Abstract: Embodiments include a technique for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives is provided. Techniques include calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server. Techniques also include calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server, sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

Abstract: Embodiments include a technique for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives is provided. Techniques include calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server. Techniques also include calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server, sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

Abstract: Embodiments include method, systems and computer program products for real-time data clustering using multiple representatives associated with the data cluster. In some embodiments, a data point may be received from a data source. A data cluster may be identified. A set of representatives associated with the data cluster may be obtained. The data point may be compared to the set of representatives associated with the data cluster. A determination may be made to add the data point to the data cluster. The data point may be associated with the data cluster. The set of representatives associated with the data cluster may be updated, where one of the set of representatives associated with the data cluster is replaced with the data point.

Abstract: Embodiments include a technique for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives is provided. Techniques include calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server. Techniques also include calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server, sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

Abstract: Embodiments include a technique for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives is provided. Techniques include calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server. Techniques also include calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server, sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

Abstract: Embodiments include method, systems and computer program products for real-time data clustering using multiple representatives associated with the data cluster. In some embodiments, a data point may be received from a data source. A data cluster may be identified. A set of representatives associated with the data cluster may be obtained. The data point may be compared to the set of representatives associated with the data cluster. A determination may be made to add the data point to the data cluster. The data point may be associated with the data cluster. The set of representatives associated with the data cluster may be updated, where one of the set of representatives associated with the data cluster is replaced with the data point.

Abstract: The present invention provides a method of preventing harmful variability and oscillation in weight based, dynamic load balancing environments. More specifically, the present invention relates to methods to effectively determine the magnitude of weight changes in dynamic load balancing environments based on the workload magnitude and server farm capacity. This method depends on the development of new multi-system characteristics such as a relative workload metric to characterize the workload of the system relative to the collective capacity of all of the systems to handle the workload.

Abstract: A method and apparatus for managing CPU resources of a logically partitioned computing environment without shared memory access. A logical partition needing additional resources sends a message requesting such resources to a central domain manager, which sends messages to other partitions in the same group requesting that they assess their ability to donate resources to the requesting partition. Upon receiving such assessment request, each logical partition assesses its ability to donate resources to the requesting partition and responds accordingly to the domain manager. If at least one partition responds that it can donate resources to the requesting partition, the domain manager sends a message to a selected donor partition requesting that it reconfigure itself to donate resources to the requesting partition.

Abstract: In datacenter environments, many copies of servicing components (application servers, http servers, etc) are used to handle larger loads. In these cases, incoming service requests typically go to a load balancer to be directed to the appropriate servicing component. Modern advances in technology, like the Server/Application State Protocol, have allowed load balancers to receive recommendations in the form of numerical weights to describe the best distribution for the incoming requests. The present invention provides a method for computing path oriented statistics that enable load balancing algorithms to transparently integrate downstream performance and resource usage statistics into load balancing weights.

Abstract: The present invention provides a method of preventing harmful variability and oscillation in weight based, dynamic load balancing environments. More specifically, the present invention relates to methods to effectively determine the magnitude of weight changes in dynamic load balancing environments based on the workload magnitude and server farm capacity. This method depends on the development of new multi-system characteristics such as a relative workload metric to characterize the workload of the system relative to the collective capacity of all of the systems to handle the workload.

Abstract: In datacenter environments, many copies of servicing components (application servers, http servers, etc) are used to handle larger loads. In these cases, incoming service requests typically go to a load balancer to be directed to the appropriate servicing component. Modern advances in technology, like the Server/Application State Protocol, have allowed load balancers to receive recommendations in the form of numerical weights to describe the best distribution for the incoming requests. The present invention provides a method for computing path oriented statistics that enable load balancing algorithms to transparently integrate downstream performance and resource usage statistics into load balancing weights.

Abstract: The present invention provides a method of preventing harmful variability and oscillation in weight based, dynamic load balancing environments. More specifically, the present invention relates to methods to effectively determine the magnitude of weight changes in dynamic load balancing environments based on the workload magnitude and server farm capacity. This method depends on the development of new multi-system characteristics such as a relative workload metric to characterize the workload of the system relative to the collective capacity of all of the systems to handle the workload.

Abstract: In datacenter environments, many copies of servicing components (application servers, http servers, etc) are used to handle larger loads. In these cases, incoming service requests typically go to a load balancer to be directed to the appropriate servicing component. Modern advances in technology, like the Server/Application State Protocol, have allowed load balancers to receive recommendations in the form of numerical weights to describe the best distribution for the incoming requests. The present invention provides a method for computing path oriented statistics that enable load balancing algorithms to transparently integrate downstream performance and resource usage statistics into load balancing weights.

Abstract: A method and apparatus for managing CPU resources of a logically partitioned computing environment without shared memory access. A logical partition needing additional resources sends a message requesting such resources to a central domain manager, which sends messages to other partitions in the same group requesting that they assess their ability to donate resources to the requesting partition. Upon receiving such assessment request, each logical partition assesses its ability to donate resources to the requesting partition and responds accordingly to the domain manager. If at least one partition responds that it can donate resources to the requesting partition, the domain manager sends a message to a selected donor partition requesting that it reconfigure itself to donate resources to the requesting partition.

Abstract: In datacenter environments, many copies of servicing components (application servers, http servers, etc) are used to handle larger loads. In these cases, incoming service requests typically go to a load balancer to be directed to the appropriate servicing component. Modern advances in technology, like the Server/Application State Protocol, have allowed load balancers to receive recommendations in the form of numerical weights to describe the best distribution for the incoming requests. The present invention provides a method for computing path oriented statistics that enable load balancing algorithms to transparently integrate downstream performance and resource usage statistics into load balancing weights.

Abstract: The present invention provides a method of preventing harmful variability and oscillation in weight based, dynamic load balancing environments. More specifically, the present invention relates to methods to effectively determine the magnitude of weight changes in dynamic load balancing environments based on the workload magnitude and server farm capacity. This method depends on the development of new multi-system characteristics such as a relative workload metric to characterize the workload of the system relative to the collective capacity of all of the systems to handle the workload.