Sperry Corporation

Unisys Corporation

Robert E Swenson

Merlin L Hanson

Lawrence D Sasscer

Thomas P Cooper

Joseph E Kessler

Dennis R Konrad

Ralph E Sipple

Michael J Hill

James F Torgerson

James R Hamstra

Don M Robinson

John R Jordan

Benjamin T George

Vladi Pusic

Anthony P vonArx

Arnold R Schmalzbauer

Larry J Kelson

Robert F Esson

Donald T Bordsen

Kevin W Bengtson

Anthony R Talarczyk

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

An outboard file cache system including file surge protection and an improved method for managing allocation of cache storage are disclosed. In an outboard cache, excessive writes to a file are detected by testing whether a segment belonging to a file specified by a nd, which logically precedes one of the segments specified in the command by a predetermined number of segments, is present in the outboard file cache, has been written, and has not been destaged. When a first surge-threshold is reached, a group of segments is selected and destaged. When a second surge-threshold is reached, the outboard cache inhibits allocation of further cache storage for the file for the purpose of writing until a selected group of segments is destaged.

Cache system with file surge protection and method for managing

The disclosure relates to a cache management method for a post-store cache. Recently referenced units of storage in the cache are identified and a round robin cache replacement method is generally used to select a unit of storage for replacement if the unit of storage has not been recently referenced. Recently used segments are identified by maintaining a Recently Used Zone ahead of the storage location presently considered for replacement. For each reference to a unit of storage in the cache which results in a miss, a portion of the cache is scanned for units of storage to destage.

Enhanced least recently used round robin cache management method and

A cache management method in a cache system having a plurality of processors for managing the cache storage. Each of the processors may reserve portions of the cache storage which future allocation of cache storage may be expedited. All the processors begin searching for a portion of cache storage to allocate with the same starting portion of cache storage when processing cache commands in which the referenced data is not present in the cache storage. A processor will advance past the starting portion of cache storage when allocation of a portion of cache storage is necessary and another processor is currently using the starting portion of cache storage.

Multiprocessor system with scheme for managing allocation and

A multiprocessor data processing system is implemented with processors, each of which may request for a temporary time the exclusive lock on an object which is stored on a data base. To achieve this a lock processor synchronizes the locking and unlocking of the objects. The requesting processor directs the storage of the object from the data base into a selected high performance storage unit, where it has exclusive rights to modify or write into the object until the object is unlocked by the processor. An audit tape or disk records all modifications made to any object during a transaction. A non-volatile cache memory is inserted in the audit trail to store a before-look image of the object that resides in the high performance storage unit. Data compaction occurs by comparison of the before-look image with an after-look image to provide a difference image, which is supplied to an audit buffer that is coupled to the audit tape. The locking processor may unlock the secured object once the after-look image has been committed from either a stored version in the non-volatile cache or from a high performance main memory unit to the data base disk. The difference image and the after-look image associated with the difference image may then be stored in the non-volatile cache, and provided to the audit tape or disk and the data base disk in a sequence which is independent of the operating sequence of the requesting processor.

Cache memory with data compaction for use in the audit trail of a data

In a cache/disk sybsystem employing the post-store technique, each file is assigned a file number. The file number is an address in a directory on a disk, this address containing pointers to the physical disk space assigned to that file. During normal operation of the subsystem a device number and address are utilized to transfer written-to segments of data from a cache store to a disk. If a failure occurs which prevents the cache-to-disk transfer of a segment of data, a storage control unit forms a status word and a sense message. The sense message includes the file number and information defining the fault. A host processor is then notified that the status word is available. The host processor obtains the status word and indexes into a status action branch table. The branch table entry directs the host processor to issue a Sense I/O command to obtain from the storage control unit the sense message identifying the fault. The executive program of the host processor may subsequently utilize the sense message, which also includes the file number, to assign new disk space at which the segment of data may be recorded. The segment of data which could not be written to the disk is first read into the host processor. The host processor then issues a command to write the segment of data into the newly assigned disk space.

Cache/disk subsystem with file number for recovery of cached data

In a data processing system including a processor, a bulk memory, a cache, and a storage control unit for controlling the transfer of data between the bulk memory and the cache, a timestamp is generated with each write command. A linked list is maintained, having an entry therein corresponding to each segment in the cache which has been written to since it was moved from the bulk memory to the cache. The timestamp accompanying a write command which is the first command to write to a segment after that segment is moved from bulk memory to the cache is entered into the list at the most recently used position. An entry in the linked list is removed from the list when the segment corresponding thereto is transferred from the cache to the bulk memory. The linked list is utilized to update a value TOLDEST, which represents the age of the oldest written-to segment in the cache that has not been returned to bulk memory since it was first written to. The processor periodically issues a command which transfers TOLDEST from the subsystem to the processor. In case of a cache failure, such as might result from a power loss, the processor may sense the latest value of TOLDEST together with other file recovery synchronization information and determine what part of the data which it sent to the cache was lost because it did not get recorded in the bulk memory.

Processor-addressable timestamp for indicating oldest written-to cache

In a system having a host processor connected through a storage control unit to a cache store and a plurality of disk devices, segments of data which have been written to, while resident in the cache store, are transferred to the disks at some later time. If an abnormal condition, such as a bad spot on the disk, prevents the transfer of a segment from the cache store to the disk, an indicator is set for that segment by the storage control unit to prevent further attempts to transfer the segment. A segment whose indicator is set remains in the cache store until the host processor issues an initialize or a reset segment command.

Cache/disk subsystem with floating entry

In a data processing system having a host processor, a cache store for storing segments of data, a bulk memory and a storage control unit for controlling transfers between the processor, cache store and bulk memory, the storage control unit normally responds to a read or write command from the host processor to control the transfer of data. If a copy of the data transferred is not resident in the cache store then a copy is written therein by the storage control unit. If the length of a data transfer exceeds a first threshold length then the storage control unit does not write a copy of the data into the cache store. If the length of a data transfer exceeds a second threshold length, and the transfer begins on a segment boundary and comprises an integral number of segments, then the storage control unit does not write a copy of the data into the cache store. The writing into the cache store is transparent to the host processor. The use of a transfer threshold prevents data from being entered into the cache store when it is not likely to be used again soon. Two thresholds are provided because the data transferred in long transfers of an integral number of segments is even less likely to be used again soon than the data transferred in long transfers that do not begin on a segment boundary or comprise an integral number of segments.

Cache/disk subsystem with cache bypass

One or more host processors issue commands to one or more storage control units which control data transfers between the host processors, a cache store and a plurality of disk devices. A command queue is maintained for each disk device to store commands waiting to be executed by the disk device. The cache store stores segments of data which have been read from, or are to be written to disk space. In response to a command from a host processor a corresponding command is added to one of the command queues. If the disk device is not busy and has no previously queued commands waiting to be executed the storage control unit issues a seek command to the disk drive device. If there are previously queued commands waiting to be executed, or if the disk device is busy, the cache store is checked to determine if it contains a copy of the data from the disk space specified by the host processor command. If a copy of the data from the specified disk space is resident in the cache store then a data transfer is initiated between the host processor and the cache store. A priority value and a sequence number are assigned to each command as it is added to a queue so that the highest priority queued command with the lowest request number is executed when the disk device corresponding to the queue becomes idle. The storage control unit may create commands and place them on command queues for execution, these commands being for the purpose of transferring the least recently used segments of data from the cache store to the disks if the segments have been written to while in the cache store.

Hierarchical memory system having cache/disk subsystem with command

A cache/disk subsystem includes a host processor, a cache store, a disk drive device for driving a disk, and a storage control unit for controlling the transfer of data between the host, disk and cache store. The cache store holds segments of data which have been read from, or are to be written to, the disk. When the host issues a normal WRITE command to the storage control unit and none or a part only of the data from the disk space specified by the command is resident in the cache store, the segment or segments of data in the disk space are staged into the cache store and overlayed with data from the host. A directory in the host keeps track of the disk segments previously written to. When the host is ready to issue a WRITE command it checks the directory and, if the segment or segments to be written to have not previously been written to, then the host issues an ACQUIRE WRITE command. The storage control unit includes controls responsive to an ACQUIRE WRITE command for bypassing the staging operation. The storage control unit also includes circuits for converting a normal WRITE command to an ACQUIRE WRITE command when the beginning and ending addresses of a normal WRITE command fall on segment boundaries. The ACQUIRE WRITE command eliminates unnecessary staging from the disk to the cache store in situations where there is no data to be staged or all of the data which would be staged would be overwritten.

Cache/disk subsystem with acquire write command

In a system wherein processors are connected through channel units to disk devices, two storage control units (SCU's) are provided between the channel units and the disk devices. The SCU's receive commands from the processor and issue seek instructions to the devices to locate the disk space specified by the commands, provided that a copy of the data from the specified disk space is not resident in a cache store. The commands may be queued for a time in a memory which is accessible by both SCU's. Before issuing a seek instruction to a device, an SCU first determines whether it is the only SCU having a path through a channel unit back to the processor which issued the command or whether there is a path from each SCU to the processor. If both SCU's have a path back to the processor, the SCU issues an untagged seek instruction to the device. If not, the SCU issues a tagged seek instruction to the device. When the disk device completes a seek resulting from a tagged instruction, it sends an interrupt to the SCU which issued the seek and that SCU then notifies the processor which issued the command. In response to the notification the processor issues the command a second time and it is then executed under the control of the SCU. When a device completes a seek resulting from an untagged seek instruction it sends an interrupt to both SCU's and either one may respond to it.

Cache/disk subsystem with load equalization

In a data processing system having plural disk devices connected to first and second storage control units, and plural processors connected to the storage control units by connection paths, a table is provided having entries defining the connection paths between each storage control unit and each processor. The storage control units may queue commands from the processors in a command queue for later execution. Either storage control unit may utilize a queued command to issue a seek signal to a disk device and report status to the processor which issued the command. When a disk device completes a seek it issues an interrupt signal which may be sampled by either storage control unit to report further status. In order to report status to the processor which issued the command, a storage control unit searches the table defining the connection paths between the storage control units and the processors and applies the status to the first available connection path that it finds. If one storage control unit, upon searching the table, finds that there is no connection path between itself and the processor which issued the command, it returns to an idle condition so that the other storage control unit may respond to the interrupt signal, search the table, and apply the status indication to the host processor which issued the command.

Multiprocessor system with cache/disk subsystem with status routing for

One or more storage control units (SCUs) interface one or more channels with a plurality of disk drives. One or more host processors are connected to the channels. Associated with the SCUs are one or more cache storage units (CSUs) having a relatively short access time as compared to direct accessing of the disk drives. The CSUs are transparent to the user who programs the hosts as though he were addressing the disk drives directly. An SCU converts the disk address (starting address and number of words to be transferred) to a CSU address and checks a segment descriptor table (SDT) to find out if the segment or segments involved in a transfer are present in a CSU. If so, the data is transferred from the CSU to the channel. If the data is not present in a CSU, then for a read transfer command the data is obtained from the disk drive, entered into the CSU, and supplied to the host. For a write command involving a partial segment, the segment is staged from disk to CSU and the host data overlays the appropriate portion of the segment in CSU without writing the segment back to disk. The post-store method is utilized for destaging to the disk drives segments of data which have been written to while in the CSU.

Cache/disk subsystem

A cache/disk subsystem includes a storage control unit, a relatively low capacity high speed cache store, and a higher capacity slower memory such as a plurality of disk drive devices. The storage control unit controls the subsystem to transfer from the cache store to the disk drive devices segments of data which have been modified while resident in the cache store and insures that space will be available in the cache store for new data if a particular operation requires that new data be transferred from a disk to the cache store for use. The subsystem may include plural storage control units and any of them may control the transfers of segments of data to the disks.

Growing community of inventors

Mendota Heights, MN, United States of America

Robert E Swenson