGP metrics.

Abstract: A method of routing communication between nodes of a mesh network is provided. The method includes each node of the mesh network advertising a hop distance relative to a head of the mesh network, the hop distance being a measure of how many hops a node of the mesh network is from a head node of the mesh network, the head node being a node of the mesh network that controls all the other nodes of the mesh network.

Abstract: An embodiment generates, responsive to receiving a declarative replication specification, a logical map comprising a plurality of paths to completing a data replication process identified by the replication specification. The embodiment determines, responsive to generating the logical map, an optimal path to completing the data replication process by searching the plurality of paths of the logical map using a cost-directed search algorithm to find the optimal path from among the plurality of paths. The embodiment also performs, responsive to finding the optimal path, a series of data replication operations for replicating source data on a target database. The embodiment also detects, during a runtime of the data replication process, a predetermined change to a runtime environment of the data replication process. The embodiment also determines, responsive to detecting the predetermined change, a revised optimal path for performing the data replication process in the changed runtime environment.

Abstract: A system described herein may provide a technique for the generation of a node map using contraction hierarchy techniques in a manner that accounts for constraints that apply to nodes or links between nodes in the node map. Candidate shortcut links, that represent paths through multiple links in the node map, may be evaluated to determine whether such candidate shortcut links are less restrictive than less costly paths having the same starting and ending nodes as the candidate shortcut links. Candidate shortcut links may be added to the node map in situations where less costly paths are available, but are more restrictive than the shortcut links. At query time, the shortcut links may apply to certain queries having parameters that violate constraints associated with the less costly paths but do not violate any applicable constraints of the shortcut links.

Abstract: A method for fetching a content from a web server to a client device is disclosed, using tunnel devices serving as intermediate devices. The client device accesses an acceleration server to receive a list of available tunnel devices. The requested content is partitioned into slices, and the client device sends a request for the slices to the available tunnel devices. The tunnel devices in turn fetch the slices from the data server, and send the slices to the client device, where the content is reconstructed from the received slices. A client device may also serve as a tunnel device, serving as an intermediate device to other client devices. Similarly, a tunnel device may also serve as a client device for fetching content from a data server. The selection of tunnel devices to be used by a client device may be in the acceleration server, in the client device, or in both.

Abstract: A system and method for diameter agent load balancing. The method including: receiving a request from a sending diameter node; parsing at least one Attribute-Value Pair (AVP) from the request; determining a partition-identifier (partition-id) from the at least one AVP; determining a receiving diameter node, based on the partition-id; and sending the request to the receiving diameter node. The system including: a message module configured to receive a request from a sending diameter node; a parsing module configured to parse at least one Attribute-Value Pair (AVP) of the message from the request and determine a partition-id from the at least one AVP and a receiving diameter node, based on the partition-id; and a forwarding module configured to send the request to the receiving diameter node.

Abstract: A network device may obtain information concerning a virtual chassis that indicates that the network device and an additional network device are to be included in the virtual chassis. The network device may determine, based on the information concerning the virtual chassis, that the network device is connected to the additional network device, wherein the network device is connected to the additional network device via a link between a network interface of the network device and a network interface of the additional network device. The network device may cause the network interface of the network device to be converted to a virtual chassis interface and the network interface of the additional network device to be converted to a virtual chassis interface to enable the network device and the additional network device to be included in the virtual chassis to allow bootstrapping of the virtual chassis as a single logical device.

Abstract: Methods, apparatus, systems and articles of manufacture to build a storage architecture for graph data are disclosed herein. Disclosed example apparatus include a neighbor identifier to identify respective sets of neighboring vertices of a graph. The neighboring vertices included in the respective sets are adjacent to respective ones of a plurality of vertices of the graph and respective sets of neighboring vertices are represented as respective lists of neighboring vertex identifiers. The apparatus also includes an element creator to create, in a cache memory, an array of elements that are unpopulated. The array elements have lengths equal to a length of a cache line. In addition, the apparatus includes an element populater to populate the elements with neighboring vertex identifiers. Each of the elements store neighboring vertex identifiers of respective ones of the list of neighboring vertex identifiers.

Abstract: Systems, methods and/or computer program products managing momentary increases in resource requirements of microservices within a service mesh using sliced replicas or pre-configured cloud functions. The service mesh tracks incoming API calls of applications requesting use of specific microservice capabilities. Over time capabilities being called are categorized by the service mesh based on call frequency. Service mesh predicts patterns of expected increases in requests for each capability of the microservices. As the increase in the number of API calls for the capability become more than a threshold level predicted by the service mesh, a sliced replica of the microservice is created or a pre-configured cloud function is enabled to handle requests specifically for the capability causing the momentary increase in resource requirements, while the service mesh continues routing API calls for other capabilities to the requested microservice.

Abstract: A method of operating an access control system comprising a plurality of access controls, the method comprising: determining an energy metric of each of the plurality of access controls; determining a latency metric of each of the plurality of access controls; transmitting the energy metric of each of the plurality of access controls; transmitting the latency metric of each of the plurality of access controls; collecting the energy metric and the latency metric at a head node or collecting energy metric at each of the plurality of access controls from a 1-hop transmission distance; and determining a data route through the plurality of access controls in response to the energy metric of each of the plurality of access controls and the latency metric of each of the plurality of access controls.

Abstract: Techniques for improved routing based on network traffic are provided. Telemetry data relating to a first network node of a plurality of network nodes in a locator ID separation protocol (LISP) fabric is received. A first portion of the telemetry data that relates to a first destination of a plurality of destinations is identified. Further, a first routing weight associated with a first interface of the first network node is revised based on the first portion of the telemetry data, where the first interface is associated with the first destination. The revised first routing weight is published to a second plurality of network nodes in the LISP fabric, wherein the second plurality of network nodes route packets to the first network node based in part on the revised first routing weight.

Abstract: In a border network having a plurality of routers, routing tables for the routers can be generated using one or more weighted parameters, such as a price associated with transmission of packets over third-party routers, capacity of the border routers, and performance information (e.g., latency) of the border routers. The routing tables can be repetitively generated to proactively ensure that the border network is operating optimally. The framework described augments a control plane with Availability, Performance, and Cost (APC) telemetry collected from existing services in a cloud environment. Such telemetry can be continuously correlated in a convex optimization framework to maintain the network in its optimal state. As a result, the framework can proactively avoid potential network degradation events (e.g. congestion) and the resulting customer impact, while reducing the cost of routing traffic through upstream peers/transits.

Abstract: In some implementations, a method is provided. The method includes receiving at an origin controller node, a task and determining a destination work node where the task is to be executed, wherein the origin controller node and the destination work node each comprise a receptor node among a plurality of receptor nodes forming a receptor network, and each of the plurality of receptor nodes functions as one or more of a router node, a controller node, and a work node. The method further includes determining a first router node comprising a first hop on an optimal path to the destination work node, the first router node determined from among a set of receptor nodes functioning at least as router nodes. The method further includes transmitting the task to the destination work node via the first router node and performing the task via the destination work node.

Abstract: A method and system for processing media content in Moving Picture Experts Group (MPEG) Network Based Media Processing (NBMP) includes obtaining a list of function repositories, each function repository storing one or more functions for processing the media content, obtaining a descriptor identifying a location of each of the function repositories identified in the list of function repositories, selecting a function repository, from among the function repositories, according to a criterion, and processing the media content using a function included in the selected function repository.

Abstract: Aspects of the subject disclosure may include, for example, determining, an application requirement for a server of a data center geographically separated from a central location. A configuration file adapted for the server according to the application requirement is obtained at the central location and a host processor located at the data center is selected according to the application requirement. A static IP address preassigned to an administrative portion of the selected host processor is identified. The selected host processor is initialized remotely over a wide area network via the administrative port without utilizing any software resources pre-existing at the data center. The initializing configures the selected host processor according to the configuration file to service the application requirement. Other embodiments are disclosed.

Abstract: A method and an apparatus for processing data are provided. The method includes: determining, in response to receiving an access request sent by a client and a target path in a complete binary tree, a time interval between the access request and a last access request as a second time interval, a value of a non-leaf node of the complete binary tree being related to a time interval of latest two access requests received by a metadata server, a leaf node being used to represent a virtual node in a distributed system; determining a change amount of a value of a node in the complete binary tree according to the target path and the second time interval; querying an access result according to the access request, and sending the access result to the client; and sending the change amount to a management server end.

Abstract: Technologies for intelligent traffic optimization include a directional flow server that receives dynamic traffic data from traffic infrastructure devices in a monitored region. The traffic data may include traffic volume data and traffic control status data. The server updates a dynamic layer of a high-definition map based on the dynamic traffic data. The high-definition map also includes a static layer and a directional flow layer. The server optimizes the directional flow in response to updating the dynamic layer. The directional flow layer is indicative of traffic direction associated with roads in the monitored region. The server may optimize each of several sub-regions and then optimize connection roads between the regions. The server may distribute the optimized directional flow layer to consumers such as traffic control devices, autonomous vehicles, and other subscribing devices. Other embodiments are described and claimed.

Abstract: A method is performed by a first front-end node, for supporting data communication. The first front-end node operates in a communications network comprising a pool of front-end nodes for providing access to a database. The pool of front-end nodes includes the first front-end node. The first front-end node receives, from a load balancer node, a first request for data from the database, from a client node. The first front-end node provides, to the client node, a first response to the received first request, the first response including a first indication indicating that the first front-end node is a preferred node for providing a subsequent response to a subsequent request for data from the client node. The subsequent response is allowed to originate from another front-end node in the pool of front-end nodes, different than the preferred node.

Abstract: A method for accelerating transaction execution across geographically dispersed clusters is disclosed. In one embodiment, such a method includes receiving a request to read data in a data replication system comprising a primary storage system and a secondary storage system in a synchronous mirroring relationship with the primary storage system. The method determines whether at least one of the following conditions is true: (1) the secondary storage system is closer from a communication standpoint; and (2) a high speed communication technology is available to communicate with the secondary storage system that is not available to communicate with the primary storage system. In the event either of the conditions is true, the method directs the request to the secondary storage system. In the event neither of the conditions is true, the method directs the request to the primary storage system. A corresponding system and computer program product are also disclosed.

Abstract: A tool may identify and revert changes that caused network hardware components or hardware servers to malfunction. The tool builds and maintains a graph that represents the hardware components and servers in the system and their dependencies. When a change is made to the system, links and weights in the graph are adjusted to account for the changes. When a component or server is reported as malfunctioning, the tool traverses the graph to locate the changes that are the most likely root causes of the malfunction. The tool may then revert the change to resolve the malfunction.

Abstract: A system designed for increasing network communication speed for users, while lowering network congestion for content owners and ISPs. The system employs network elements including an acceleration server, clients, agents, and peers, where communication requests generated by applications are intercepted by the client on the same machine. The IP address of the server in the communication request is transmitted to the acceleration server, which provides a list of agents to use for this IP address. The communication request is sent to the agents. One or more of the agents respond with a list of peers that have previously seen some or all of the content which is the response to this request (after checking whether this data is still valid).

Abstract: Provided is a matching system including a policy and rule processing unit configured to extract SC units for each application service required, combine the SC units, configure a transaction unit, and determine an amount to be charged for a transaction according to a charging rule; a matching processing unit configured to match the extracted SC units to NC units; and an NC combination processing unit configured to combine the NC units used for the matching and deliver the combination of the NC units and detailed operating conditions to the networking management system.

Abstract: Embodiments of the present disclosure relate to a method and an apparatus for processing data. The method can include: determining, in response to receiving an access request, a time interval between the access request and a last access request as a first time interval; acquiring a preset complete binary tree from a management server end, a value of a non-leaf node of the complete binary tree being related to a time interval of latest two access requests received by a metadata server, a leaf node being used to represent a virtual node in a distributed system, and the virtual node corresponding to the metadata server; selecting a target path from the complete binary tree according to the first time interval; and sending the access request and the target path to a metadata server corresponding to a leaf node of the target path.

Abstract: The present teaching, which includes methods, systems and computer-readable media, relates to techniques to manage representation of a content topic in a content pool. The disclosed techniques may include determining availability of content related to the content topic based on a set of content items in the content pool, and determining a level of interest of a set of users in the content topic based at least on information related to interaction of the set of users with the set of content items. A level of representation of the content topic in the content pool may be determined based at least on the determined availability of content and the determined level of interest. Based on the level of representation, at least some of the set of content items related to the content topic may be updated, e.g., content items may be added to or removed from the content pool.

Abstract: A system designed for increasing network communication speed for users, while lowering network congestion for content owners and ISPs. The system employs network elements including an acceleration server, clients, agents, and peers, where communication requests generated by applications are intercepted by the client on the same machine. The IP address of the server in the communication request is transmitted to the acceleration server, which provides a list of agents to use for this IP address. The communication request is sent to the agents. One or more of the agents respond with a list of peers that have previously seen some or all of the content which is the response to this request (after checking whether this data is still valid).

Abstract: Systems and methods are provided for managed file transfer. A managed file transfer server may receive a request from a sender to send a file (bulk data) to a recipient and may determine a location server that is closest to the location of the recipient. A server-to-server transfer can be automatically initiated to move the file to the location server that is closest to the location of the recipient.

Abstract: An exemplary method comprises receiving data points, selecting a first subset of the data points to generate an initial set of landmarks, each data point of the first subset defining a landmark point and for each non-landmark data point: calculating first data point distances between a respective non-landmark data point and each landmark point of the initial set of landmarks, identifying a first shortest data point distance from among the first data point distances between the respective non-landmark data point and each landmark point of the initial set of landmarks, and storing the first shortest data point distance as a first landmark distance for the respective non-landmark data point. The method further comprising identifying a non-landmark data point with a longest first landmark distance in comparison with other first landmark distances and adding the identified non-landmark data point associated as a first landmark point to the initial set of landmarks.

Abstract: A method, computer program product, and computer system for defining, on at least one processor, one or more potential overflow paths for processing at least a portion of a data load through a plurality of stream operators deployed on a plurality of computing devices. Embodiments may include determining, on the at least one processor, a conditional availability for each computing device of the plurality of computing devices. Embodiments may also include identifying, on the at least one processor, a change in one or more of a configuration of the plurality of computing devices and the at least a portion of the data load. Embodiments may further include dynamically activating, on the at least one processor, one or more potential overflow paths in response to the identified change based upon, at least in part, the conditional availability for each computing device of the plurality of computing devices.

Abstract: A plurality of forwarding servers is communicatively coupled to a plurality of sending systems via a network. A first forwarding server associated with a first geographical region includes an interface that receives a first request for a resource from a first sending system via the network. A determination logic determines whether one or more requests that precede the first request are received from the first sending system at a second forwarding server associated with a second geographical region. A causation logic causes the first request to be forwarded to a second endpoint associated with the second geographical region rather than a first endpoint associated with the first geographical region at least partly based on an inference that the first sending system is located in the second geographical region.

Abstract: A method is provided in one example embodiment and may include generating an interest message to be transmitted using each face of a plurality of faces associated with a node, wherein the interest message is to be discarded by each of a neighboring node of a plurality of neighboring nodes that receive the interest message; determining, for each face of the plurality of faces associated with the node, a one-hop propagation delay associated with each face of the plurality of faces; and selecting a particular face through which to transmit another interest message based, at least in part, on the one-hop propagation delay of the particular face and a number of pending interests associated with the particular face.

Abstract: A communication analysis device updates statistical information on a content access based on the communication session log of a user. The content access is an aggregation of a plurality of communication sessions needed for browsing one website. The communication analysis device acquires session information pertaining to one communication session, and determines whether a lapsed time from the start time or end time of a base communication session to the start time or end time of the one communication session is equal to or less than a threshold value. When the lapsed time is equal to or less than the threshold value, the communication analysis device determines that the one communication session is included in the same content access as the base communication session, and updates the statistical information of the content access. The communication analysis device stochastically sets the threshold value based on a prescribed probability distribution.

Abstract: A system and method for efficiently allocating device resources is disclosed. A composer application is executed by a management controller. A fabric box includes fabric switches and a fabric controller. A network is coupled to the management controller and the fabric box. Host servers are each coupled to an upstream port of one of the fabric switches. Multiple device resources are coupled to downstream ports of the fabric switches. The fabric controller determines the path distances between each of the downstream ports and each of the upstream ports. The composer application selects a device based on the shortest path distance to a requested host server in response to an allocation command.

Abstract: This invention relates to the proposal of component carrier (de)activation message that is allowing a activation or deactivation of one or more component carriers in the uplink or downlink. Furthermore, the invention relates to the use of the new component carrier (de)activation message in methods for (de)activation of downlink component carrier(s) configured for a mobile terminal, a base station and a mobile terminal. To enable efficient and robust (de)activation of component carriers, the invention proposes to use component carrier-specific or cell-RNTI(s) for the scrambling of the CRC of the component carrier (de)activation message, and to explicitly indicate the intended recipient of the component carrier (de)activation message in a corresponding field in the message. Furthermore, the invention further proposes different designs of the component carrier (de)activation message and further uses thereof, so as to trigger CQI reporting and/or SRS transmission by a mobile terminal.

Abstract: Techniques are described for receiving manufacturing data and events over real time and non-real time interfaces and associating the data with one another. In one example, real time data is received, the real time data associated with a counter value assigned by a precision counter. The received real time data is stored in a storage buffer, and non-real time data is received, where the non-real time data associated with a counter value assigned by the precision counter. In response to receiving the non-real time data, the buffer is searched for real time data having a matching counter value and, in response to identifying stored real time data associated with a matching counter value, the non-real time data is associated with the real time data based on the match. Data packages are generated including related real time and non-real time data associated with matching counter values.

Abstract: Embodiments of the present disclosure include an Orchestrator to commission a point of presence (PoP) node based on receiving an internet protocol (IP) address, the IP address associated with a geographic location; and a global routing manager for assigning a client device to a PoP node based on the received IP address. Embodiments also include a performance manager to acquire performance information about each PoP node.

Abstract: A community safety system with inter-organizational communication, allowing for members of a first organization to provide alerts when located within a security zone associated with a second organization. A notification management entity, responsible for managing the different safety systems associated with each organization, determines where a user is located and whether the user is associated with any of the organizations responsible for the security zone in which the user is located.

Abstract: A web service request is received, and a stored performance metric value, generated for other web service requests of the same type, is compared to a threshold value. If the stored performance metric value meets the threshold value, then the received web service request is routed to a sandboxed instance of the web service for execution. If the stored performance metric value does not meet the threshold value then the received web service request is routed to a default web service instance for execution. In either case, the performance metric is monitored during execution of the received web service request, and the stored performance metric value is updated based upon the monitored performance metric generated for the received web service request.

Abstract: Automated reconfiguration of real time data stream processing may be implemented. A processing function that describes one or more operations to be performed with respect to one or more data streams may be executed at one or more processing nodes. Performance metrics describing the performance of the processing function at the processing nodes may be collected and monitored. A reconfiguration event may be detected for the processing function. A different execution configuration for the processing function may be determined and initiated in response to detecting the reconfiguration event.

Abstract: Methods, systems, and computer readable media for sharing identification information of network nodes in an IMS network are disclosed. One method includes determining, by a packet data network gateway (PGW), policy and charging rules function (PCRF) identification information corresponding to a PCRF node designated to support an IMS-based subscriber session requested by a user equipment device and establishing, by the PGW, a subscriber Gx session with the PCRF node corresponding to the determined PCRF identification information. The method further includes determining, by the PGW, application function (AF) identification information corresponding to an AF server designated to support the IMS-based subscriber session and generating, by the PGW, a create session response message including at least the PCRF identification information and the AF identification information. The method also includes sending, by the PGW, the create session response message containing the at least the PCRF identification inform.

Abstract: Provided is an opportunistic forwarding method and apparatus for a content-based information centric network in a delay tolerant network environment. The opportunistic forwarding method for a content-based information centric network in a delay tolerant network environment having a processor and a memory includes when a first node contacts a second node, updating a first delivery predictability and a second delivery predictability of the first node based on contact record information between the first node and the second node and data that the second node has, updating a first message of the first node by comparing the first message including first interest information and first data that the first node has with a second message including second interest information and second data that the second node has, and delivering the first message of the first node to the second node based on the first and second delivery predictabilities of the first node.

Abstract: Systems and methods are provided for managed file transfer. A managed file transfer server may receive a request from a sender to send a file (bulk data) to a recipient and may determine a location server that is closest to the location of the recipient. A server-to-server transfer can be automatically initiated to move the file to the location server that is closest to the location of the recipient.

Abstract: The present invention provides a registration management method capable of suppressing influence of data file forwarding load upon processing, and suitable for application to for example VNF Package on-boarding. A data file required for registration is placed in advance on a side of a management apparatus that performs registration processing, and at a registration stage, a registration request is made to the management apparatus by specifying the data file placed on the management apparatus side.

Abstract: A network system control method includes intercepting a flow modification message sent by a controller from a network protocol path between a switch and the controller so as to obtain a new flow entry; accessing a flow table of the switch so as to obtain a plurality of flow entries; inserting at least one redundant flow entry according to the new flow entry and the plurality of flow entries; performing an aggregation operation to the new flow entry, the plurality of flow entries and the at least one redundant flow entry so as to generate a set of aggregated flow entries; and updating the flow table using the set of aggregated flow entries.

Abstract: The invention provides a cloud-computing graphic server. In one embodiment, the cloud-computing graphic server is coupled to a client host via a network, and includes a plurality of back-end graphic servers and at least one front-end graphic server. The graphic server is coupled to the back-end graphic servers via a high-speed network, receives a request from the client host via the network, determines a plurality of application programs required by the request, and selects a plurality of used back-end graphic servers respectively corresponding to the application programs from the back-end graphic servers. The used back-end graphic servers execute the application programs according to instructions from the front-end graphic server to generate a plurality of graphic surfaces, and the front-end graphic server blends the graphic surfaces to obtain a windows surface datastream and sends the windows surface datastream back to the client host for display via the network.