Unisys Corporation

Thomas P Cooper

Dennis R Konrad

Michael J Hill

Thomas L Nowatzki

Ralph E Sipple

Michael J Heideman

John R Jordan

Ronald Q Smith

Robert E Swenson

James A Elke

Lester Y Lee

James R Mettling

James M Plasek

Lee-Chin Hsu Liu

James F Torgerson

Carl R Crandall

David A Novak

Craig B Johnson

Haeng D Park

Jennifer J Smith

Joseph E Kessler

Michael C Otto

Thomas N DeVries

Anthony P vonArx

Rajeev P Tandon

Michael J Bierma

Robert F Esson

Donald T Bordsen

Katsuhiko Kihara

Sheng-te Chiang

An apparatus for and method of implementing a cluster lock processing system having a relatively large number of commodity cluster instruction processors which are managed by a highly scalable, off the shelf communication processor. Because the commodity processors have virtually no system viability features such as memory protection, failure recovery, etc., the communication processor assumes the responsibility for providing these functions. The low cost of the commodity cluster instruction processors makes the system almost linearly scalable. Furthermore, having a fully scalable communication processor ensures a completely scalable system. The cluster/locking, caching, and mass storage accessing functions are fully integrated into a single hardware platform. The architecture may be implemented using any of a wide variety of proprietary or non-proprietary operating system environments.

Cluster lock server: ability to support multiple standard and proprietary locking protocols

An apparatus for and method of implementing a cluster lock processing system having a relatively large number of commodity cluster instruction processors which are managed by a highly scalable, off the shelf communication processor. Because the commodity processors have virtually no system viability features such as memory protection, failure recovery, etc., the communication processor assumes the responsibility for providing these functions. The low cost of the commodity cluster instruction processors makes the system almost linearly scalable. Furthermore, having a fully scalable communication processor ensures a completely scalable system. The cluster/locking, caching, and mass storage accessing functions are fully integrated into a single hardware platform.

Standard channel I/O processor (SCIOP)

Method and apparatus for tracking messages and transactions communicated between a number of users and a number of hosts, where the hosts are executing concurrent applications. During a first user session, a first user communicates with a first host. A counter associated with the first user session is incremented when the application creates an output message. The counter is decremented when the output message is released to the first user. A flag associated with the first user session is set and the counter is incremented once the flag is set, if an input message or committing transaction is received from the first user. The flag is cleared and the counter is decremented when the input message or committing transaction is processed by the concurrent application. The counter being decremented indicates there are no dependencies during the first user session with regard to the output message, input message, or committing transaction. If a second user session is initiated with a second host, the second host determines if there are any dependencies with the first user session. A dependency exists if the output message is created but not released, or if the input message or committing transaction is received but not processed. If there are dependencies, the first user session is reestablished to resolve the dependencies, otherwise, the second session is continued.

Protecting duplicate/lost updates against host failures

A system and method for providing transaction indivisibility in a transaction processing system through the use of commonly-accessible modules for monitoring and maintaining proper source message sequencing is provided. A source message is transmitted from the host processing unit upon recovery of a failure of the host processing unit, where the source message includes information destined for the database, and an identifying sequence number. The identifying sequence number is compared to a stored sequence number, where the stored sequence number is associated with an immediately preceding source message received prior to the failure of the host processing unit. A source message indivisibility failure is indicated where the identifying sequence number is not consecutive with respect to the stored sequence number, while the source message is added to a message execution queue if the identifying sequence number is consecutive with respect to the stored sequence number.

System for providing transaction indivisibility in a transaction

A system and method for assembling database restoration data according to transaction in a transaction processing system. Database restoration data is collated by chronologically storing updated database records in distinct storage banks partitioned according to its corresponding transaction. Resulting database record blocks from storage banks associated with completed database transactions are queued in the order that their corresponding transactions were completed. The queued database record blocks are transferred to storage media, whereby the queued database record blocks are arranged on the storage media according to transaction, and in the order that the active transactions were completed.

System and method for arranging database restoration data for efficient

Method and apparatus for providing backup memory storage for audit trail data within a computer system having a main memory storage, a non-volatile memory storage, and backup memory storage. The computer system executes a number of transaction programs which generate audit trail entries. As the audit trail entries are generated, a portion of the audit trail entries are stored in a portion of the main memory storage to create an audit trail. The audit trail entries are accumulated in the portion of the main memory storage until a request is received to write the portion of the main memory storage to a corresponding portion of the non-volatile memory. Subsequent portions of the audit trail entries are accumulated in subsequent portions of the main memory storage. An output transfer to the backup memory storage is made once a sufficient number of portions of the main memory storage have been written to corresponding portions of the non-volatile memory in order to maximize the transfer efficiency to the backup memory storage.

XPC backup for in-process audit

Method and apparatus for managing the allocation and release of memory space within a number of files shared by a number of hosts in a data processing system. A number of allocate and release tables are provided which are accessible by the number of hosts for managing the allocation and release of space within the number of files. Each one of the number of allocate and release tables corresponds to one of the number of files and provides a file location, file size, and indication to the number of hosts of which of the number of blocks within the file have been allocated and which of the number of blocks have been released. Any host may allocate one or more of the number of blocks to store messages created or received by the host by determining from the allocate and release tables which one or more of the number of blocks within one of the number of files are available. The host may then release the one or more of the number of blocks once the one or more of the number of blocks are no longer required to store the message. Once the one or more of the number of blocks are released, any of the number of hosts may allocate and release additional messages.

Shared file allocation and release

A system and method for organizing and managing recovery information on a per-transaction basis using distinct memory structures is provided. A transaction processing system includes one or more transaction request units for initiating data transactions with a database. A memory architecture includes storage modules to store transaction recovery information that is used in recovering from a transaction processing failure, and further includes transaction descriptor memory structures each of which correspond to a distinct transaction. The transaction descriptors allow access to the transaction recovery information to be isolated from other transaction recovery information. A control information field in each transaction descriptor stores a transaction identifier that identifies the distinct transaction to which it corresponds. The transaction descriptor also includes multiple information address fields, each of which includes a pointer to one of the storage modules that stores part of the transaction recovery information associated with the transaction identifier.

System and method for managing recovery information in a transaction

A data processing system including a first and second host, a first and second outboard file cache connected to the first host, and a first and second secondary storage device connected to the first host. The system operation includes the first host reading file data from the first or second secondary storage device after the data is cached on both the first and second outboard file caches. File data is updated by writing to both first and second outboard file caches. File data is destaged by writing data from the first outboard file cache only, to first and second secondary storage devices. Failure of a single outboard file cache is handled by the first host not reading and writing to the failed outboard file cache. Site-wide failure of first host, first outboard file cache, and first secondary storage device is handled by establishing communication from second host to both second outboard file cache and second secondary storage device and resuming processing.

Dual XPCS for disaster recovery

A computer system having dual outboard devices for generating audit trail sequence numbers and performing file locking. One embodiment includes a master and a slave outboard device, where lock request handling includes a lock request first sent to the master, lock grant awaited, followed by a request sent to the slave. Unlock request handling includes sending an unlock request to slave first, followed by sending an unlock request to the master. Obtaining a system sequence number includes always reading the slave outboard device, then reading the master outboard device. The computer system includes a method for switching host processors from a single outboard device mode to a dual outboard device mode and back again in the event of failure of one of the outboard devices.

Dual XPCS for disaster recovery in multi-host computer complexes

A system and method for balancing database transaction request distribution between various hosts in a multiprocessor transaction processing system is provided. The transaction processing system includes a database and multiple host processors each coupled to at least one database transaction request unit. Database transaction requests sent from the host processors are collectively entered into a commonly-accessible load balancing queue. Each database transaction request is accompanied by a source identifier that identifies the database transaction request unit which initiated the corresponding database transaction request. The queued database transaction requests from the load balancing queue are processed by currently-available host processors, regardless of which host processor initiated the database transaction request. A transaction request acknowledgment for each of the processed transaction requests is created and transferred to the database transaction request unit identified by their respective source identifiers.

Balanced input/output task management for use in multiprocessor

A system and method are described for caching files of data in a cache which is beyond the input/output boundary of a host. A host references a file with file access commands containing a logical file-identifier and a logical offset into the file. An outboard file cache coupled to the input/output section of the host receives the file access commands. The outboard file cache is transparent to users who program the host. Generation of input/output channel programs and mapping the data referenced to a physical address in secondary storage are eliminated when the referenced data is present in the cache. A file descriptor table in the outboard file cache identifies the logical portions of the logical files which are present in the cache. If the data referenced by the logical file-identifier and logical offset in a file access command is present in the outboard file cache, the data is transferred from the outboard file cache to the host memory. Otherwise, a miss status is returned to the host, and the host stages data from secondary storage to the outboard file cache. The outboard file cache further includes a lock table for storing file locks which inhibit access to selected files. In an outboard cache, excessive writes to a file are detected, the outboard cache having the first division for storage of selected portions of normal files, and a second division for storage of selected the second division is monitored and automatically converted to the first division when the storage used in the second division of cache memory falls below the periodic minimum.

Outboard file cache system

A system for concurrent processing of queries and transactions against a shared database. The system includes multiple processors wherein a processor is available for processing queries and another processor is available for concurrently processing transactions. A query buffer is established for performing the query search while the data accessed by transactions is available in a database cache. Control logic in a database management system distinguishes between transactions and queries and initiates file-read control for reading the file containing the database. File-read control contains logic for logical sequential reading and logical non-sequential reading of the file. Control structures provide a means for control over the load that the query is allowed to place on the system.

System for optimally processing a transaction and a query to the same

A system for parallel reading and processing of a file. The system includes multiple disks for storing the file. The disks are coupled to a data processing system via multiple input-output channels. A file buffer is established in the memory of the data processing system, wherein the file buffer is shared by an instruction processor that initiates a parallel read request and manipulates the file data once it is read, and multiple input-output processors that are coupled to the input-output channels. Multiple input requests are issued to the multiple disks to be processed in parallel. The input-output processors write file data to a first portion of the file buffer in parallel with the reading of a second portion of the file buffer by the instruction processor. Control structures provide a means for control over the input processing demands that the parallel read request is allowed to place on the system.

System for parallel reading and processing of a file

A transaction processing audit and recovery system is disclosed. After processing a transaction, an audit manager logs in audit records only changed blocks of data of a segment of a database. Upon failure of database backing storage, a prior copy of the database is reloaded to database backing storage that is available and a recovery manager reads the audit records and copies the changed blocks back to the database backing storage. An outboard file cache system is used in conjunction with the recovery manager to recover the database. The outboard file cache provides cache storage for segments of the database and writes non-contiguous blocks of one or more segments as directed in a single input/output request initiated from the recovery manager.

Growing community of inventors

New Brighton, MN, United States of America

Thomas P Cooper