International Business Machines Corporation

Robert H Bell, Jr

Hong Lam Hua

Michael Jay Shapiro

Louis Bennie Capps, Jr

William John Starke

Mysore Sathyanaranyana Srinivas

Venkat Rajeev Indukuru

Guy L Guthrie

Jeffrey A Stuecheli

Wen-Tzer Thomas Chen, Jr

Rajendra D Panda

Michael D Roll

Donald R DeSota

Thomas Edward Cook

Men-Chow Chiang

Madhavi Gopal Valluri

Pattabi Michael Seshadri

Michael A Paolini

Hugh Shen

Robert Alan Cargnoni

William A Maron

Octavian Florin Herescu

Vernon Walter Miller

Ronald Edward Newhart

Luigi Brochard

Jose Raul Escalera

Derek E Williams

Leo James Clark

Ram Raghavan

Luis Alfonso Lastras-Montano

Jason Frederick Cantin

Mysore Sathyanarayana Srinivas

Sameh Sherif Sharkawi

Olga Yiparaki

Joseph H Robichaux

Francois Thomas

Thomas W Chen, Jr

Daniel Mark Dreps

Naresh Nayar

Richard James Eickemeyer

Diane Garza Flemming

Glenn Graham Daves

Steven Paul Hartman

Sergio Reyes

Matthew Accapadi

Bradly George Frey

William Elton Burky

Mathew Accapadi

John-David Wellman

Herman Dietrich Dierks

Bernadette Ann Pierson

Thomas J Dewkett

Alex E Mericas

Thomas Michael Capasso

Donald R Desota

Danie M Dreps

Pattabi R Seshadri

Louis B Capps

According to one aspect of the present disclosure a method and technique for managing data transfer includes receiving and storing a plurality of different data patterns anticipated to be encountered by a processor unit of a data processing system corresponding to a particular application being processed. Responsive to receiving a read request for data, the requested data is read from a memory subsystem, and the read data is compared by the memory subsystem to the stored data patterns. Responsive to determining that the read data matches at least one of the stored data patterns, the memory subsystem replaces the matching read data with a pattern tag corresponding to the matching data pattern. The pattern tag is transmitted over a communication link in response to the request.

System, method and computer program product for data transfer management

According to one aspect of the present disclosure a method and technique for managing data transfer includes receiving and storing a plurality of different data patterns anticipated to be encountered by a processor unit of a data processing system corresponding to a particular application being processed. Responsive to receiving a read request for data, the requested data is read from a memory subsystem, and the read data is compared by the memory subsystem to the stored data patterns. Responsive to determining that the read data matches at least one of the stored data patterns, the memory subsystem replaces the matching read data with a pattern tag corresponding to the matching data pattern. The pattern tag is transmitted to the processor unit instead of the requested data as a response to the read request, and the processor unit replaces the pattern tag with the corresponding data pattern.

Techniques for preserving memory affinity in a computer system is disclosed. In response to a request for memory access to a page within a memory affinity domain, a determination is made if the request is initiated by a processor associated with the memory affinity domain. If the request is not initiated by a processor associated with the memory affinity domain, a determination is made if there is a page ID match with an entry within a page migration tracking module associated with the memory affinity domain. If there is no page ID match, an entry is selected within the page migration tracking module to be updated with a new page ID and a new memory affinity ID. If there is a page ID match, then another determination is made whether or not there is a memory affinity ID match with the entry with the page ID field match. If there is no memory affinity ID match, the entry is updated with a new memory affinity ID; and if there is a memory affinity ID match, an access counter of the entry is incremented.

Technique for preserving memory affinity in a non-uniform memory access data processing system

A multiprocessor system having plural heterogeneous processing units schedules instruction sets for execution on a selected of the processing units by matching workload processing characteristics of processing units and the instruction sets. To establish an instruction set's processing characteristics, the homogeneous instruction set is executed on each of the plural processing units with one or more performance metrics tracked at each of the processing units to determine which processing unit most efficiently executes the instruction set. Instruction set workload processing characteristics are stored for reference in scheduling subsequent execution of the instruction set.

Multicore processor and method of use that configures core functions based on executing instructions

According to one aspect of the present disclosure a method and technique for managing data transfer is disclosed. The method includes comparing, by a processor unit of a data processing system, data to be written to a memory subsystem to a stored data pattern and, responsive to determining that the data matches the stored data pattern, replacing the matching data with a pattern tag corresponding to the matching data pattern. The method also includes transmitting the pattern tag to the memory subsystem.

Execution of a store instruction to modify an instruction at a memory location identified by a memory address is requested. A cache controller stores the memory address and the modified data in an associative memory coupled to a data cache and an instruction cache. In addition, the modified data is stored in a second level cache without invalidating the memory location associated with the instruction cache.

Virtual unifed instruction and data caches including storing program instructions and memory address in CAM indicated by store instruction containing bit directly indicating self modifying code

According to one aspect of the present disclosure, a method and technique for performance-driven cache line memory access is disclosed. The method includes: receiving, by a memory controller of a data processing system, a request for a cache line; dividing the request into a plurality of cache subline requests, wherein at least one of the cache subline requests comprises a high priority data request and at least one of the cache subline requests comprises a low priority data request; servicing the high priority data request; and delaying servicing of the low priority data request until a low priority condition has been satisfied.

Performance-driven cache line memory access

According to one aspect of the present disclosure a system and technique for performance-driven cache line memory access is disclosed. The system includes: a processor, a cache hierarchy coupled to the processor, and a memory coupled to the cache hierarchy. The system also includes logic executable to, responsive to receiving, a request for a cache line: divide the request into a plurality of cache subline requests, wherein at least one of the cache subline requests comprises a high priority data request and at least one of the cache subline requests comprises a low priority data request; service the high priority data request; and delay servicing of the low priority data request until a low priority condition has been satisfied.

A method utilizes information provided by performance monitoring hardware to dynamically adjust the number of levels of speculative branch predictions allowed (typically 3 or 4 per thread) for a processor core. The information includes cycles-per-instruction (CPI) for the processor core and number of memory accesses per unit time. If the CPI is below a CPI threshold; and the number of memory accesses (NMA) per unit time is above a prescribed threshold, the number of levels of speculative branch predictions is reduced per thread for the processor core. Likewise, the number of levels of speculative branch predictions could be increased, from a low level to maximum allowed, if the CPI threshold is exceeded or the number of memory accesses per unit time is below the prescribed threshold.

Performance of processors is improved by limiting number of branch prediction levels

Some embodiments of the inventive subject matter are directed to operations that include determining that an access request to a computer memory results in a cache miss. In some examples, the operations further include determining an amount of cache resources used to service additional cache misses that occurred within a period prior to the cache miss. Furthermore, in some examples, the operations further include servicing the access request to the computer memory based, at least in part, on the amount of the cache resources used to service the additional cache misses within the period prior to the cache miss.

Dynamic prioritization of cache access

A method, system and computer-usable medium are disclosed for managing transient instruction streams. Transient flags are defined in Branch-and-Link (BRL) instructions that are known to be infrequently executed. A bit is likewise set in a Special Purpose Register (SPR) of the hardware (e.g., a core) that is executing an instruction request thread. Subsequent fetches or prefetches in the request thread are treated as transient and are not written to lower-level caches. If an instruction is non-transient, and if a lower-level cache is non-inclusive of the L1 instruction cache, a fetch or prefetch miss that is obtained from memory may be written in both the L1 and the lower-level cache. If it is not inclusive, a cast-out from the L1 instruction cache may be written in the lower-level cache.

Performance of emerging applications in a virtualized environment using transient instruction streams

According to one aspect of the present disclosure a method and technique for managing memory access is disclosed. The method includes setting a memory databus utilization threshold for each of a plurality of processors of a data processing system to maintain memory databus utilization of the data processing system at or below a system threshold. The method also includes monitoring memory databus utilization for the plurality of processors and, in response to determining that memory databus utilization for at least one of the processors is below its threshold, reallocating at least a portion of unused databus utilization from the at least one processor to at least one of the other processors.

Reallocating unused memory databus utilization to another processor when utilization is below a threshold

Provided are a computer program product, system, and method for determining server write activity levels to use to adjust write cache size. Server write activity information on server write activity to the cache is gathered. The server write activity information is processed to determine a server write activity level comprising one of multiple write activity levels indicating a level of write activity. The determined server write activity level is transmitted to a storage server having a write cache, wherein the storage server uses the determined server write activity level to determine whether to adjust a size of the storage server write cache.

Determining server write activity levels to use to adjust write cache size

Growing community of inventors

Austin, TX, United States of America

Robert H Bell, Jr