International Business Machines Corporation

Jia Ming Zhang

Andrew Gary Hourselt

Ping Li

Hui Calvin Yang

Min Cai

Elena Schneider

Pei Hu

Brad J Behle

Ashaki A R Reid

A method for scheduling execution of an application is provided. The method comprises monitoring a usage state of resources of a first virtual machine that executes the application, so as to determine whether the usage state reaches a predetermined state. The method further comprises migrating an application module consuming the resources to a second virtual machine having corresponding resources, if the usage state reaches a predetermined state.

Policy-based workload performance optimization for dynamically distributed OSGi application

A plurality of network addresses identify a source device in the network. A plurality of subscriptions for the source device to receive messages from the target devices each identify one of the target devices and one of the source network addresses the identified target device uses to communicate with the source device. Upon detecting a change to a changed network address, a determination is made of a subscription indicating the changed network address and the target device indicated in the determined subscription is determined. A selection is made of a selected network address, other than the changed network address is selected. If the selected network address can be used to communicate between the determined target device and the source device, then the selected network address is used and provided to the determined target device.

Determining a network address for managed devices to use to communicate with manager server in response to a change in a currently used network address

A method for preventing a browser timeout resulting from multiple requests sent from a client to a server is provided. In response to receiving a request from the client, an application thread is initialized on the server for submitting data to the client. The application thread is observed using a daemon observation thread. If a loading time of the application thread exceeds a timeout, the application thread is terminated. In response to terminating the application thread, an empty response is sent to the client. In response to receiving the empty response to the client, a determination is made whether to make an additional request to the server.

Prevention of browser timeout

Growing community of inventors

Shanghai, China

Jia Ming Zhang