Abstract: A method, apparatus, system, and signal-bearing medium that, in an embodiment, determine a cipher strength based on a trust level associated with a request, create a unit of work based on the request, encrypt the unit of work into a message based on the cipher strength, and send the message to grid servers. In various embodiments, the trust level may be determined based on a security token associated with the request or based on a zone from which the request originates. In various embodiments, the request originates from a client that belongs to the zone or originates from one of the grid servers that belongs to the zone. In an embodiment, a request from a grid server may be associated with a response to a previous unit of work that the grid server executed.

Abstract: A method, apparatus, system, and signal-bearing medium that, in an embodiment, send units of work to grid executors, create training data based on the performance of the grid executors, and train a neural network via the training data. The training data includes pairs of input and output data, where the input data is the types of the units of work and the output data is the service strengths of the grid executors. Once the neural network has been trained, subsequent units of work have their grid executors selected by inputting the types of the units of work to the neural network and receiving a service strength from the neural network as output. The grid executors are then selected based on the output service strength from the neural network. In this way, in an embodiment, the grid performance may be increased.

Abstract: A method, apparatus, system, and signal-bearing medium that, in an embodiment, determine units of work in a page, join selected pairs of the units of work with lowest complexity metrics until all of the units of work have complexity metrics that are less than a threshold. The joining is subject to rules of a dependency graph, which indicates data dependency relationships of the units of work. The selected joined pairs are then encapsulated into a processing unit, which is sent to multiple grid servers. Responses from the grid servers are used to assemble dynamic content into the page. In an embodiment, the complexity metrics are then modified based on an exponential moving average of responses times of the units of work. In this way, in an embodiment, the performance of assembling the page may be increased.

Abstract: A method, apparatus, system, and signal-bearing medium that, in an embodiment, receive configuration data from a development environment, determine a component based on reading an identifier of the component from a list of all possible components, install the component to an application server if the component exists in the configuration data but is not already installed in the application server, and remove the component from the application server if the component is already installed in the application server but does not exist in the configuration data. A rule is determined for the component based on policy data, where the configuration data may include the policy data. Either the application server or the component is configured via the rule. In this way, components may be automatically installed and tuned in an application server.

Abstract: A server computer system includes a keep-alive autonomic adjustment mechanism that monitors current operating conditions and dynamically adjusts the keep-alive specification to optimize the keep-alives for the current operating conditions. The keep-alive autonomic adjustment mechanism monitors the total number of clients, and the average number of requests per client. The value of a keep-alive specification that determines how log to keep a connection open may be dynamically and autonomically adjusted by the keep-alive autonomic adjustment mechanism according to the monitored current operating conditions. The keep-alive autonomic adjustment mechanism may also include the ability to autonomically enable and disable the keep-alives. In this manner the performance of the server may be tuned to optimize the effect of using keep-alives according to current operating conditions.