公开(公告)号：US07562090B2

公开(公告)日：2009-07-14

申请号：US10324362

申请日：2002-12-19

申请人： Guy Maring Lohman ,  Nimrod Megiddo ,  Jun Rao ,  Chun Zhang

发明人： Guy Maring Lohman ,  Nimrod Megiddo ,  Jun Rao ,  Chun Zhang

IPC分类号： G06F7/00 ,  G06F12/00

CPC分类号： G06F17/30584 ,  G06F17/30445 ,  Y10S707/99943 ,  Y10S707/99953

摘要： A system for automating data partitioning in a parallel database includes plural nodes connected in parallel. Each node includes a database server and two databases connected thereto. Each database server includes a query optimizer. Moreover, a partitioning advisor communicates with the database server and the query optimizer. The query optimizer and the partitioning advisor include a program for recommending and evaluating data table partitions that are useful for processing a workload of query statements. The data table partitions are recommended and evaluated without requiring the data tables to be physically repartitioned.

摘要翻译： 用于在并行数据库中自动化数据划分的系统包括并行连接的多个节点。 每个节点包括一个数据库服务器和两个连接到其上的数据库。 每个数据库服务器都包含一个查询优化器。 此外，分区顾问与数据库服务器和查询优化器进行通信。 查询优化器和分区顾问程序包括一个用于推荐和评估对处理查询语句的工作负载有用的数据表分区的程序。 建议和评估数据表分区，而不需要物理重新分区数据表。

公开(公告)号：US20090005902A1

公开(公告)日：2009-01-01

申请号：US12055801

申请日：2008-03-26

申请人： Nimrod Megiddo

发明人： Nimrod Megiddo

IPC分类号： G06F17/00

CPC分类号： G06Q10/08

摘要： Disclosed is a method for operating and managing a re-fueling business. The method includes contracting with a customer to fuel a fuel consuming vehicle for the customer, providing a locking device with a fuel tank of the vehicle, the locking device being unlockable by the re-fueling business and the customer, and transporting a fuel containing structure from the re-fueling business to various locations of the customer's vehicle whenever a fueling of the vehicle is desirable to the business and the customer, the locations being remote of the business, unlocking the locking device of the fuel tank upon arrival at the locations, filling the fuel tank with an amount of fuel from the fuel containing structure upon the transporting of the fuel container to the locations, and re-locking the locking device after filling the fuel tank with the amount of fuel.

摘要翻译： 公开了一种操作和管理再加油业务的方法。 该方法包括与客户订约以为顾客燃料消耗车辆，提供具有车辆燃料箱的锁定装置，锁定装置可通过再加油业务和客户解锁，以及运送含燃料的结构 从客户车辆的各个地点到车辆的燃料供应对于企业和客户都是希望的，远离业务的位置，在到达位置时解锁燃料箱的锁定装置， 在将燃料容器运送到位置时，从燃料容纳结构向燃料箱填充燃料箱，并且在用燃料量填充燃料箱之后重新锁定锁定装置。

公开(公告)号：US20080306891A1

公开(公告)日：2008-12-11

申请号：US12134073

申请日：2008-06-05

申请人： Elad Eliezer Hazan ,  Nimrod Megiddo

发明人： Elad Eliezer Hazan ,  Nimrod Megiddo

IPC分类号： G06F15/18

CPC分类号： G06N99/005 ,  G06N5/02 ,  G06Q30/0244

摘要： Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.

公开(公告)号：US20080222093A1

公开(公告)日：2008-09-11

申请号：US12125221

申请日：2008-05-22

申请人： Wei Fan ,  Guy Maring Lohman ,  Volker Gerhard Markl ,  Nimrod Megiddo ,  Jun Rao ,  David Everett Simmen ,  Julia Stoyanovich

发明人： Wei Fan ,  Guy Maring Lohman ,  Volker Gerhard Markl ,  Nimrod Megiddo ,  Jun Rao ,  David Everett Simmen ,  Julia Stoyanovich

IPC分类号： G06F17/30

CPC分类号： G06F17/30469

摘要： A method and system for automatically and adaptively determining query execution plans for parametric queries. A first classifier trained by an initial set of training points is generated. A query workload and/or database statistics are dynamically updated. A new set of training points is collected off-line. Using the new set of training points, the first classifier is modified into a second classifier. A database query is received at a runtime subsequent to the off-line phase. The query includes predicates having parameter markers bound to actual values. The predicates are associated with selectivities. A mapping of the selectivities into a plan determines the query execution plan. The determined query execution plan is included in an augmented set of training points, where the augmented set includes the initial set and the new set.

摘要翻译： 一种用于自动和自适应地确定参数查询的查询执行计划的方法和系统。 产生由初始训练点训练的第一分类器。 动态更新查询工作负载和/或数据库统计信息。 离线收集了一套新的培训点。 使用新的一组训练点，第一个分类器被修改为第二个分类器。 在离线阶段之后的运行时间接收数据库查询。 该查询包括具有绑定到实际值的参数标记的谓词。 谓词与选择性相关联。 将选择性映射到计划中确定查询执行计划。 确定的查询执行计划被包括在增强的训练点集合中，其中增强集合包括初始集合和新集合。

公开(公告)号：US07117130B1

公开(公告)日：2006-10-03

申请号：US09607513

申请日：2000-06-28

申请人： Nimrod Megiddo

发明人： Nimrod Megiddo

IPC分类号： G06F17/10

CPC分类号： G05B13/024

摘要： Stochastic control problems of linear systems in high dimensions are solved by modeling a structured Markov Decision Process (MDP). A state space for the MDP is a polyhedron in a Euclidean space and one or more actions that are feasible in a state of the state space are linearly constrained with respect to the state. One or more approximations are built from above and from below to a value function for the state using representations that facilitate the computation of approximately optimal actions at any given state by linear programming.

摘要翻译： 通过对结构化马尔可夫决策过程（MDP）进行建模来解决高维度线性系统的随机控制问题。 用于MDP的状态空间是欧几里得空间中的多面体，并且在状态空间的状态中可行的一个或多个动作相对于状态是线性约束的。 一个或多个近似是从上面和从下面建立到使用表示的状态的值函数，这些表示便于通过线性规划在任何给定状态下计算近似最佳动作。

公开(公告)号：US20050235114A1

公开(公告)日：2005-10-20

申请号：US11151363

申请日：2005-06-13

申请人： Nimrod Megiddo ,  Dharmendra Modha

发明人： Nimrod Megiddo ,  Dharmendra Modha

IPC分类号： G06F12/00 ,  G06F12/12

CPC分类号： G06F12/123 ,  G06F12/122 ,  G06F12/127 ,  G06F2212/502

摘要： An adaptive replacement cache policy dynamically maintains two lists of pages, a recency list and a frequency list, in addition to a cache directory. The policy keeps these two lists to roughly the same size, the cache size c. Together, the two lists remember twice the number of pages that would fit in the cache. At any time, the policy selects a variable number of the most recent pages to exclude from the two lists. The policy adaptively decides in response to an evolving workload how many top pages from each list to maintain in the cache at any given time. It achieves such online, on-the-fly adaptation by using a learning rule that allows the policy to track a workload quickly and effectively.

摘要翻译： 自适应替换高速缓存策略除了缓存目录之外，还动态维护两个页面列表，新近度列表和频率列表。 该策略将这两个列表大致相同的大小，缓存大小c。 一起，两个列表记住了适合缓存的页面数量的两倍。 在任何时候，策略都会选择要从两个列表中排除的最新页面的可变数量。 该策略自适应地决定响应于不断变化的工作负载，在任何给定时间，来自每个列表的顶页数量将保持在高速缓存中。 它通过使用允许策略来快速有效地跟踪工作负载的学习规则来实现在线，即时的适应。

公开(公告)号：US20050114849A1

公开(公告)日：2005-05-26

申请号：US10723850

申请日：2003-11-25

申请人： Nimrod Megiddo

发明人： Nimrod Megiddo

IPC分类号： G06F9/45

CPC分类号： G06F8/443

摘要： Disclosed is a system and method for instructing a computer program to self-optimize comprising inputting commands into the computer program and allowing a learning protocol in the computer program to determine an approximately optimal policy of operation of the computer program based on the commands. The commands comprise operational choices for the computer program to select from including an approximately optimal choice for optimizing the operation of the program. The commands comprise a selection command for selecting any function in a list of instructions inputted into the program, wherein the function provides a basis of making an approximately optimal choice. Additionally, the commands comprise a rule command for instructing the computer program of how to make an approximately optimal choice. Moreover, the commands comprise a reward command for instructing the program which of the operational choices results in an approximately optimal choice for optimizing the operation of the computer program.

摘要翻译： 公开了一种用于指示计算机程序自我优化的系统和方法，包括将命令输入到计算机程序中，并允许计算机程序中的学习协议基于命令确定计算机程序的近似最佳操作策略。 这些命令包括用于选择的计算机程序的操作选择，以包括用于优化程序的操作的近似最佳选择。 这些命令包括用于选择输入到程序中的指令列表中的任何功能的选择命令，其中该功能提供了进行近似最佳选择的基础。 此外，命令包括用于指示计算机程序如何进行大致最佳选择的规则命令。 此外，命令包括用于指示程序的奖励命令，哪个操作选择导致用于优化计算机程序的操作的近似最佳选择。

公开(公告)号：US06898665B1

公开(公告)日：2005-05-24

申请号：US09481233

申请日：2000-01-11

申请人： Nimrod Megiddo

发明人： Nimrod Megiddo

IPC分类号： G06F12/00 ,  G11B5/55

CPC分类号： G11B5/5547

摘要： A system and method that utilize a scheduling algorithm to profile the seek operation of an I/O command based on its available rotational time, for improving the average command access time and hence the disk drive throughput. The seek profile of a command is adjusted so that the seek operation is performed in a servo dual mode: a conservative mode at a slower seek velocity, and an aggressive mode at a higher seek velocity. The scheduling algorithm reorders the I/O commands based on the seek length from the current command to the next command, and on the rotational distance that reflects the rotational time. If the rotational time for a particular command is long relative to its seek length, the seek operation is performed in the conservative mode with a very high probability of success within the available time. If, however, the rotational time for a particular command is short relative to its seek length, whereby the conservative mode is expected to require an additional revolution, the seek operation is performed in the aggressive mode with a significant probability of success within the available rotational time. In the event of a resulting miss, a revolution will be added to the rotational time. This dual servo increases the overall probability of success of the seek operation, resulting in a smaller average access time and thereby a higher throughput of the disk drive.

摘要翻译： 一种利用调度算法基于其可用旋转时间对I / O命令的查找操作进行分析的系统和方法，用于改善平均命令访问时间，从而提高磁盘驱动器吞吐量。 调整命令的寻道曲线，使得以伺服双模式执行寻道操作：以较慢的寻道速度进行保守模式，并以更高的寻道速度进行攻击模式。 调度算法根据从当前命令到下一个命令的查找长度以及反映旋转时间的旋转距离对I / O命令进行重新排序。 如果特定命令的旋转时间相对于其寻道长度较长，则在可用时间内成功的概率非常高的保守模式下执行寻道操作。 然而，如果特定命令的旋转时间相对于其寻道长度较短，由此预期保守模式需要额外的旋转，则在攻击模式下执行寻道操作，在可用旋转中具有显着的成功概率 时间。 在遇到错误的情况下，旋转时间将会增加一场革命。 这种双伺服机构增加了寻道操作成功的总体概率，导致较小的平均访问时间，从而提高磁盘驱动器的吞吐量。

公开(公告)号：US06742179B2

公开(公告)日：2004-05-25

申请号：US09902595

申请日：2001-07-12

申请人： Nimrod Megiddo ,  Bilha Mendelson

发明人： Nimrod Megiddo ,  Bilha Mendelson

IPC分类号： G06F944

CPC分类号： G06F8/4441

摘要： A program product and method of compiling a computer program to optimize performance of a computer program. First, after initialization, a profiling run is done on computer code which may include program code blocks and program data. Execution of each computer program step is monitored and each occurrence of each individual code unit is logged, e.g. each instruction block or block of data. Frequently occurring code units are identified periodically as hot blocks. An initial snapshot of hot blocks is logged, e.g., when identified hot blocks exceed an initial block number. Profiling continues until the profiling run is complete, updating identified hot blocks and logging a new current snapshot whenever a current set of identified hot blocks contains a selected percentage of different hot blocks. Snapshots are selected as representative to different program modes. The program is optimized according to program modes.

摘要翻译： 一种编译计算机程序以优化计算机程序性能的程序产品和方法。 首先，在初始化之后，对可能包括程序代码块和程序数据的计算机代码进行分析运行。 监视每个计算机程序步骤的执行，并记录每个单独代码单元的每个出现。 每个指令块或数据块。 经常出现的代码单元被定期识别为热块。 记录热块的初始快照，例如，当识别的热块超过初始块号时。 性能持续到分析运行完成，更新已识别的热块，并记录新的当前快照，每当一组已识别的热块包含所选百分比的不同热块。 选择快照作为不同程序模式的代表。 该程序根据程序模式进行优化。

公开(公告)号：US06665649B1

公开(公告)日：2003-12-16

申请号：US09523100

申请日：2000-03-10

申请人： Nimrod Megiddo

发明人： Nimrod Megiddo

IPC分类号： G06F1760

CPC分类号： G06Q30/08 ,  G06Q40/04

摘要： The present invention provides a smooth ending system for an auction on the Internet. The system randomly selects an end time for the auction by picking an auction duration from a probability distribution, so that bidders have no preferences over the time of bidding. At any given time, assuming that the auction is still in process, there is a fixed probability, independent of the time that has elapsed since the beginning of the auction, that the auction will end within the next time unit.

摘要翻译： 本发明提供了一种用于互联网拍卖的平滑结束系统。 系统通过从概率分布中挑选拍卖持续时间来随机选择拍卖的结束时间，以便投标人在投标时没有偏好。 在任何给定的时间，假设拍卖仍在进行中，独立于拍卖开始以来已经过去的时间是固定的概率，即拍卖将在下一个时间单位内结束。