Abstract: A method may include various operations and a computer program product may cause a processor to perform various operations. The operations include receiving a task to be performed by a software application including numerous processing services, assigning a task identifier to the task before any processing service processes the task, and storing, for each processing service that processes the task, the task identifier in a record that associates the processing service identity with the task identifier in response to the processing service obtaining the task. Optional operations may further include detecting that the task has experienced a service event, identifying each of the processing services that processed the task prior to the service event, and collecting a service data log from each of the identified processing services. A service data archive may be formed with the service data logs collected from each of the identified processing services.

Abstract: Effects of a physical memory fault are mitigated. In one example, to facilitate mitigation, memory is allocated to processing entities of a computing environment, such as applications, operating systems, or virtual machines, in a manner that minimizes impact to the computing environment in the event of a memory failure. Allocation includes using memory structure information, including, information regarding fault containment zones, to allocate memory to the processing entities. By allocating memory based on fault containment zones, a fault only affects a minimum number of processing entities.

Abstract: Effects of a physical memory fault are mitigated. In one example, to facilitate mitigation, memory is allocated to processing entities of a computing environment, such as applications, operating systems, or virtual machines, in a manner that minimizes impact to the computing environment in the event of a memory failure. Allocation includes using memory structure information, including, information regarding fault containment zones, to allocate memory to the processing entities. By allocating memory based on fault containment zones, a fault only affects a minimum number of processing entities.