Abstract: An apparatus and method provide client-side routing of database requests to enhance the performance of a partitioned database system. In a first embodiment, the database client determines the correct database server for a transaction, and routes requests for the transaction to the correct database server. In a second embodiment, the database client sends a first database request in a transaction to a selected database server, determines from the response the correct database server, and routes all subsequent requests for the transaction to the correct database server. In a third embodiment, the database client sends a first database request in a transaction to a selected database server, which routes the first request to the correct database server, which returns the results directly to the database client via direct connection. The database client then routes all subsequent database requests in the transaction directly to the correct database server.

Abstract: An apparatus, program product and method utilize distributed coordinating members to distribute member-specific state data to subscribing members in a distributed computing system. Member-specific state data is associated with different subjects, and coordinating members are configured to manage state data associated with particular subjects such that a reporting member that needs to report member-specific state data associated with a particular subject forwards that state data to a coordinating member associated with that subject. Once received, the coordinating member then distributes the member-specific state data to any members that have subscribed to receive state data associated with that subject.

Abstract: A method, apparatus, system, and signal-bearing medium that in an embodiment determine the active partitions of an application and assign the active partitions among servers via a routing table. The routing table is sent to a client, who uses the routing table to send requests and/or messages to the active partitions. In an embodiment, the routing table includes identifiers of the active partitions and the associated partitions. If one of the servers fails or if the partitions have a loading imbalance, the partitions are reassigned via the routing table.

Abstract: A method, apparatus, system, and signal-bearing medium that in an embodiment switch between blocking I/O and non-blocking I/O based on the number of concurrent connections. If the number of concurrent connections is greater than a high threshold, then blocking I/O is switched to non-blocking I/O. If the number of concurrent connections is less than a low threshold, then non-blocking I/O is switched to blocking I/O. In this way, I/O may be optimized depending on the number of concurrent connections, which increases performance.

Abstract: In a Java-based environment, a dynamic proxy runtime provides a client with the ability to asynchronously invoke an interface method, thereby unblocking the client and allowing it to perform other tasks while waiting for results to be returned from the runtime. The original interface is not redefined nor is the real target object implementation modified. Rather, the runtime defines rules which the client employs to define an additional asynchronous interface to supplement the original interface.

Abstract: A method, apparatus, system, and signal-bearing medium that in an embodiment enforce ordering of messages sent from a queue to clients. If a total order indicator is on for a queue associated with a get message request, the next message is sent from the queue to the client if the queue does not have an associated in-doubt transaction. An in-doubt transaction may be a transaction for which the client has not received a commit request. In another embodiment, an authorized client is selected and messages are only sent from the queue to the authorized client.

Abstract: Methods, articles of manufacture, and systems are provided that may be utilized to increase the rate in which orders may be processed for a given security by balancing the trading load over a number of local books. For example, when a monitored amount of order volume for a book reaches a threshold amount, an additional book may be created and the order volume may be distributed among multiple books. As a result, multiple processors may be used to process the orders locally (e.g., at the same exchange), thus increasing order processing bandwidth. When the order volume for the security declines, books (e.g.

Abstract: In a preferred aspect of the present invention, a method for augmenting a hierarchy of layered applications and corresponding protocols can include applying a discrimination algorithm to a selection process in which a particular application/protocol layer in a listing of adjacent application/protocol layers is selected to receive traffic flowing through the hierarchy. A new application/protocol layer can be inserted adjacent to the particular application/protocol layer in the hierarchy. Also, the new application/protocol layer can be added to the listing. Finally, the discrimination algorithm can be replaced with another discrimination algorithm programmed to consider the new application/protocol layer during the selection process. Significantly, each of the steps of performing the inserting, adding and replacing steps can be performed without decoupling or disabling other applications and protocols in the hierarchy.