Hewlett-packard Development Company, L.P.

Hewlett-packard Company

Scott Alan Rhine

Sharad Singhal

Xiaoyun Zhu

Francisco J Romero

Hyun Duk Kim

Michey N Mehta

David M Bernardo

Cliff McCarthy

Virtual machine threads are mapped to virtual processors. Groups of virtual processors are formed. The number of virtual processors in each group is selected to match a number of unassigned physical processors in a proximity. The virtual processors of each group are selected so that the total estimated interactivity for the group is significantly above average for groups having the same number of those virtual processors.

Proximity mapping of virtual-machine threads to processors

According to one embodiment, a request is received for executing a process on a computer. Information is accessed for a plurality of cells associated with the computer. A cell is selected that is the preferred candidate among the plurality of cells for executing the process using memory, which resides on the selected cell, based, at least in part, on the accessed information and the request.

Selecting a cell that is a preferred candidate for executing a process

In one embodiment, a method comprises (i) identifying a group associated with an executable that is using a resource of a computer system, (ii) decrementing a group utilization limit for the resource when the group utilization limit is greater than zero, (iii) decrementing a utilization reserve for the group when the group utilization limit for the resource equals zero, wherein operations (i)-(iii) are performed by a software routine responsive to system interrupts, and (iv) scheduling another executable to use the resource, wherein the scheduling verifies that (a) the another executable belongs to a group that has a non-zero group utilization limit for the resource or (b) the another executable belongs to a group that has a non-zero utilization reserve.

Controlling computer resource utilization

A system and method is provided for assigning a plurality of executable processes to a plurality of physical processors in a multi-processor computer system using a minimum processor share and a maximum processor share defined for each executable process. In an embodiment, the method can include allocating shares of total processor time to each executable process in proportion to the minimum processor shares up to the maximum processor shares to form target share allocations. The target share allocations can be used to map processes to the physical processors.

Process assignment to physical processors using minimum and maximum processor shares

In an embodiment, a computing system comprises a plurality of resources, a first manager process for allocating the plurality of resources on a dynamic basis according to service level parameters, and a plurality of computing domains, wherein at least one application, a respective second manager process, and a respective performance monitor process are executed within each computing domain, and wherein the performance monitor generates performance data related to the execution of the at least one application and the second manager process requests additional resources from the first manager process in response to analysis of performance data in view of at least one service level parameter.

System and method for allocating a plurality of resources between a plurality of computing domains

In one embodiment, a computer system comprises a plurality of processors, first software code for registering executables, creating partitions, and assigning executables to partitions, wherein the first software code assigns non-registered executables to a first partition, second software code for monitoring a utilization rate associated with the first partition, examining executables of the first partition when the utilization rate exceeds a first limit, and transferring executables having accessed an amount of resources greater than a second limit to a second partition.

Computer system and method for transferring executables between partitions

A computer-implemented method for distributing processes among a plurality of resources is disclosed. The processes are associated with a plurality of priority groups. The method includes ascertaining whether a first priority group of the plurality of priority groups is rich, the first priority group of the plurality of groups being deemed rich if the first priority group is entitled to at least a whole one of the plurality of resources, the first priority group of the plurality of groups being deemed poor if the first priority group is entitled to less than the whole one of the plurality of resources. The method also includes distributing processes of the first priority group among at least a first subset of the plurality of resources in accordance with a first computer-implemented methodology if the first priority group is deemed a rich. The method additionally includes distributing processes of the first priority group among at least a second subset of the plurality of resources in accordance with a second computer-implemented methodology if the first priority group is deemed poor, the second computer-implemented methodology being different from the first computer-implemented methodology.

Systems and methods for facilitating fair and efficient scheduling of processes among multiple resources in a computer system

A computer-implemented method for filtering, responsive to a logical specification of a network farm and a communication model for the farm, a tree topology network specification to obtain a filtered tree topology network specification is disclosed. The filtered tree topology network specification has fewer devices therein relative to the tree topology network specification, devices needed for constructing the farm being available in the filtered tree topology network specification. The method includes removing a first set of devices and a first set of switching elements from the tree topology network specification starting at the lowest level of the tree topology network specification and working toward the root node of the tree topology network specification. Each device in the first set of devices represents one of a redundant device, a device having a type unneeded by the farm, and a device unavailable for constructing the farm. Each switching element in the first set of switching elements represents a switching element whose children devices are all members of the first set of devices.

Need-based filtering for rapid selection of devices in a tree topology network

Methods and apparatus for creating a network farm from a tree topology network are disclosed. One method includes receiving a logical specification of the farm, the logical specification specifying at least one of device types required by the farm and a number of devices for each of the device types. The method further includes receiving a communication model for the farm, the communication model specifying expected bandwidth usage for devices of the farm. The method additionally includes ascertaining, responsive to the logical specification and the communication model, a solution set of devices among devices of the tree topology network for constructing the farm. The ascertaining includes performing partitioned greedy recursion on switching elements of the tree topology network, which includes recursively ascertaining, starting from the top level of the tree topology network and recursing toward the lowest switching element level of the tree topology network, the lowest cost solution at each switching element level of the tree topology network, wherein the lowest cost solution that completely satisfies resource requirements of the farm and that is found after the tree topology network is traversed by the partitioned greedy recursion is deemed the solution set of devices.

Systems and methods for rapid selection of devices in a tree topology network

A retrieval system and method which allows a computer system to adjust the data retrieval start times of the data retrieval requests from the different users in order to achieve overall system objectives. Each user is assigned a percentage availability which then determines a factor for all of the applications run by that user. This factor is associated with each request for data retrieval from any application. For any given batch of data requests there is calculated a start time for each data request, with the start time dependent upon the assigned factor. The start times from all users for that batch are then ordered in accordance with the calculated start times. The system will reorder the start times when efficient operation requires such reordering.

Growing community of inventors

Frisco, TX, United States of America

Scott Alan Rhine