International Business Machines Corporation

Richard I Baum

Glen A Brent

David Bruce Lindquist

Donald H Gibson

Casper A Scalzi

Michael Gerard Mall

Richard J Schmalz

Kenneth E Plambeck

Alan G Ganek

Bhaskar Sinha

James Walter Rymarczyk

Terry L Borden

James Lum

Thomas James Heller, Jr

Carl Edward Clark

Justin R Butwell

William Habacker

Charles H Brotman

Moon Jung Kim

Balakrishna Raghavendra Iyer

Satya Prakash Sharma

Inderpal Singh Narang

Ronald M Smith

Brent Allen Carlson

Michael Ignatowski

Lee Hardy Wilson

Julian Thomas

David Richard Page

Gururaj Seshagiri Rao

Hatem Mohamed Ghafir

Gururaj S Rao

Carol E Clark

Frederick E Sakalay

Hatem M Ghafir

Lee H Wilson

A computing system processes memory transactions for parallel processing of multiple threads of execution by support of which an application need not be aware. The computing system transactional memory support provides a Transaction Table in memory and a method of fast detection, of potential conflicts between multiple transactions. Special instructions may mark the boundaries of a transaction and identify memory locations applicable to a transaction. A 'private to transaction' (PTRAN) tag, directly addressable as part of the main data storage memory location, enables a quick detection of potential conflicts with other transactions that are concurrently executing on another thread of said computing system. The tag indicates whether (or not) a data entry in memory is part of a speculative memory state of an uncommitted transaction that is currently active in the system.

Computing system with optimized support for transactional memory

A multiprocessor computer system has a plurality of processing nodes which use processor state information to determine which coherent caches in the system are required to examine a coherency transaction produced by a single originating processor's storage request. A node of the computer has dynamic coherency boundaries such that the hardware uses only a subset of the total processors in a large system for a single workload at any specific point in time and can optimize the cache coherency as the supervisor software or firmware expands and contracts the number of processors which are being used to run any single workload. Multiple instances of a node can be connected with a second level controller to create a large multiprocessor system. The node controller uses the mode bits to determine which processors must receive any given transaction that is received by the node controller. The second level controller uses the mode bits to determine which nodes must receive any given transaction that is received by the second level controller. Logical partitions are mapped to allowable physical processors. Cache coherence regions which encompass subsets of the total number of processors and caches in the system are chosen for their physical proximity. A distinct cache coherency region can be defined for each partition using a a hypervisor.

Multiprocessor system with dynamic cache coherency regions

Makeup-eyeglasses

Makeup eyeglasses

A coexecutor for executing functions offloaded from central processors (CPs) in a data processing system, as requested by one or more executing control programs, which include a host operating system (host OS), and subsystem programs and applications executing under the host OS. The offloaded functions are embodied in code modules. Code modules execute in the coexecutor in parallel with non-offloaded functions being executed by the CPs. Thus, the CPs do not need to execute functions which can be executed by the coexecutor. CP requests to the coexecutor specify the code modules which are accessed by the coexecutor from host shared storage under the same constraints and access limitations as the control programs. The coexecutor may emulate host dynamic address translation, and may use a provided host storage key in accessing host storage. The restricted access operating state for the coexecutor maintains data integrity. Coexecutors can be of the same architecture or of a totally different architecture from the CPs to provide an efficient processing environment for the offloaded functions. The coexecutor interfaces host software which provides the requests to the coexecutor. Offloaded modules, once accessed by the coexecutor, may be cached in coexecutor local storage for use by future requests to allow subsequent invocations to proceed without waiting to again load the same module.

Coexecution processor isolation using an isolation process or having

A processor functioning as a coprocessor attached to a central processing complex provides efficient execution of the functions required for database processing: sorting, merging, joining, searching and manipulating fields in a host memory system. The specialized functional units: a memory interface and field extractor/assembler, a Predicate Evaluator, a combined sort/merge/join unit, a hasher, and a microcoded control processor, are all centered around a partitioned Working Store. Each functional unit is pipelined and optimized according to the function it performs, and executes its portion of the query efficiently. All functional units execute simultaneously under the control processor to achieve the desired results. Many different database functions can be performed by chaining simple operations together. The processor can effectively replace the CPU bound portions of complex database operations with functions that run at the maximum memory access rate improving performance on complex queries.

Apparatus for realigning database fields through the use of a crosspoint

Database engine predicate evaluator

Database engine

A processor functioning as a coprocessor attached to a central processing complex provides efficient execution of the functions required for database processing: sorting, merging, joining, searching and manipulating fields in a host memory system. The specialized. functional units: a memory interface and field extractor/assembler, a Predicate Evaluator, a combined sort/merge/join unit, a hasher, and a microcoded control processor, are all centered around a partitioned Working Store. Each functional unit is pipelined and optimized according to the function it performs, and executes its portion of the query efficiently. All functional units execute simultaneously under the control processor to achieve the desired results. Many different database functions can be performed by chaining simple operations together. The processor can effectively replace the CPU bound portions of complex database operations with functions that run at the maximum memory access rate improving performance on complex queries.

Allows instructions and data to be located in any one or more of plural sections of a large-size real memory of a data processing system. Any memory section is located by concatenating a conventional small real/absolute address with an address extender used with conventional small-size memory. A Central Processor Extended Address Mode (CPEAM) register content indicates the location of extenders in an AR(s), ASTE(s), STE(s) or PTE(s) for use by a central processor or I/O operations. An Input-Output Extended Address Mode (IOEAM) register content indicates the location of the extenders in ORB(s), CCW(s) or IDAW(s) for use by I/O operations. A compatible mode sets the content to zero for either or both of the CPEAM and IOEAM if either or both is not to be used.

System for addressing a very large memory with real or virtual addresses

A method and apparatus for translating a large logical address as a large virtual address (LVA) when a dynamic address translation (DAT) mode is on. Each LVA is separated into three concatenated parts: 1. a highest-order part (ADEN) for indexing into an access directory (AD) to locate an entry (ADE) for locating one access list (AL); 2. an intermediate part (ALEN) for indexing into a selected AL to access an entry (ALE) that enables location of an associated conventional address translation table which represents a conventional size virtual address space; and 3. a low-order DAT virtual address (VA) part having the same size as a conventional type of virtual address. The low-order DAT VA part is translated by the associated conventional address translation table. If a carry signal is generated during the creation of the low-order DAT VA part, then a change in the selection of an ALE results. An offset value of ALEs can be utilized to generate an effective ADEN and ALEN, which are utilized for the address translation of the LVA.

Large logical addressing method and means

A computer system has general purpose registers, control registers and access registers for containing information to allow address space capability. A linkage stack uses protected address space to store state information during program call and program return operations. The linkage stack contains information relating to state entries for the saved information and header and trailer entries to point to other linkage stack sections. A control register contains the pointer to the current linkage stack entry and is changed as the program call or return moves through the stack.

Linkage mechanism for program isolation

Growing community of inventors

Poughkeepsie, NY, United States of America

Richard I Baum