公开(公告)号：US08914795B2

公开(公告)日：2014-12-16

申请号：US13359825

申请日：2012-01-27

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06F9/455 ,  G06F9/46 ,  G06F15/173 ,  G06F9/50

CPC分类号： G06F9/45558 ,  G06F9/5066 ,  G06F9/5077 ,  G06F2009/4557 ,  G06F2209/5019

摘要： An information processing apparatus having a prediction section that determines the predicted peak usage amount of physical resources for each time interval for individual clusters each including a plurality of virtual machines having the same function; a setting section sets a constraint that ensures that, for individual combinations of a first physical machine, a second physical machine, and a time interval, the total predicted peak usage amount of a physical resource predicted for the first physical machine if the second physical machine stops during the time interval does not exceed a physical resource amount prepared for the first physical machine; and an allocation-configuration deriving section derives an allocation configuration by calculating, in accordance with the constraint, a solution to an optimization problem for minimizing, as an objective function, the sum total of the physical resource amounts of all of physical machines to which the virtual machines are allocated.

摘要翻译： 一种信息处理装置，具有：预测部，其对于每个包含具有相同功能的多个虚拟机的各个簇确定每个时间间隔的物理资源的预测峰值使用量; 设置部分设置一个约束，其确保对于第一物理机器，第二物理机器和时间间隔的单独组合，如果第二物理机器预测的第一物理机器预测的物理资源的总预测峰值使用量 在时间间隔期间停止不超过为第一物理机准备的物理资源量; 并且分配配置导出部分通过根据约束计算优化问题的解决方案来导出分配配置，以使作为目标函数的所有物理机器的物理资源量的总和最小化为 分配虚拟机。

公开(公告)号：US20120317573A1

公开(公告)日：2012-12-13

申请号：US13590651

申请日：2012-08-21

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06F9/455

CPC分类号： G06F9/45558 ,  G06F9/5066 ,  G06F9/5077 ,  G06F2009/4557 ,  G06F2209/5019

摘要： A computer-executable method for determining an allocation configuration for allocating virtual machines to physical machines. The method includes the steps of determining a predicted peak usage amount of physical resources for each time interval for individual clusters wherein the clusters include a plurality of virtual machines; setting a constraint that ensures the total predicted peak usage amount of the physical resource predicted for the first physical machine does not exceed the physical resource amount for the first physical machine; and deriving an allocation configuration by calculating, in accordance with the constraint, a solution to an optimization problem for minimizing, as an objective function, the sum total of the physical resource amounts of the plurality of physical machines to which the virtual machines are allocated.

摘要翻译： 一种用于确定用于将虚拟机分配给物理机器的分配配置的计算机可执行方法。 该方法包括以下步骤：确定每个群集的每个时间间隔的物理资源的预计峰值使用量，其中，群集包括多个虚拟机; 设置一个约束，确保为第一物理机预测的物理资源的总预测峰值使用量不超过第一物理机的物理资源量; 以及通过根据所述约束计算解决最优化问题来导出分配配置，以使作为目标函数的虚拟机被分配到的多个物理机的物理资源量的总和最小化。

公开(公告)号：US09135564B2

公开(公告)日：2015-09-15

申请号：US13589407

申请日：2012-08-20

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06N7/00 ,  G06N99/00

CPC分类号： G06N7/005 ,  G06N99/005

摘要： A method for determining an optimum policy by using a Markov decision process in which T subspaces each have at least one state having a cyclic structure includes identifying, with a processor, subspaces that are part of a state space; selecting a t-th (t is a natural number, t≦T) subspace among the identified subspaces; computing a probability of, and an expected value of a cost of, reaching from one or more states in the selected t-th subspace to one or more states in the t-th subspace in a following cycle; and recursively computing a value and an expected value of a cost based on the computed probability and expected value of the cost, in a sequential manner starting from a (t−1)-th subspace.

公开(公告)号：US09135563B2

公开(公告)日：2015-09-15

申请号：US13586385

申请日：2012-08-15

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06N99/00 ,  G06N7/00

CPC分类号： G06N7/005 ,  G06N99/005

摘要： A method for determining an optimum policy by using a Markov decision process in which T subspaces each have at least one state having a cyclic structure includes identifying, with a processor, subspaces that are part of a state space; selecting a t-th (t is a natural number, t≦T) subspace among the identified subspaces; computing a probability of, and an expected value of a cost of, reaching from one or more states in the selected t-th subspace to one or more states in the t-th subspace in a following cycle; and recursively computing a value and an expected value of a cost based on the computed probability and expected value of the cost, in a sequential manner starting from a (t−1)-th subspace.

摘要翻译： 一种用于通过使用马尔科夫决定过程来确定最佳策略的方法，其中T个子空间各自具有至少一个具有循环结构的状态，包括用处理器识别属于状态空间的子空间的子空间; 在所识别的子空间中选择第t（t是自然数，t＆nlE; T）子空间; 计算在所选择的第t个子空间中的一个或多个状态到下一周期中的第t个子空间中的一个或多个状态的成本的概率和期望值; 并且以从第（t-1）个子空间开始的顺序方式，基于计算出的成本的概率和期望值，递归计算成本的值和期望值。

公开(公告)号：US20130085983A1

公开(公告)日：2013-04-04

申请号：US13586385

申请日：2012-08-15

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06N5/02

CPC分类号： G06N7/005 ,  G06N99/005

摘要： A method for determining an optimum policy by using a Markov decision process in which T subspaces each have at least one state having a cyclic structure includes identifying, with a processor, subspaces that are part of a state space; selecting a t-th (t is a natural number, t≦T) subspace among the identified subspaces; computing a probability of, and an expected value of a cost of, reaching from one or more states in the selected t-th subspace to one or more states in the t-th subspace in a following cycle; and recursively computing a value and an expected value of a cost based on the computed probability and expected value of the cost, in a sequential manner starting from a (t−1)-th subspace.

摘要翻译： 一种用于通过使用马尔科夫决定过程来确定最佳策略的方法，其中T个子空间各自具有至少一个具有循环结构的状态，包括用处理器识别属于状态空间的子空间的子空间; 在所识别的子空间中选择第t（t是自然数，t＆nlE; T）子空间; 计算在所选择的第t个子空间中的一个或多个状态到下一周期中的第t个子空间中的一个或多个状态的成本的概率和期望值; 并且以从第（t-1）个子空间开始的顺序方式，基于计算出的成本的概率和期望值，递归计算成本的值和期望值。

公开(公告)号：US08914797B2

公开(公告)日：2014-12-16

申请号：US13590651

申请日：2012-08-21

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06F9/455 ,  G06F9/46 ,  G06F15/173 ,  G06F9/50

CPC分类号： G06F9/45558 ,  G06F9/5066 ,  G06F9/5077 ,  G06F2009/4557 ,  G06F2209/5019

摘要： A computer-executable method for determining an allocation configuration for allocating virtual machines to physical machines. The method includes the steps of determining a predicted peak usage amount of physical resources for each time interval for individual clusters wherein the clusters include a plurality of virtual machines; setting a constraint that ensures the total predicted peak usage amount of the physical resource predicted for the first physical machine does not exceed the physical resource amount for the first physical machine; and deriving an allocation configuration by calculating, in accordance with the constraint, a solution to an optimization problem for minimizing, as an objective function, the sum total of the physical resource amounts of the plurality of physical machines to which the virtual machines are allocated.

摘要翻译： 一种用于确定用于将虚拟机分配给物理机器的分配配置的计算机可执行方法。 该方法包括以下步骤：确定每个群集的每个时间间隔的物理资源的预计峰值使用量，其中，群集包括多个虚拟机; 设置一个约束，确保为第一物理机预测的物理资源的总预测峰值使用量不超过第一物理机的物理资源量; 以及通过根据所述约束计算解决最优化问题来导出分配配置，以使作为目标函数的虚拟机被分配到的多个物理机的物理资源量的总和最小化。

公开(公告)号：US20130085974A1

公开(公告)日：2013-04-04

申请号：US13589407

申请日：2012-08-20

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06N5/02

CPC分类号： G06N7/005 ,  G06N99/005

摘要： A method for determining an optimum policy by using a Markov decision process in which T subspaces each have at least one state having a cyclic structure includes identifying, with a processor, subspaces that are part of a state space; selecting a t-th (t is a natural number, t≦T) subspace among the identified subspaces; computing a probability of, and an expected value of a cost of, reaching from one or more states in the selected t-th subspace to one or more states in the t-th subspace in a following cycle; and recursively computing a value and an expected value of a cost based on the computed probability and expected value of the cost, in a sequential manner starting from a (t −1)-th subspace.

摘要翻译： 一种用于通过使用马尔可夫决策过程来确定最佳策略的方法，其中T个子空间各自具有至少一个具有循环结构的状态，包括用处理器识别属于状态空间的子空间的子空间; 在所识别的子空间中选择第t（t是自然数，t＆nlE; T）子空间; 计算在所选择的第t个子空间中的一个或多个状态到下一周期中的第t个子空间中的一个或多个状态的成本的概率和期望值; 并且以从第（t-1）个子空间开始的顺序方式，以计算出的成本的概率和预期值递归地计算成本的值和期望值。

公开(公告)号：US20120198447A1

公开(公告)日：2012-08-02

申请号：US13359825

申请日：2012-01-27

申请人： Takayuki Osogami ,  Raymond H. Rudy

发明人： Takayuki Osogami ,  Raymond H. Rudy

IPC分类号： G06F9/455

CPC分类号： G06F9/45558 ,  G06F9/5066 ,  G06F9/5077 ,  G06F2009/4557 ,  G06F2209/5019

摘要： An information processing apparatus having a prediction section that determines the predicted peak usage amount of physical resources for each time interval for individual clusters each including a plurality of virtual machines having the same function; a setting section sets a constraint that ensures that, for individual combinations of a first physical machine, a second physical machine, and a time interval, the total predicted peak usage amount of a physical resource predicted for the first physical machine if the second physical machine stops during the time interval does not exceed a physical resource amount prepared for the first physical machine; and an allocation-configuration deriving section derives an allocation configuration by calculating, in accordance with the constraint, a solution to an optimization problem for minimizing, as an objective function, the sum total of the physical resource amounts of all of physical machines to which the virtual machines are allocated.

摘要翻译： 一种信息处理装置，具有：预测部，其对于每个包含具有相同功能的多个虚拟机的各个簇确定每个时间间隔的物理资源的预测峰值使用量; 设置部分设置一个约束，其确保对于第一物理机器，第二物理机器和时间间隔的单独组合，如果第二物理机器预测的第一物理机器预测的物理资源的总预测峰值使用量 在时间间隔期间停止不超过为第一物理机准备的物理资源量; 并且分配配置导出部分通过根据约束计算优化问题的解决方案来导出分配配置，以使作为目标函数的所有物理机器的物理资源量的总和最小化为 分配虚拟机。