公开(公告)号：JP2011081800A

公开(公告)日：2011-04-21

申请号：JP2010225316

申请日：2010-10-05

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： RUBIO JUAN CARLOS ,  RAJAMANI KARTHICK ,  WARE MALCOLM SCOTT ,  FLOYD MICHAEL STEPHEN

IPC: G06F1/26

CPC classification number: G06F1/3296 ,  G06F1/3203 ,  G06F1/324 ,  Y02D10/126 ,  Y02D10/172

Abstract: PROBLEM TO BE SOLVED: To provide a managing operation associated with one or more voltage changes and one or more frequency changes. SOLUTION: A voltage change request and a frequency change request are associated with dynamic voltage and frequency scaling (DVFS) operations. The DVFS operation is transmitted by a processor to be executed by one or more direct current assemblies. A sequence associated with the one or more voltage changes and a sequence associated with the one or more frequency changes are detected by a system. The sequences are dynamically modified to enable insertion of an additional voltage change, whereby the additional voltage change indicates completion of one or more previous voltage change requests. Completion of the voltage change request enables one or more subsequent voltage change requests to be processed. When a voltage change request is not successfully completed, one or more future voltage changes are suspended. COPYRIGHT: (C)2011,JPO&INPIT

Abstract translation: 要解决的问题：提供与一个或多个电压变化和一个或多个频率变化相关联的管理操作。 解决方案：电压变化请求和频率变化请求与动态电压和频率缩放（DVFS）操作相关联。 DVFS操作由处理器发送以由一个或多个直流组件执行。 与一个或多个电压变化相关联的序列和与一个或多个频率变化相关联的序列由系统检测。 这些序列被动态地修改以允许插入额外的电压变化，由此额外的电压变化指示完成一个或多个先前的电压变化请求。 完成电压变化请求使得能够处理一个或多个后续电压变化请求。 当电压变化请求未成功完成时，将暂停一个或多个未来的电压变化。 版权所有（C）2011，JPO＆INPIT

公开(公告)号：JP2004326808A

公开(公告)日：2004-11-18

申请号：JP2004131849

申请日：2004-04-27

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ARIMILLI RAVI KUMAR ,  FLOYD MICHAEL STEPHEN ,  REICK KEVIN FRANKLIN

IPC: G06F13/14 ,  G06F9/445 ,  G06F13/00 ,  G06F13/12 ,  G06F13/40 ,  G06F15/16 ,  G06F15/177

CPC classification number: G06F13/4081

Abstract: PROBLEM TO BE SOLVED: To provide a data processing system having a function of adding and removing a hot plug of individual hot plug possible constitution elements without intervening it to a present operation of a whole processing system. SOLUTION: The processing system includes a mutual connection structure including a hot plug connector which can connect the outer hot plug possible constitution element to the data processing system. A logical constitution element includes constitution logic, routing and operation logic. When the hot plug possible constitution element is connected to the hot plug connector, a service element automatically detects connection and selects a correct constitution file for an extension system. When the constitution file is once loaded and that the operation of the new element is prepared is shown by system check of the new element, the new element is integrated in the existed system. OS allocates a work load to the new element. The whole processes are performed without turning off power or without intervening it to the operation of the existed element from a viewpoint of a customer. COPYRIGHT: (C)2005,JPO&NCIPI

公开(公告)号：JP2005011319A

公开(公告)日：2005-01-13

申请号：JP2004131814

申请日：2004-04-27

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ARIMILLI RAVI KUMAR ,  FLOYD MICHAEL STEPHEN ,  REICK KEVIN FRANKLIN

IPC: G06F13/14 ,  G06F1/18 ,  G06F3/00 ,  G06F9/445 ,  G06F9/46 ,  G06F9/54 ,  G06F13/00 ,  G06F13/40 ,  G06F15/177

CPC classification number: G06F13/4081

Abstract: PROBLEM TO BE SOLVED: To provide a data processing system equipped with a hot plug function without intervention for several main hardware components, such as a processor, a memory, and an input/output (I/O) channel. SOLUTION: The data processing system comprises former processors connected to each other via an attachment feature, a former memory, and a former input and the output (I/O) channel. Furthermore, the data processing system also comprises a service element and an operating system (OS). The attachment feature comprises an interconnect line, a hardware component, and a software logic component, which enables the processing system to embody functions of hot plug addtion (or removal) of reconstruction functions, which are the addition and the removal of the processor, the memory, and the I/O channel. Various components are added to the system without interfering the existing component processing and used in the expanded system they can be used immediately. COPYRIGHT: (C)2005,JPO&NCIPI

Abstract translation: 要解决的问题：提供一种配备有热插拔功能的数据处理系统，无需干预几个主要硬件组件，如处理器，存储器和输入/输出（I / O）通道。 解决方案：数据处理系统包括通过附件功能，前一个存储器和前一个输入和输出（I / O）通道彼此连接的前处理器。 此外，数据处理系统还包括服务元件和操作系统（OS）。 附件特征包括互连线，硬件组件和软件逻辑组件，这使得处理系统能够实现作为处理器的添加和移除的重新配置功能的热插拔（或移除）功能， 内存和I / O通道。 将各种组件添加到系统中，而不会干扰现有的组件处理，并在扩展系统中使用它们可以立即使用。 版权所有（C）2005，JPO＆NCIPI

公开(公告)号：JP2004326809A

公开(公告)日：2004-11-18

申请号：JP2004131894

申请日：2004-04-27

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ARIMILLI RAVI KUMAR ,  FLOYD MICHAEL STEPHEN ,  REICK KEVIN FRANKLIN

IPC: G06F11/30 ,  G06F9/445 ,  G06F9/46 ,  G06F9/48 ,  G06F9/54 ,  G06F13/00

CPC classification number: G06F11/2043 ,  G06F11/2028 ,  G06F11/2289

Abstract: PROBLEM TO BE SOLVED: To provide a method and a system for dynamically detecting a component having a problem in a hot plug processing system without intervening it to a whole processing of the system and for automatically removing the component having the problem by a hot removing method and to provide a data processing system. SOLUTION: The data processing system providing a non-intervened hot plug function is designed with additional logic to make a hot plug possible component start and complete a test sequence of a factory level and to judge whether the component appropriately functions or not. When the component does not appropriately functions, OS re-allocates a work load of the component to the other component of the system. When OS completes re-allocation, a service element starts hot removal of the component. Thus, the component is logically and electrically separated from the system. COPYRIGHT: (C)2005,JPO&NCIPI

Abstract translation: 要解决的问题：提供一种用于在热插拔处理系统中动态地检测具有问题的部件的方法和系统，而不将其干涉到系统的整个处理，并且通过一个 热去除方法并提供数据处理系统。

解决方案：提供非干预热插拔功能的数据处理系统设计有额外的逻辑，以使热插拔可能的组件启动并完成工厂级的测试序列，并判断组件是否适当地起作用。 当组件没有适当的功能时，OS会将组件的工作负载重新分配给系统的其他组件。 当操作系统完成重新分配时，服务元素将启动组件的热删除。 因此，组件与系统逻辑和电气分离。 版权所有（C）2005，JPO＆NCIPI

公开(公告)号：DE112013002011B4

公开(公告)日：2018-02-15

申请号：DE112013002011

申请日：2013-03-27

Applicant: IBM

Inventor： ALLEN-WARE MALCOLM S ,  DRAKE ALAN JAMES ,  HUANG WEI ,  FLOYD MICHAEL STEPHEN ,  WEN HUAJUN

IPC: G06F1/32

Abstract: Verfahren in einem Datenverarbeitungssystem zum Minimieren der Leistungsaufnahme für den Betrieb einer Verarbeitungseinheit mit fester Frequenz, wobei das Verfahren aufweist: Zählen einer Anzahl von Zeitschlitzen in einem Zeitfenster, in denen ein Drosseln der Verarbeitungseinheit mit fester Frequenz ausgeführt wird; Teilen der Anzahl von Zeitschlitzen, in denen ein Drosseln ausgeführt wird, durch eine Gesamtanzahl von Zeitschlitzen im Zeitfenster, wobei ein Arbeitsleistungsverlust-(PLOSS-)Wert erzeugt wird; Ermitteln, ob der der Verarbeitungseinheit mit fester Frequenz zugehörige (PLOSS-)Wert größer als ein zulässiger Arbeitsleistungsverlust-(APLOSS-)Wert ist; und in Reaktion darauf, dass der PLOSS-Wert kleiner als der oder gleich dem APLOSS-Wert ist, Einleiten eines Absenkens der an die Verarbeitungseinheit mit fester Frequenz gelieferten Spannung; Ermitteln, ob bei der Verarbeitungseinheit mit fester Frequenz ein Zeitsteuerungsproblem besteht; und in Reaktion auf ein bestehendes Zeitsteuerungsproblem Drosseln des Betriebs der Verarbeitungseinheit mit fester Frequenz, wobei die Reaktion zum Lösen des Zeitsteuerungsproblems schneller ist als die Reaktion zum Anpassen der an die Verarbeitungseinheit mit fester Frequenz gelieferten Spannung.

公开(公告)号：ES2286630T3

公开(公告)日：2007-12-01

申请号：ES04727310

申请日：2004-04-14

Applicant: IBM

Inventor： ARMSTRONG WILLIAM JOSEPH ,  FLOYD MICHAEL STEPHEN ,  LEITNER LARRY SCOTT ,  KALLA RONALD NICK ,  SINHAROY BALARAM

IPC: G06F11/34 ,  G06F9/38

Abstract: Un método para contabilizar el tiempo de uso de un procesador que da soporte a la ejecución concurrente de una pluralidad de hilos de hardware y que tiene una pluralidad de recursos usados por dicha pluralidad de hilos de hardware, comprendiendo el método: en cada ciclo de reloj de procesador, determinar el uso relativo de recursos en base a si cada hilo de hardware está o no en un estado particular de ciclo de procesador; y en respuesta a dicha determinación, actualizar una pluralidad de cuentas de uso de recursos de procesador, cada una de ellas asociada a un hilo particular de los hilos de hardware en conformidad con dicho uso relativo de recursos, caracterizado porque dicha actualización se realiza por igual en cada una de dicha pluralidad de dichas cuentas de uso de recursos de procesador en respuesta a la determinación de que ninguno de dicha pluralidad de hilos de hardware se encuentra en dicho estado particular de ciclo.

公开(公告)号：DE602004006858D1

公开(公告)日：2007-07-19

申请号：DE602004006858

申请日：2004-04-14

Applicant: IBM

Inventor： ARMSTRONG WILLIAM JOSEPH ,  FLOYD MICHAEL STEPHEN ,  LEITNER LARRY SCOTT ,  KALLA RONALD NICK ,  SINHAROY BALARAM

IPC: G06F11/34 ,  G06F9/38

Abstract: Processor time accounting is enhanced by per-thread internal resource usage counter circuits that account for usage of processor core resources to the threads that use them. Relative resource use can be determined by detecting events such as instruction dispatches for multiple threads active within the processor, which may include idle threads that are still occupying processor resources. The values of the resource usage counters are used periodically to determine relative usage of the processor core by the multiple threads. If all of the events are for a single thread during a given period, the processor time is allocated to the single thread. If no events occur in the given period, then the processor time can be equally allocated among threads. If multiple threads are generating events, a fractional resource usage can be determined for each thread and the counters may be updated in accordance with their fractional usage.

公开(公告)号：BRPI0409710A

公开(公告)日：2006-05-02

申请号：BRPI0409710

申请日：2004-04-14

Applicant: IBM

Inventor： ARMSTRONG WILLIAM JOSEPH ,  FLOYD MICHAEL STEPHEN ,  LEITNER LARRY SCOTT ,  KALLA RONALD NICK ,  SINHAROY BALARAM

IPC: G06F9/38 ,  G06F11/34

Abstract: Processor time accounting is enhanced by per-thread internal resource usage counter circuits that account for usage of processor core resources to the threads that use them. Relative resource use can be determined by detecting events such as instruction dispatches for multiple threads active within the processor, which may include idle threads that are still occupying processor resources. The values of the resource usage counters are used periodically to determine relative usage of the processor core by the multiple threads. If all of the events are for a single thread during a given period, the processor time is allocated to the single thread. If no events occur in the given period, then the processor time can be equally allocated among threads. If multiple threads are generating events, a fractional resource usage can be determined for each thread and the counters may be updated in accordance with their fractional usage.

公开(公告)号：DE112013002011T5

公开(公告)日：2015-01-08

申请号：DE112013002011

申请日：2013-03-27

Applicant: IBM

Inventor： ALLEN-WARE MALCOLM S ,  DRAKE ALAN JAMES ,  HUANG WEI ,  FLOYD MICHAEL STEPHEN ,  WEN HUAJUN

IPC: G06F1/32

Abstract: Es wird ein Mechanismus zum Minimieren der Leistungsaufnahme für den Betrieb einer Verarbeitungseinheit mit fester Frequenz bereitgestellt. Es wird eine Anzahl von Zeitschlitzen in einem Zeitfenster gezählt, in denen eine Drosselung der Verarbeitungseinheit mit fester Frequenz ausgeführt wird. Die Anzahl von Zeitschlitzen, in denen eine Drosselung ausgeführt wird, wird durch eine Gesamtanzahl von Zeitschlitzen im Zeitfenster geteilt, wobei ein Arbeitsleistungsverlust-(PLOSS-)Wert erzeugt wird. Es wird ermittelt, ob der der Verarbeitungseinheit mit fester Frequenz zugehörige (PLOSS-)Wert größer als ein zulässiger Arbeitsleistungsverlust-(APLOSS-)Wert ist. In Reaktion darauf, dass der PLOSS-Wert kleiner als der oder gleich dem APLOSS-Wert ist, wird ein Absenken der an die Verarbeitungseinheit mit fester Frequenz gelieferten Spannung eingeleitet.

公开(公告)号：CA2518468C

公开(公告)日：2008-09-30

申请号：CA2518468

申请日：2004-04-14

Applicant: IBM

Inventor： KALLA RONALD NICK ,  LEITNER LARRY SCOTT ,  FLOYD MICHAEL STEPHEN ,  ARMSTRONG WILLIAM JOSEPH ,  SINHAROY BALARAM

IPC: G06F11/34 ,  G06F9/38

Abstract: An accounting method and logic for determining per-thread processor resource utilization in a simultaneous multi-threaded (SMT) processor provides a mechanism for accounting for processor resource usage by programs and thread s within programs. Relative resource use is determined by detecting instructio n dispatches for multiple threads active within the processor, which may inclu de idle threads that are still occupying processor resources. If instructions a re dispatched for all threads or no threads, the processor cycle is accounted equally to all threads. Alternatively if no threads are in a dispatch state, the accounting may be made using a prior state, or in conformity with ratios of the threads' priority levels. If only one thread is dispatching, that thread is accounted the entire processor cycle. If multiple threads are dispatching, but less than all threads are dispatching (in processors supporting more than two threads), the processor cycle is billed evenly acro ss the dispatching threads. Multiple dispatches may be detected for the threads and a fractional resource usage determined for each thread and the counters may be updated in accordance with their fractional usage.