公开(公告)号：US07225223B1

公开(公告)日：2007-05-29

申请号：US09665934

申请日：2000-09-20

申请人： Clifford A. McCarthy ,  Raja Daoud

发明人： Clifford A. McCarthy ,  Raja Daoud

IPC分类号： G06F15/16

CPC分类号： G06F9/5061 ,  G06F9/5005

摘要： A software method and computer system using said method are disclosed for efficiently reallocating unused system resources in a system in which unused system resources are initially allocated among groups that may become active or inactive. The method reallocates unused resources among active groups by scaling up these groups' initial entitlement shares, subject to maximum caps that may apply. For each group, the method calculates a scaling ratio equal to the cap divided by the entitlement, and then sorts the groups according their scaling ratios. The method processes each group in increasing order of their scaling ratios and reaches a maximal share of system resources to be allotted to each group, where the maximal share for an active group is between, or equal to, the group's initial entitlement value and its cap. Inactive groups have maximal shares equal to zero, reflecting that they receive no system resources. As the method processes the groups, it attempts to simply scale each active group by its scaling ratio. If scaling all unprocessed groups by the current group's scaling ratio would exceed the unallocated resources, then the remaining groups are scaled by the remaining resources divided by the sum of the entitlements of the remaining groups. Because the groups are sorted in order of increasing scaling ratios, the remaining active groups may be scaled as described without exceeding their caps.

摘要翻译： 公开了一种使用所述方法的软件方法和计算机系统，用于在系统中有效地重新分配未使用的系统资源，其中在可能变为活动或不活动的组中最初分配未使用的系统资源。 该方法通过按照可能适用的最大上限扩大这些组的初始权利份额，重新分配活动组中未使用的资源。 对于每个组，该方法计算等于上限除以权利的缩放比例，然后根据缩放比例对组进行排序。 该方法按照其缩放比例的递增顺序处理每个组，并达到要分配给每个组的系统资源的最大份额，其中活动组的最大份额在或等于组的初始授权价值和其上限 。 非活动组的最大份额等于零，反映它们不会收到系统资源。 随着该方法处理组，它尝试简单地按比例缩放每个活动组。 如果按照当前组的缩放比例缩放所有未处理的组，将超过未分配的资源，则剩余的组除以剩余组的权利总和除以剩余的组。 因为这些组按照递增缩放比例的顺序进行排序，所以剩余的活动组可按照描述进行缩放而不超过其上限。

公开(公告)号：US07143413B2

公开(公告)日：2006-11-28

申请号：US10144991

申请日：2002-05-15

申请人： Clifford A. McCarthy ,  Canming Jin

发明人： Clifford A. McCarthy ,  Canming Jin

IPC分类号： G06F9/46 ,  G06F15/173

CPC分类号： G06F9/5011

摘要： A method is disclosed for allocating computer system resources, such as memory, among applications. Input parameters are received for applications. The parameters include a minimum entitlement value, a maximum entitlement value, and a weight. The minimum entitlement is the minimum amount of resources that the application should receive. The maximum entitlement is the maximum amount of resources that the application should receive. The weight specifies the priority of the application relative to other applications, for purposes of distributing system resources. Computer system resources are then allocated among applications based on the parameters.

摘要翻译： 公开了一种在应用程序之间分配诸如存储器之类的计算机系统资源的方法。 接收到应用程序的输入参数。 参数包括最小权利值，最大权利值和权重。 最低权利是应用程序应该获得的资源的最小数量。 最大权利是应用程序应获得的最大资源量。 权重指定应用程序相对于其他应用程序的优先级，以便分发系统资源。 然后根据参数在应用程序中分配计算机系统资源。

公开(公告)号：US08336054B2

公开(公告)日：2012-12-18

申请号：US11489967

申请日：2006-07-20

申请人： Ludmila Cherkasova ,  Jerome Rolia ,  Clifford A. McCarthy

发明人： Ludmila Cherkasova ,  Jerome Rolia ,  Clifford A. McCarthy

IPC分类号： G06F9/46 ,  G06F15/173

CPC分类号： G06F9/4881

摘要： A method comprises defining a scheduler parameter for a maximum allocation of capacity of a shared resource to a consumer for a scheduling interval. Utilization of an allocated capacity of the shared resource by the consumer during a given scheduling interval is measured, and when the allocated capacity of the shared resource is completely utilized by the consumer during the given scheduling interval, the scheduler increases the allocated capacity of the shared resource to the defined maximum allocation for the consumer for a next scheduling interval. Thus, rather than gradually increasing the allocation of capacity over many intervals, the scheduler immediately increases the allocation to a predefined maximum amount in response to an allocated amount of capacity being completely utilized during a scheduling interval.

摘要翻译： 一种方法包括在调度间隔中定义用于向消费者的共享资源的容量的最大分配的调度器参数。 测量给定调度间隔期间消费者分配的共享资源容量的利用率，并且当在给定调度间隔内消费者完全利用共享资源的分配容量时，调度器增加共享资源的分配容量 资源到消费者下一个调度间隔的定义的最大分配。 因此，不是在多个间隔中逐渐增加容量的分配，所以调度器响应于在调度间隔期间完全利用的分配的容量而立即将分配增加到预定的最大量。

公开(公告)号：US20120117392A1

公开(公告)日：2012-05-10

申请号：US13259118

申请日：2009-03-25

申请人： Thomas E. Turicchi, JR. ,  Clifford A. McCarthy ,  Charles W. Cochran

发明人： Thomas E. Turicchi, JR. ,  Clifford A. McCarthy ,  Charles W. Cochran

IPC分类号： G06F1/26

CPC分类号： G06F1/263 ,  G06F1/28 ,  G06Q10/04 ,  G06Q10/06 ,  G06Q50/06 ,  H05K7/1498

摘要： Apparatus, methods, and other embodiments associated with providing a correlation between a power distribution unit(s) and a device(s) are described. One example method includes storing first time series data that identifies, on a per power distribution unit (PDU) basis, current drawn from a set of PDUs. The example method may include storing second time series data that identifies, on a per device basis, power used by a set of related devices. With the two time series data available, the method may then provide a PDU-device correlation signal that identifies a correlation between current drawn from a PDU and power used by a device.

摘要翻译： 描述与提供配电单元和设备之间的相关性相关联的装置，方法和其它实施例。 一个示例性方法包括存储在每个功率分配单元（PDU）的基础上识别从一组PDU中提取的电流的第一时间序列数据。 示例性方法可以包括存储在每个设备的基础上识别一组相关设备使用的功率的第二时间序列数据。 利用两个时间序列数据可用，该方法然后可以提供PDU设备相关信号，其识别从PDU中抽取的电流和由设备使用的功率之间的相关性。

公开(公告)号：US07941804B1

公开(公告)日：2011-05-10

申请号：US11263140

申请日：2005-10-31

申请人： Daniel E. Herington ,  Clifford A. McCarthy

发明人： Daniel E. Herington ,  Clifford A. McCarthy

IPC分类号： G06F9/46 ,  G06F15/00

CPC分类号： G06F9/5083 ,  G06F9/45533 ,  G06F9/5077 ,  G06F2009/45587 ,  G06F2209/5022

摘要： In one embodiment, a system comprises a plurality of partitions that provide isolated operating environments for execution of software processes, wherein the plurality of partitions are arranged in a tiered manner with different partition types at different tiers, a data structure for representing the plurality of partitions, wherein the data structure comprises multiple nodes corresponding to the plurality of partitions and links between the nodes representing how the plurality of partitions are arranged in the tiers, and an arbiter software module for allocating resources between the plurality of partitions, wherein the arbiter software module receives requests to allocate resources to the plurality of partitions and traverses the data structure to determine which requests to satisfy.

摘要翻译： 在一个实施例中，系统包括提供用于执行软件过程的隔离操作环境的多个分区，其中所述多个分区以不同层级的不同分区类型以分层方式排列，用于表示所述多个分区的数据结构 其中，所述数据结构包括对应于所述多个分区的多个节点和表示所述多个分区如何布置在所述层中的所述节点之间的链接;以及仲裁器软件模块，用于在所述多个分区之间分配资源，其中所述仲裁器软件模块 接收向多个分区分配资源的请求，并遍历数据结构以确定哪个请求满足。

公开(公告)号：US20080022282A1

公开(公告)日：2008-01-24

申请号：US11489968

申请日：2006-07-20

申请人： Ludmila Cherkasova ,  Jerome Rolia ,  Clifford A. McCarthy

发明人： Ludmila Cherkasova ,  Jerome Rolia ,  Clifford A. McCarthy

IPC分类号： G06F9/46

CPC分类号： G06F9/505 ,  G06F9/5044 ,  G06F11/3461 ,  G06F11/3495 ,  G06F2209/5022 ,  G06F2209/508

摘要： A system comprises a workload manager evaluator operable to receive a representative workload that is representative of competing workloads that share access to at least one shared computing resource. The workload manager evaluator is operable to evaluate performance of a scheduler that schedules access of the competing workloads to the shared computing resource according to defined control parameter values, wherein the workload manager evaluator evaluates performance of the scheduler under the representative workload for a plurality of different values of the control parameters. In certain embodiments, the workload manager evaluator determines an optimal value for the control parameters of the scheduler for scheduling access to the at least one shared computing resource for the representative workload to satisfy defined performance desires of the system.

摘要翻译： 系统包括工作负载管理器评估器，其可操作以接收代表共享对至少一个共享计算资源的访问的竞争工作负载的代表性工作负载。 工作负载管理器评估器可用于评估调度器的性能，该调度器根据定义的控制参数值来调度对共享计算资源的访问，其中工作负载管理器评估器评估多个不同的代表性工作负载下的调度器的性能 控制参数的值。 在某些实施例中，工作负载管理器评估器确定调度器的控制参数的最优值，以调度对代表性工作负载的至少一个共享计算资源的访问以满足系统的定义的性能期望。

公开(公告)号：US07228546B1

公开(公告)日：2007-06-05

申请号：US09493753

申请日：2000-01-28

申请人： Clifford A. McCarthy ,  Raja Daoud

发明人： Clifford A. McCarthy ,  Raja Daoud

IPC分类号： G06F9/46

CPC分类号： G06F9/5061

摘要： The inventive work load manager (WLM) dynamically responds to changes in workload characteristics. The WLM bases response on performance goals set by the administrator, and manipulates the resources to achieve these goals. The WLM receives performance information from performance monitors. The WLM examines the information from the performance monitors and compares the information with the goals using a Proportional Integral and Derivative controller. Based on the comparison, the WLM may increase, decrease, or leave unchanged, the resources devoted to an application. If the performance of an application is performance is lagging, e.g., if transactions are taking longer than the goal, then the WLM increases the entitlement. If an application is over-achieving, then the WLM will decrease its entitlement and allocate it to another application. The WLM uses an arbiter which allocates the resources between the different applications, based on the priorities of the applications. Consequently, the WLM creates a feedback loop between application performance and CPU entitlements.

摘要翻译： 本发明的工作负载管理器（WLM）动态响应工作负载特性的变化。 WLM基于管理员设定的绩效目标，并操纵资源实现这些目标。 WLM从性能监视器接收性能信息。 WLM检查性能监视器中的信息，并使用比例积分和微分控制器将信息与目标进行比较。 根据比较，WLM可能会增加，减少或保持不变，用于应用程序的资源。 如果应用程序的性能落后，例如，如果事务占用的时间长于目标，则WLM会增加权限。 如果应用程序过度实现，则WLM将减少其权限并将其分配给另一个应用程序。 WLM使用仲裁器，它根据应用程序的优先级分配不同应用程序之间的资源。 因此，WLM在应用程序性能和CPU授权之间创建一个反馈循环。