Abstract: A message processing facility provides a send function for sending messages to designated locations. The facility allows a default target for a message, initiates parallel execution, adds additional processes to an already parallel execution, determines that a problem has completed and performs the appropriate completion actions. The message processing facility enables coordination of the results of a network that has been widened in its parallelism, without creating subproblems. Processes within the parallel network do not need to be aware of the topology or the position of a given process within the topology. The width of the parallelism involved in a particular request is tracked, and therefore, the appropriate number of results expected can be determined automatically and dynamically. Therefore, despite dynamic changes in topology, completion can be determined without awareness of the processes that create the results.

Abstract: One system of a multisystem environment takes over log entries owned by another system of the environment. When a system owning entries on a primary log stream is inactive (or a sync point manager on the system is inactive), another system of the multisystem environment writes the log entries corresponding to the inactive system to other locations. The entries within the other locations are not owned by any system. Thus, the resource managers associated with those log entries are eligible for restarting on any system of the multisystem environment. However, all resource managers involved in a common set of transactions are to restart on the same system. The other resource managers can restart on another system. When one of the resource managers restarts on a system, the system takes back ownership of the entries associated with the common set of transactions.

Abstract: A laser-writable information storage blank comprises a substantially planar disk made of a material which is susceptible to modification by impingement of laser radiation. The blank medium includes a metallic track pattern, preferably in the form of a spiral pattern. The track pattern provides two reference edges from which exact distances may be determined for writing and reading binary information stored between adjacent track portions.

Abstract: One system of a multisystem environment takes over log entries owned by another system of the environment. When a system owning entries on a primary log stream is inactive (or a sync point manager on the system is inactive), another system of the multisystem environment writes the log entries corresponding to the inactive system to other locations. The entries within the other locations are not owned by any system. Thus, the resource managers associated with those log entries are eligible for restarting on any system of the multisystem environment. However, all resource managers involved in a common set of transactions are to restart on the same system. The other resource managers can restart on another system. When one of the resource managers restarts on a system, the system takes back ownership of the entries associated with the common set of transactions.

Abstract: A d.c. motor together with a hot gas bypass valve is incorporated into a cooling system specifically designed for removing heat from computer systems. Unlike typical refrigeration systems, the cooling system herein runs continuously and responds to changes in thermal load. This allows the unit to operate within a wide range of ambient conditions and at various thermal load levels unlike other systems which were capable of operation only at a single, predesigned load level. The cooling system is modular and is easily added to or removed from a redundant system which includes a single evaporator with multiple refrigerant loops which provides yet another aspect of continuous operation due to the inherent redundancy thus provided.

Abstract: Frequency differences between differing clock sources are compensated for by an adaptive filtering mechanism. An amount of frequency drift between two clock sources is determined. Then, based on that amount of frequency drift, a filtering value is selected to be used in tracking the frequency drift. If the frequency drift is determined to be large, then a minimum filtering value is selected. However, if it is determined to be small, then a maximum filtering value is selected. The selected filtering value is used to adjust the address(es) of one or more data bits being transmitted and received using the two clock sources, such that the frequency drift is properly tracked.

Abstract: A redundant refrigeration system is used in conjunction with a dual passage cold plate to provide reliable cooling for electronic circuit modules particularly to such modules which incorporate computer processors for which long term "up time" is very highly desirable. Separate compressor-expansion device-condenser refrigeration systems are coupled to a single cold plate which maintains flow-wise isolation between refrigerant streams while at the same time providing close thermal coupling.

Abstract: A logical positioning mechanism is provided within a storage controller, thus enabling the storage controller to determine which logical position in a storage device is to be accessed next. The logical positioning mechanism includes one or more logical position indicators defined, maintained and controlled by the storage controller. The logical position indicators include the address of the logical position to be next accessed and are incremented and decremented by the storage controller depending upon the operation accessing the logical position.

Abstract: A method and system are provided for executing a service processor request within a data processing system having one or more processors within a central processing complex, each of the processors within the central processing complex including allocatable processor resources. Each of the processors within the central processing complex is provided with the capability of processing selected service processor requests by reserving a portion of the allocatable processor resources within each of the processors for such purpose. A service processor request within the central processing complex is initially processed utilizing at least one of the processors in response to receiving a service processor request, if sufficient processor resources are available to process the service processor request within the reserved portion of the allocatable resources.

Abstract: An access method and system in a logically partitioned data processing system having a plurality of logical partitions. A requesting logical partition requests information pertaining to a resource of the data processing system as the resource relates to a requested logical partition of the system. The requesting logical partition is given access to the information when the requesting and requested logical partitions are authorized to access the resource. Thus, access to logical partition information is provided on a per resource basis. Additionally, access is limited such that the requesting logical partition can access information relating to less than all of the logical partitions of the logically partitioned data processing system.

Abstract: In a multiprocessing system, data which is stored on an external storage media and shared by a plurality of processors is converted from an old format to a new format without bringing the multiprocessing system down. One processor at a time is provided the capability to view the data in the new format, while the other processors view the data in the old format. After all of the processors are capable of viewing the data in the new format, the data is converted from the old format to the new format and the data in the new format is stored on the external storage media. If for some reason the new format turns out to be unacceptable, procedures are also in place for falling back to the old format without disrupting system availability.

Abstract: When the hardware needs to perform a disruptive hardware action, such as a stop of one or more central processors, the hardware provides advance notification to all affected operating systems. The advance notification includes the time that the proposed disruptive action is intended to be initiated and its projected duration. In one instance, the hardware does not proceed with the disruptive action until it receives return acknowledgment from the operating systems that the disruptive action can be performed. The operating systems may respond back to the hardware to proceed as planned with the disruptive action or to reinitiate it only after a specified delay. This allows the operating systems time to prepare for any consequences of the disruptive action.