Abstract: A Shared Data Access Serialization mechanism for sharing data among a plurality of systems while maintaining data integrity. User data is maintained on a primary and optionally an alternate data store. Each data store contains a set of lock blocks, one for each system sharing the data. The contents of the lock blocks, normally a time-of-day value, indicate system ownership status of the associated data. "Lock Rules" are disclosed for determining resource ownership, as well as a "lock stealing" mechanism for obtaining resource ownership from a temporarily stopped system. Suffix records and check records are used to insure data integrity. Error indications deduced from inconsistent suffix and/or check records are used to trigger a data recovery mechanism, and the recovery mechanism can synchronize a primary and secondary data store without the necessity of suspending access to the primary during the synchronization process.

Abstract: A tightly-coupled processor complex comprises two or more processors, the complex being asymmetric in that a feature available on one processor is not available on at least one other processor. A work selection mechanism selects one of a set of one or more ready work units, each capable of execution on one or more of the asymmetric processors. A processor set identification function identifies an "indirect idle" set of processors which can participate as hosts in work reassignment to make use of a previously idle processor, and identifies an "indirect bump" set of processors which can participate as hosts in work reassignment to displace a lowest priority work unit previously executing - any work reassignment being initiated by an assigner means and comprising an optimized number of work reassignment steps reassigning work among one of the processor sets to accomplish a related assignment goal (making use of a previously idle processor, or displacing a lowest priority work unit).

Abstract: A large data processing system includes a channel path management program that is periodically executed during operation of the system to gather control unit and subchannel utilization statistics and to monitor such statistics. If the degree of utilization of a channel path or an I/O device exceeds a predetermined threshold value, one or more change(s) is(are) dynamically made in the channel path configuration.

Abstract: A single entity contains source lines, being operated on by one or more end users. Within the same entity are entity version and level control data. Individual source lines contain version-related identification variables. After each version or level update by a user, a comparison is made between new and old versions; source line identification variables are modified, and new source lines are added; dependent version information is stored in the entity, and control data is updated. Subsequent retrievals of a version are responsive to the dependent version information, and produce indications of any changes that had been made to dependent versions.

Abstract: An installation creates a source I/O definition file (IODF), defining a current system I/O configuration, and a target IODF, defining a future I/O configuration. An activate function is initiated to change the system's hardware and software definition to correspond to the target IODF. A compare function compares the source and target IODFs, and creates a Configuration Change Block (CCB) to efficiently drive the change process. A validation function validates that the proposed change is valid, and the change process updates the hardware and software configuration definitions. If the change process is successful, a notification is sent; if not, a back-out process restores the I/O configuration to its initial state.

Abstract: An application, executing on a first processing element in a MP system without an asymmetric feature, issues an instruction requiring that feature to complete. A program check interruption gives control to interrupt handlers, which create a high-priority, non-preemptable work unit control block and enters the dispatcher to enqueue the work unit on a processor-related queue associated with a second processing element having the asymmetric feature. When the dispatcher executes in the second processing element, it executes the non-preemptable work unit, which transfers control to the application at the point of interruption. Subsequently the application has only whatever processor affinity obtained prior to the program check.

Abstract: A request is made by a system in a first logical partition, within a logically partitioned data processing system, to dynamically change the I/O configuration of the host system in a way that affects a system in a second logical partition. The hypervisor intercepts the request, ensures the serialization of such dynamic I/O requests, and allows dynamic reconfiguration to proceed. Subsequently, the hypervisor determines the effect of the reconfiguration on the second partition, and notifies the second partition of the change.

Abstract: A virtual lookaside facility maintains named data objects in class-related data spaces by virtual storage, for use by user programs. Concurrently executing service processes use local copies of an access token to access the data in the data spaces. Deletion of one of the data spaces invalidates the access token so that active processes will encounter a program check on access after data deletion. An error recovery means, recognizing the invalid token as cause of the program check, provides an appropriate return code to the invoker of the service process.

Abstract: Paging data is used to manage address space working set size, system multiprogramming level, and job mix, to improve system productive CPU utilization. System paging characteristics are monitored to determine when to perform management functions. When needed, address space paging data is collected for monitored address spaces, and target working set sizes set for managed address spaces. If necessary, address spaces are "forced" to exhibit characteristic paging patterns. Reallocation of working set pages among active address spaces, and swap in-swap out actions, are assessed, and impacts on productive CPU utilization is estimated.

Abstract: An installation creates a source I/O definition file (IODF), defining a current system I/O configuration, and a target IODF, defining a future I/O configuration. When the configuration definition is dynamically changed from the current to the target, a serialization mechanism--comprising a "device pin" technique and a "group serialization" technique to handle changed device group definitions--insures that data integrity is not lost on devices undergoing reconfiguration; and a change detection mechanism insures that changes to control structures affected by the dynamic reconfiguration, are noticed by programs accessing those control structures while the structures are changing.

Abstract: A cryptographic facility implements a multiple key part import procedure. The installation manager can verify that a key part has been correctly entered and has not been compromised. The security requirement for the procedure is that no single party can subvert the system security by misusing the procedure. This is accomplished by the use of a control-vector-dependent verification pattern to indicate that each key part has been accepted by using the proper control vector and the use of different key switch positions to specify whether the key part is a master key part or an operational key part and whether the key part is a first part or a subsequent key part. The apparatus provides an automatic reset of the key part register at the completion of each key-entry instruction so that each key part can be imported only once. This prevents the same key part from being imported twice as different key part types. The apparatus also prevents a key part from being combined with itself to create a known key.

Abstract: A switching error detection and recovery mechanism recognizes that an I/O configuration has changed. Comparing new configuration data to previous data maintained in a CDT table structure, and detecting unequal path information, an error condition is recognized. After recognizing the error condition, recovery actions are taken to notify the operator and insure data integrity. The CDT is structured so that redundant data is eliminated, and so records may be efficiently and dynamically added and deleted.

Abstract: A working key of a certain key type is to be transmitted from a first system (having a first usage-control value associated with keys of the certain type) and a second system (having a second usage-control value associated with keys of the certain type). A translation control value, associated with the certain key type, is generated, functionally relating the first and second usage-control values. The translation control value is used in a cryptographic function to send or receive the working key between systems, the cryptographic function being designed to produce valid results when the correct translation control value, and usage-control values, are employed, and unpredictable results otherwise. Effectively, the first usage-control value is translated to the second usage-control value.

Abstract: A non-main storage data space (NMDS) is presented, whose virtual pages may be backed by expanded storage, by a combination of expanded and auxiliary storage, but not by main storage. Mechanisms are provided to prevent direct referencing of data in a NMDS by instructions, and to bring data into and out of NMDS's. The I/O mechanisms minimize physical I/O by exchanging pointers to backing pages in appropriate cases. Additionally, control is returned immediately to I/O-initiating applications, when I/O is required between an address space and a NMDS, without the need for synchronization by the application: the mechanism provides for synchronization by the system's page fault processing.

Abstract: Page management mechanisms provide candidates for page stealing and prefetching from a main storage data cache of shared data when the jobs sharing the data are accessing it in a sequential manner. Pages are stolen behind the first reader in the cache, and thereafter at locations least likely to be soon re-referenced by trailing readers. A "clustering" of readers may be promoted to reduce I/O contention. Prefetching is carried out so that the pages most likely to be soon referenced by one of the readers are brought into the cache.

Abstract: A virtual lookaside faclity is provided for maintaining named data objects in class-related data spaces in virtual storage, readily retrievable by user programs. A search order is associated with each user, specifying an ordered list of "major names" which are, in effect, sequentially searched for a specified "minor name", or data object, to obtain a virtual storage copy of that data object. As data objects are placed into a virtual cache, existence information, implicit in the naming structure, is captured and saved. This information is relied on later in retrieving objects from the cache. The data isolation provided by maintaining class data and control blocks in individual data spaces is exploited to prevent failures relating to one class of objects from affecting the others, and to handle latent program users, following failures, effectively.An LRU-like trimming technique is used to remove less useful objects from the cache when cache space is fully utilized.

Abstract: A switching error detection and recovery mechanism recognizes that an I/O configuration has changed. Comparing new configuration data to previous data maintained in a CDT table structure, and detecting unequal path information, an error condition is recognized. After recognizing the error condition, recovery actions are taken to notify the operator and insure data integrity. The CDT is structured so that redundant data is eliminated, and so records may be efficiently and dynamically added and deleted.

Abstract: A linear data set is mapped to one or more non-main storage virtual data spaces. Portions of this data space are then selectively mapped to a "window" in an address space in which an application is executing, and changes made in this "window" are temporarily saved in the data space. After completion of processing, the application may permanently save changed data from the data space to the linear data set.The technique for mapping the data space to the address space may be used to map between two address spaces, and may be extended to encompass third and subsequent spaces, so that a reference to a mapped address in the first space will ultimately be interpreted as a reference to an address in the last mapped space.

Abstract: A facility for making dynamic changes to a system master key without stopping the system, and without loss of integrity to ongoing cryptographic operations. A version number is generated and associated with the current master key. A dynamic change is made to the master key, resulting in the then current master key becoming the old master key, and a "new" current master key (with a new version number) being placed into operation. Subsequent cryptographic requests using a supplied key enciphered under the old master key are identified by means of a supplied version number associated with the supplied key. This identification triggers a reencipher operation, reenciphering the supplied key under the now current master key--after which the cryptographic operation proceeds. Unique patterns are generated to verify the contents of the master key registers, and to authorize normal use of the cryptographic facility, and issuers of key-change operations.

Abstract: Within a data processing system, a control mechanism for supporting a data space without common segments in addition to traditional address spaces containing common segments. Logic for eliminating duplication of lookaside table entries for virtual addresses within shared segments, but not for identical virtual addresses within data address spaces is provided, as well as for overriding low address protection for store operations into data spaces. Thus, the entire virtual addressing range is available to programs wishing to use such data spaces for data isolation and data sharing.