公开(公告)号：US20060282620A1

公开(公告)日：2006-12-14

申请号：US11152557

申请日：2005-06-14

申请人： Sujatha Kashyap ,  Mysore Srinivas

发明人： Sujatha Kashyap ,  Mysore Srinivas

IPC分类号： G06F12/00

CPC分类号： G06F12/128

摘要： The present invention provides a method, system, and apparatus for communicating to associative cache which data is least important to keep. The method, system, and apparatus determine which cache line has the least important data so that this less important data is replaced before more important data. In a preferred embodiment, the method begins by determining the weight of each cache line within the cache. Then the cache line or lines with the lowest weight is determined.

摘要翻译： 本发明提供了一种用于向相关缓存进行通信的方法，系统和装置，该数据对于保持最不重要。 方法，系统和装置确定哪个高速缓存行具有最不重要的数据，以便在更重要的数据之前替换该较不重要的数据。 在优选实施例中，该方法通过确定高速缓存中每个高速缓存线的权重开始。 然后确定具有最低权重的高速缓存行或行。

公开(公告)号：US20050257020A1

公开(公告)日：2005-11-17

申请号：US10845549

申请日：2004-05-13

申请人： Sujatha Kashyap ,  Mysore Srinivas

发明人： Sujatha Kashyap ,  Mysore Srinivas

IPC分类号： G06F9/50 ,  G06F12/00

CPC分类号： G06F9/5072

摘要： A method, system, and program for dynamic memory management of unallocated memory in a logical partitioned data processing system. A logical partitioned data processing system typically includes multiple memory units, processors, I/O adapters, and other resources enabled for allocation to multiple logical partitions. A partition manager operating within the data processing system manages allocation of the resources to each logical partition. In particular, the partition manager manages allocation of a first portion of the multiple memory units to at least one logical partition. In addition, the partition manager manages a memory pool of unallocated memory from among the multiple memory units. Responsive to receiving a request for a memory loan from one of the allocated logical partitions, a second selection of memory units from the memory pool is loaned to the requesting logical partition. The partition manager, however, is enabled to reclaim the loaned selection of memory units from the requesting logical partition at any time.

摘要翻译： 用于逻辑分区数据处理系统中未分配存储器的动态存储器管理的方法，系统和程序。 逻辑分区数据处理系统通常包括多个存储器单元，处理器，I / O适配器和能够分配给多个逻辑分区的其他资源。 在数据处理系统内操作的分区管理器管理资源到每个逻辑分区的分配。 特别地，分区管理器管理多个存储器单元的第一部分到至少一个逻辑分区的分配。 此外，分区管理器管理多个存储器单元中的未分配存储器的存储器池。 响应于从所分配的逻辑分区之一接收对存储器借用的请求，来自存储器池的存储器单元的第二选择被借给请求逻辑分区。 然而，分区管理器可以在任何时间从请求的逻辑分区中回收借出的存储单元选择。

公开(公告)号：US20100275206A1

公开(公告)日：2010-10-28

申请号：US12427746

申请日：2009-04-22

申请人： Greg Mewhinney ,  Diane Flemming ,  David Whitworth ,  William Maron ,  Mysore Srinivas

发明人： Greg Mewhinney ,  Diane Flemming ,  David Whitworth ,  William Maron ,  Mysore Srinivas

IPC分类号： G06F9/46 ,  G06F15/00

CPC分类号： G06F9/5033 ,  G06F9/4843 ,  G06F9/50 ,  G06F9/5044

摘要： Standalone software performance optimizer systems for hybrid systems include a hybrid system having a plurality of processors, memory operably connected to the processors, an operating system including a dispatcher loaded into the memory, a multithreaded application read into the memory, and a static performance analysis program loaded into the memory; wherein the static performance analysis program instructs at least one processor to perform static performance analysis on each of the threads, the static performance analysis program instructs at least one processor to assign each thread to a CPU class based on the static performance analysis, and the static performance analysis program instructs at least one processor to store each thread's CPU class. An embodiment of the invention may also include the dispatcher optimally mapping threads to processors using thread CPU classes and remapping threads to processors when a runtime performance analysis classifies a thread differently from the static performance analysis.

摘要翻译： 用于混合系统的独立软件性能优化器系统包括具有多个处理器的混合系统，可操作地连接到处理器的存储器，包括加载到存储器中的调度器，读入存储器的多线程应用的操作系统和静态性能分析程序 加载到内存中; 其中所述静态性能分析程序指示至少一个处理器对每个所述线程执行静态性能分析，所述静态性能分析程序指示至少一个处理器基于所述静态性能分析将每个线程分配给CPU类，并且所述静态 性能分析程序指示至少一个处理器存储每个线程的CPU类。 本发明的实施例还可以包括调度器，当运行时性能分析将线程与静态性能分析不同的方式进行分类时，线程CPU类将线程最优地映射到处理器并将线程重新映射到处理器。

公开(公告)号：US20070101333A1

公开(公告)日：2007-05-03

申请号：US11260611

申请日：2005-10-27

申请人： Greg Mewhinney ,  Mysore Srinivas

发明人： Greg Mewhinney ,  Mysore Srinivas

IPC分类号： G06F9/46

CPC分类号： G06F9/526 ,  G06F9/485

摘要： A first collection of threads which represent a collection of tasks to be executed by at least one of a collection of processing units is monitored. In response to detecting a request by a thread among the first collection of threads to access a shared resource locked by a second thread among the collection of threads, the first thread attempts to access a list associated with the shared resource. The list orders at least one thread among the collection of threads by priority of access to the shared resource. In response to determining the list is locked by a third thread among the collection of threads, the first thread is placed into a sleep state to be reawakened in a fixed period of time. In response to determining that at least one of the collection of processing units has entered into an idle state, the first thread is awakened from the sleep state before the fixed period of time has expired. Also, in response to awakening the first thread from the sleep state, the first thread is assigned to at least one of the collection of processing units and the first thread retries its attempt to access the list.

摘要翻译： 监视表示要由处理单元的集合中的至少一个执行的任务的集合的线程的第一集合。 响应于在线程的第一集合中检测到线程的请求以访问由线程集合中的第二线程锁定的共享资源的请求，第一线程尝试访问与共享资源相关联的列表。 该列表通过访问共享资源的优先级来排序线程集合中的至少一个线程。 响应于确定列表被线程集合中的第三线程锁定，第一线程被置于休眠状态以在固定的时间段内被重新唤醒。 响应于确定处理单元的集合中的至少一个已经进入空闲状态，在固定时间段到期之前，第一线程从睡眠状态唤醒。 而且，响应于将第一线程从睡眠状态唤醒，第一线程被分配给处理单元集合中的​​至少一个，并且第一线程重试其访问列表的尝试。

公开(公告)号：US20110258468A1

公开(公告)日：2011-10-20

申请号：US12763431

申请日：2010-04-20

申请人： Vaijayanthimala K. Anand ,  William A. Maron ,  Mysore Srinivas ,  Diane Garza Flemming

发明人： Vaijayanthimala K. Anand ,  William A. Maron ,  Mysore Srinivas ,  Diane Garza Flemming

IPC分类号： G06F1/32

CPC分类号： G06F1/3203 ,  G06F1/324 ,  G06F9/5077 ,  G06F9/5094 ,  Y02D10/126 ,  Y02D10/22 ,  Y02D10/36

摘要： Handling requests for power reduction by first enabling a request for an amount of power change, e.g. reduction by any partition. In response to the request for power reduction, an equal proportion of the whole amount of power reduction is distributed between each of a set of cores providing the entitlements to the partitions, and the entitlement of the requesting partition is reduced by an amount corresponding to the whole amount of the power change.

摘要翻译： 通过首先启用功率变化量的请求来处理功率降低的请求，例如， 减少任何分区。 响应于减少功率的请求，在提供分配权利的一组核心的每一个之间分配全部功率减少的相等比例，并且请求分区的授权减少对应于 整个功率的变化量。

公开(公告)号：US20070033371A1

公开(公告)日：2007-02-08

申请号：US11197616

申请日：2005-08-04

申请人： Andrew Dunshea ,  Greg Mewhinney ,  Mysore Srinivas

发明人： Andrew Dunshea ,  Greg Mewhinney ,  Mysore Srinivas

IPC分类号： G06F12/00

CPC分类号： G06F12/12 ,  G06F12/0842 ,  G06F12/0862 ,  G06F2212/6024

摘要： A computer implemented method, apparatus, and computer usable code for managing cache information in a logical partitioned data processing system. A determination is made as to whether a unique identifier in a tag associated with a cache entry in a cache matches a previous unique identifier for a currently executing partition in the logical partitioned data processing system when the cache entry is selected for removal from the cache, and saves the tag in a storage device if the partition identifier in the tag matches the previous unique identifier.

摘要翻译： 一种用于在逻辑分区数据处理系统中管理高速缓存信息的计算机实现的方法，装置和计算机可用代码。 当选择高速缓存条目以从高速缓存中移除时，确定与高速缓存条目相关联的标签中的唯一标识符与高速缓存条目中的高速缓存条目中的唯一标识符是否匹配在逻辑分区数据处理系统中的当前执行分区的先前唯一标识符， 并且如果标签中的分区标识符与先前的唯一标识符匹配，则将标签保存在存储设备中。

公开(公告)号：US20060288186A1

公开(公告)日：2006-12-21

申请号：US11463100

申请日：2006-08-08

申请人： Jos Accapadi ,  Andrew Dunshea ,  Li Li ,  Grover Neuman ,  Mysore Srinivas ,  David Hepkin

发明人： Jos Accapadi ,  Andrew Dunshea ,  Li Li ,  Grover Neuman ,  Mysore Srinivas ,  David Hepkin

IPC分类号： G06F12/00

CPC分类号： G06F12/023

摘要： A system and method for dynamically altering a Virtual Memory Manager (VMM) Sequential-Access Read Ahead settings based upon current system memory conditions is provided. Normal VMM operations are performed using the Sequential-Access Read Ahead values set by the user. When low memory is detected, the system either turns off Sequential-Access Read Ahead operations or decreases the maximum page ahead (maxpgahead) value based upon whether the amount of free space is simply low or has reached a critically low level. The altered VMM Sequential-Access Read Ahead state remains in effect until enough free space is available so that normal VMM Sequential-Access Read Ahead operations can be performed (at which point the altered Sequential-Access Read Ahead values are reset to their original levels).

摘要翻译： 提供了一种基于当前系统内存条件动态更改虚拟内存管理器（VMM）顺序访问预读设置的系统和方法。 使用用户设置的顺序访问读取前值可以执行正常的VMM操作。 当检测到低内存时，系统会根据自由空间量是否很低或已经达到极低的水平，关闭顺序访问预读操作或者减小最大页面前提（maxpgahead）值。 改变的VMM顺序访问预读状态在有足够的可用空间可用之前保持有效，以便可以执行正常的VMM顺序访问预读操作（此时，改变的顺序访问读取前置值被重置为其原始级别） 。

公开(公告)号：US08407499B2

公开(公告)日：2013-03-26

申请号：US12763431

申请日：2010-04-20

申请人： Vaijayanthimala K Anand ,  Diane Garza Flemming ,  William A Maron ,  Mysore Srinivas

发明人： Vaijayanthimala K Anand ,  Diane Garza Flemming ,  William A Maron ,  Mysore Srinivas

IPC分类号： G06F1/32 ,  G06F1/24 ,  G06F9/00

CPC分类号： G06F1/3203 ,  G06F1/324 ,  G06F9/5077 ,  G06F9/5094 ,  Y02D10/126 ,  Y02D10/22 ,  Y02D10/36

摘要： Handling requests for power reduction by first enabling a request for an amount of power change, e.g. reduction by any partition. In response to the request for power reduction, an equal proportion of the whole amount of power reduction is distributed between each of a set of cores providing the entitlements to the partitions, and the entitlement of the requesting partition is reduced by an amount corresponding to the whole amount of the power change.

摘要翻译： 通过首先启用功率变化量的请求来处理功率降低的请求，例如， 减少任何分区。 响应于减少功率的请求，在提供分配权利的一组核心的每一个之间分配全部功率减少的相等比例，并且请求分区的授权减少对应于 整个功率的变化量。

公开(公告)号：US20070038809A1

公开(公告)日：2007-02-15

申请号：US11201642

申请日：2005-08-11

申请人： Jos Accapadi ,  Andrew Dunshea ,  Greg Mewhinney ,  Mysore Srinivas

发明人： Jos Accapadi ,  Andrew Dunshea ,  Greg Mewhinney ,  Mysore Srinivas

IPC分类号： G06F12/00

CPC分类号： G06F12/0897 ,  G06F12/0891

摘要： A computer implemented method, apparatus, and computer usable code for managing cache data. A partition identifier is associated with a cache entry in a cache, wherein the partition identifier identifies a last partition accessing the cache entry. The partition identifier associated with the cache entry is compared with a previous partition identifier located in a processor register in response to the cache entry being moved into a lower level cache relative to the cache. The cache entry is marked if the partition identifier associated with the cache entry matches the previous partition identifier located in the processor register to form a marked cache entry, wherein the marked cache entry is aged at a slower rate relative to an unmarked cache entry.

摘要翻译： 计算机实现的方法，装置和用于管理高速缓存数据的计算机可用代码。 分区标识符与高速缓存中的高速缓存条目相关联，其中分区标识符标识访问高速缓存条目的最后一个分区。 与高速缓存条目相关联的分区标识符与位于处理器寄存器中的先前分区标识符进行比较，以响应于高速缓存条目相对于高速缓存移动到较低级高速缓存。 如果与高速缓存条目相关联的分区标识与位于处理器寄存器中的先前分区标识符相匹配以形成标记的高速缓存条目，则标记高速缓存条目，其中标记的高速缓存条目相对于未标记的高速缓存条目以较慢的速率进行老化。

公开(公告)号：US20060155936A1

公开(公告)日：2006-07-13

申请号：US11006371

申请日：2004-12-07

申请人： Andrew Dunshea ,  Satya Sharma ,  Mysore Srinivas

发明人： Andrew Dunshea ,  Satya Sharma ,  Mysore Srinivas

IPC分类号： G06F13/00

CPC分类号： G06F12/0837 ,  G06F9/3004 ,  G06F9/30087 ,  G06F9/3834 ,  G06F9/3836 ,  G06F9/3857

摘要： A multiprocessor data processing system (MDPS) with a weakly-ordered architecture providing processing logic for substantially eliminating issuing sync instructions after every store instruction of a well-behaved application. Instructions of a well-behaved application are translated and executed by a weakly-ordered processor. The processing logic includes a lock address tracking utility (LATU), which provides an algorithm and a table of lock addresses, within which each lock address is stored when the lock is acquired by the weakly-ordered processor. When a store instruction is encountered in the instruction stream, the LATU compares the target address of the store instruction against the table of lock addresses. If the target address matches one of the lock addresses, indicating that the store instruction is the corresponding unlock instruction (or lock release instruction), a sync instruction is issued ahead of the store operation. The sync causes all values updated by the intermediate store operations to be flushed out to the point of coherency and be visible to all processors.

摘要翻译： 具有弱有序体系结构的多处理器数据处理系统（MDPS）提供处理逻辑，用于在运行良好的应用的每个存储指令之后基本上消除发出同步指令。 良好的应用程序的指令由弱有序的处理器转换和执行。 处理逻辑包括一个锁定地址跟踪实用程序（LATU），它提供一种算法和一个锁定地址表，当弱锁定处理器获取锁时，锁存地址被存储在该地址中。 当在指令流中遇到存储指令时，LATU将存储指令的目标地址与锁定地址表进行比较。 如果目标地址与其中一个锁定地址匹配，指示存储指令是相应的解锁指令（或锁定释放指令），则在存储操作之前发出同步指令。 同步使得由中间存储操作更新的所有值被刷新到一致性点，并且对于所有处理器可见。