International Business Machines Corporation

Alan Hollingshead

Philip Ivor Wakelin

Julian Charles Horn

Michael D Brooks

Mike Brooks

Russell I Wilson

Katherine Jane Susan Robinson

Mark John Pocock

A method for intermediary client reconnection to a preferred server in a high availability server cluster. The method includes monitoring a persistent connection of a logical connection to a preferred server to identify an unavailability of the preferred server, establishing a temporary persistent connection for the logical connection to an available server to replace an unavailable preferred server, prompting attempts to reconnect to the preferred server at intervals, and providing a connection switching pool and simultaneously re-establishing a persistent connection with the preferred server for the logical connection using the connection switching pool while terminating the temporary persistent connection to the available server. The connection switching pool and a main connection pool used by the persistent connection allow for simultaneously maintaining two connections to different servers in a same server cluster for the logical connection.

Intermediary client reconnection to a preferred server in a high availability server cluster

Systems, methods, and computer program products for the dynamic management of the capacity of connections to a server. One method includes receiving a request to change a first connection to a server and determining active connections to the server. The method also includes determining a corresponding current capacity of each active connection and determining a new capacity for the first connection and the active connections to the server. The method further includes determining that a spare capacity of the server is insufficient for the first connection. The method also includes allocating a capacity to the first connection based on the request. The method further includes adjusting the current capacity of the active connections, including reducing the current capacity of at least a portion of the active connections in proportion to the respective current capacity of the one or more active connections.

Dynamic connection capacity management

Systems, methods, and computer program products for the dynamic management of the capacity of long-lived connections to a server are provided. A request to change a first connection to a server is received. The server has a total capacity to process multiple concurrent requests from one or more clients over respective connections, and each connection is configured with a capacity for communicating concurrent client requests. One or more active connections to the server and corresponding current capacities are determined. A new capacity for one or more connections of the first connection and the active connections to the server is determined based on the request, the total capacity and the current capacities of the one or more active connections.

Controlling a server latency response is presented. One example comprises monitoring, at a server, a number of requests from a client that cause a failure response. The method then comprises, responsive to the number of requests meeting a predetermined requirement, delaying processing of the requests from the client by a predetermined delay time.

Controlling a server response latency

Dynamic management of concurrent client connections to a server is achieved by a server, in that the server monitors its load state and when the server's maximum permitted number of concurrent requests has been exceeded it reduces the clients' maximum numbers of permitted concurrent requests until such time as the overloading situation no longer persists. In this way, the capacity of each of multiple connections can be controlled by the server individually or as a group, wherein the control is dynamic, so that the client-server connections do not have to be accurately configured in advance.

Server connection capacity management

Methods, systems, and computer program products for handling multi-pipe connections is provided. A primary pipe request for a connection between a client and a server is received. The connection includes a plurality of pipes between the client and the server. The primary pipe request includes a number corresponding to the amount of pipes required for the connection. A first response acknowledging the receipt of the primary pipe request is transmitted. The first response includes a token. A plurality of subsequent pipe requests are received for the connection between the client and the server. Each subsequent pipe request includes the token. A subsequent response to each subsequent pipe request is transmitted. A count of the number of pipe requests received is maintained. The established pipes are released after a period of time has elapsed when the number of pipe requests received doesn't reach the number of pipes required for the connection.

Handling multi-pipe connections

A computer-implemented method for data communication. In one example method, a first system sends an enhanced capability exchange (CAPEX) request message to a second system. The CAPEX request message includes a request to change the number of connection pipes on an established socket-based connection between the first system and the second system. The first system receives an enhanced CAPEX response message from the second system. The CAPEX response message accepts the request to change the number of connection pipes on the established connection. The first system changes the number of connection pipes on the established connection in accordance with the accepted request.

Method for managing connections for data communications

A transaction processing server and associated clients which send requests to the server via client-server connections. The server has a transaction processing region operable to process requests and issue responses to the requesting client. The server maintains a record of historical processing times taken to process recent client requests. The server has a connection manager that is operable to monitor incoming client requests and extract from them a goal response time, which is a maximum time within which the client expects the request to be processed which is included in each request. The server accepts or rejects each incoming client request based on deciding whether it is likely to be processed within the goal response time. If a request is rejected, the server transmits a rejection message to the originating client.

Server request management

A system for data communication. In one example method, a first system sends an enhanced capability exchange (CAPEX) request message to a second system. The CAPEX request message includes a request to change the number of connection pipes on an established socket-based connection between the first system and the second system. The first system receives an enhanced CAPEX response message from the second system. The CAPEX response message accepts the request to change the number of connection pipes on the established connection. The first system changes the number of connection pipes on the established connection in accordance with the accepted request.

Managing connections for data communications

An apparatus, computer-implemented method and computer program product manages connections for data communication. A first system initiates a process for establishing a new connection pipe of a socket-based connection between the first system and a second system. The new connection pipe is a replacement for an existing connection pipe. The existing connection pipe is accessed by a respective existing socket at each of the first and second systems. The first system receives acknowledgement from the second system confirming that the replacement connection pipe is established. The replacement connection pipe is accessed by a respective replacement socket at each of the first and second systems. The first system re-associates sessions of the existing socket at the first system to the replacement socket at the first system.

Managing connections for data communications following socket failure

A system, computer-implemented method and computer program performs heartbeat messaging for managing connections for data communications. In one example method, an indication of a problem associated with a connection for data communication in a computing system or network is received. In response to receiving the indication, a heartbeat message is sent over the connection. The method monitors for a heartbeat response to the heartbeat message within a predefined heartbeat time interval. If a heartbeat response is received within the predefined heartbeat time interval, the method determines that the indication of a problem associated with the connection is invalid. If a heartbeat response is not received within the predefined heartbeat time interval, the method determines that the indication of a problem associated with the connection is valid.

Growing community of inventors

Eastleigh, United Kingdom

Alan Hollingshead