32 results about "Delayed binding" patented technology

A content-aware application switch and methods thereof intelligently switch client packets to one server among a group of servers in a server farm. The switch uses Layer 7 or application content parsed from a packet to help select the server and to schedule the transmitting of the packet to the server. This enables refined load-balancing and Quality-of-Service control tailored to the application being switched. In another aspect of the invention, a slow-start server selection method assigned an initially boosted server load metric to a server newly added to the group of servers under load balancing. This alleviates the problem of the new server being swamped initially due to a very low load metric compared to that of others. In yet another aspect of the invention, a switching method dependent on Layer 7 content avoids delayed binding in a new TCP session. Layer 7 content is not available during the initial handshaking phase of a new TCP session. The method uses the Layer 7 content from a previous session as an estimate to help select the server and uses a default priority to scheduling the transmitting of the handshaking packets. Updated Layer 7 content available after the handshaking phase is then used to reset the priority for the transmit schedule and becomes available for use in load balancing of the next TCP session.

A method and system for transporting traffic having disparate qualities of service classes across a packet-switched network includes receiving at an ingress node of a network a plurality of packets each comprising a quality of service (QoS) class defined externally to the network. Packets having a QoS class comprising delay bound guarantees and a low drop priority are combined into a first internal QoS class. Packets having a QoS class comprising a flexible drop priority and no delay bound guarantees are combined into a second internal QoS class. Packets having a QoS class including no delivery guarantees are combined into a third internal QoS class. The packets are transmitted in the network based on their internal QoS class.

A radio link protocol for a communications system ensures that delivery of Internet protocol data packets occurs within a set delay bound for the packets, in order to satisfy specified quality of service levels. Data packets arriving at a transmitter are subdivided into data blocks. As each block is transmitted, the transmitter starts an associated acknowledgement timer. The timer is turned off, before it expires, if the transmitter timely receives from a receiver a message informing the transmitter that the associated block was successfully received. If such message is not received, the timer expires and the transmitter sends an acknowledgement request signal to the receiver and starts an associated panic timer. The panic timer is turned off, before it expires, if the transmitter subsequently timely receives a message that the associated block was successfully received. If such message not received, the panic timer expires and the transmitter sends one of more copies of the corresponding block to the receiver before occurrence of the delay bound. If the transmitter receives a negative acknowledgement message from the receiver, that a block is missing or corrupted, the transmitter retransmits a copy of the block to the receiver. To reduce messaging traffic, the transmitter cancels acknowledgement and panic timers based upon information contained in negative acknowledgement messages, and the receiver can periodically send acknowledgement messages to inform the transmitter of successfully received blocks.

A method and system for providing dynamic ARQ, on a per packet basis, in a wireless telecommunications access network. By taking into consideration both per packet QoS based on, for example, packet delay bound, and potential delays caused by retransmission, a dynamic NAK-based ARQ scheme is presented that dynamically adapts the retransmission parameters to the per packet QoS to provide improved radio link quality, particularly for non-delay sensitive services and applications. The ARQ parameters that are dynamically adjusted can include the number of retransmission rounds, and the number of retransmissions in each round.

The uplink data rate is dynamically adjusted in a mobile station according to the mobile station's ability to meet QoS (quality of service) requirements of applications executing on the mobile station. In an embodiment, the transition probabilities received from a mobile network are first compensated according to the ability to meet QoS requirements, and the uplink data rate is then computed based on the compensated transition probabilities. According to another aspect of the present invention, the base station dynamically adjusts the transition probabilities based on delay bound requirement of packets and/or average delay or data flow rate requirement.

A method and system for providing delay bound and prioritized packet dropping are disclosed. The system limits the size of a queue configured to deliver packets in FIFO order by a threshold based on a specified delay bound. Received packets are queued if the threshold is not exceeded. If the threshold is exceeded, a packet having a precedence level less than that of the precedence level of the received packet is dropped. If all packets in the queue have a precedence level greater than that of the packet received, then the received packet is dropped if the threshold is exceeded.

Methods, systems, and media to reduce delays associated with port assignments are disclosed. More specifically, embodiments include hardware and/or software to delay binding of selected port numbers to instances of applications. Some embodiments identify port numbers for which applications generate multiple bind calls that may connect with unique remote ports. The port numbers are stored in a port list that is accessible to the protocol stack and memory is allocated for storing flags and additional port configuration data. When a bind call is associated with a port number bound to another application instance and the port number is included in the port list, the bind call may then be delayed until a connect is received from the corresponding application instance. When the resulting four tuple is unique for the computer system, the corresponding application instance is bound to the port number and connected with the remote port.

An apparatus and method for DBWRR (Delay Bound Weighted Round Robin) cell scheduling in an ATM (Asynchronous Transfer Mode) switch. More particularly, the present invention provides an apparatus and method for DBWRR cell scheduling in a high-speed ATM switch which can meet requirements for a cell transfer delay of real-time traffic in the ATM switch and minimize a processing overhead of the switch.

A method for performing congestion control in a communication system. The method includes transmitting a request message to a node using a first packet with a highest priority and receiving a response message corresponding to the request message using the first packet and at least one of a plurality of second packets, estimating a one-way delay of the first packet according to a round trip time for the first packet, converting one of the one-way delay of the first packet and the one-way delay of at least one of the plurality of second packets into a quality of service (QoS) level using a predetermined upper delay bound for a corresponding packet, and converting the converted QoS level into an effective congestion level representing a congestion degree in a network and performing congestion control using the effective congestion level.

The invention discloses a method for a wireless device to estimate an end-to-end delay probability characteristic in a multi-hop wireless network, which comprises the following steps that: each wireless device in the multi-hop wireless network monitors the performance of a local link to obtain own probability parameter gamma indicating that a data packet of the wireless device is directly served so that the wait delay of the wireless device is 0, and a descent speed parameter theta in an exponential distribution; secondly, each wireless device broadcasts the own probability parameter gamma and the descent speed parameter theta in the multi-hop wireless network so that each wireless device establishes a parameter comparison table of every wireless device in the multi-hop wireless network to the probability parameter gamma and the descent speed parameter theta of the wireless device; and finally, each wireless device calculates the end-to-end exceeding delay bound probability of a data packet of the wireless device to be transmitted according to the own established parameter comparison table and a route of the own data packet to be transmitted to another wireless device. Through simulation comparison, the method disclosed by the invention is well fitted with theoretical analysis results, and a completely perfect explanation can be given to the end-to-end delay probability characteristic in the multi-hop wireless network.

The invention discloses a DCS communication loose coupling management method. A communication management server builds a DCS communication management middleware and re-defines a communication path; a communication management server side registers the communication path; receiving-transmitting equipment receives registration information via a communication proxy to perform instantiation; when starting, the receiving-transmitting equipment firstly inquiries whether a remote communication management server has update content via the communication proxy of the equipment, if the update content exists, a library update process is performed, delayed binding initialization is performed after instantiation again, if the update content does not exist, delayed binding initialization; an instance which has completed delayed binding is activated; a communication link is built for transmitting or receiving communication information. According to the method provided by the invention, resetting and debugging complexity degrees caused by modification and upgrade of the system are reduced, so that the communication process of the DCS is automatic, openness of the DCS system is kept and improved, and complexity of a heterogonous software and hardware platform of the distributed system is reduced.

A system for scheduling packets to be transmitted is provided. The system includes a soft delay bound calculator module and a frame determination module. The soft delay bound calculator module calculates a soft delay bound for a non-real-time packet based on a packet size of the non-real-time packet and a minimum reserved traffic rate of a channel. The frame determination module determines whether a real-time packet must be transmitted at a current frame according to a delay bound, a transmission time, and a possible retransmission time thereof, and whether a non-real-time packet must be transmitted at a current frame according to a soft delay bound, a transmission time, and a possible retransmission time thereof. Thus, it is possible to improve the performance of the system while keeping the QoS thereof in a mixed service environment.

Curve servo is generated on a master medium by determining a series of offsets between the radial alignment reference and the desired curve servo sector, and expressing each offset as a set delay for master writer control sector chart. Incorporating delays into the master writer allows data mapping to be employed to generate curved servo sector patterns. In this way, a curved servo sector pattern with desired resolution is produced without increasing computational overhead or increasing memory requirements, or reducing yield. The delay may be transformed into control settings such as sector delay, clock cycle delay, and residual fractional delay settings each representing a fraction of the delay. In some embodiments, a piecewise linear approximation of the curve shape can be implemented to reduce the amount of control data required, and a master write controller that generates multiple laser pulses can be implemented.