341 results about "Slow-start" 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 equipment for accelerating throughput of communication segments from a sender (30) to a receiver (40) each of which include a protocol layer (30a 40a) for providing such segments, the method in case of TCP including a first step (20a) in which the sender protocol layer (30a) initializes a congestion window (cwnd) to some initial size (iwnd) and increases the congestion window by one segment each time it receives an ACK, but upon receiving an indication of low congestion, performs a step (20c) of accelerated start in which it sets a slow start threshold (SSTHRES) to a standard initial value (typically 65535) and re-initializes cwnd to a new value, and the sender protocol layer (30a) then increments cwnd at one or another predetermined rate—depending on whether the start immediately follows a new connection or instead follows a transfer of an existing connection to a new path—in respect to received ACKs.

The fast dynamic measurement of bandwidth in a TCP network environment utilizes a single pair of packets to calculate bandwidth between two entities on a network (such as the Internet). This calculation is based upon the packet-pair technique. This bandwidth measurement is extremely quick. On its journey across a network, communication devices may delay the packet pairs. In particular, TCP networks have two algorithms designed to delay some packets with the goal of increasing the overall throughput of the network. However, these algorithms effectively delay a packet pair designed to measure bandwidth. Therefore, they distort the measurement. These algorithms are Nagle and Slow Start. The fast dynamic measurement of bandwidth implements countermeasures to overcome the delays imposed by these algorithms. Such countermeasures include disabling the application of the Nagle Algorithm; minimizing the buffering of packets by sending a “push” packet right after the packet pair; and avoiding the Slow Start Algorithm by priming it with a dummy packet.

A novel approach is provided for avoiding the “slow-start restart” in TCP communications when network conditions dictate such. The disclosed approach serves to overcome the latency related to the “slow-start restart” by assessing network conditions and avoiding the “slow-start restart” when network conditions justify the avoidance. The disclosed methods, applications and devices implement the periodic communication of a probe packet that is generated and transmitted from a TCP sender to a TCP receiver during the period that the TCP network connection remains idle. Receipt of the probe packet by the TCP receiver will trigger transmission of an acknowledgment message by the TCP receiver. Upon receipt of the acknowledgement message the TCP sender will restart measurement of the idleness period and, as such the “slow-start restart” is avoided.

In a TCP / IP network, congestion control techniques such as slow start and congestion avoidance are employed. Such networks include wired and wireless links. However, normal operation of the wireless links exhibit different latencies than those exhibited over the wired link. The protocols employed in the wired network do not lend themselves well to efficient communication over wireless connections, and can cause slow start to be triggered. Determining when a sender will timeout due to non-receipt of an ACK, and intervening with a suppression message having an advertised window of zero to pause the user, are employed to prevent congestion control mechanisms such as slow start and congestion avoidance from activation.

A transport layer protocol such as the Stream Control Transmission Protocol uses a new traffic control technique instead of the TCP slow start algorithm. The procedure assumes that the network on which it is implemented has a fixed bandwidth assigned for the connection, and that the allotted bandwidth roughly matches the traffic load. Based on this, under message loss conditions it is only necessary to ensure that signaling traffic emitted into the network by the sender is no greater than the fixed bandwidth that has been allocated to the connection. That is, retransmissions take bandwidth away from a fixed allocation that has been made for the connection, but do not cause the connection itself to reduce the overall traffic it generates into the network; rather, it maintains the same traffic level. This technique prevents congestion in the network from increasing when message loss occurs; at the same time it does not reduce bandwidth for the association as rapidly as the slow start procedure.