摘要:
Provided are a method, computer program product and system for providing indexed queue support for efficient retrieval of messages from a queue. An index key for expediting message retrieval is assigned to a message when the operation of placing the message on a queue is committed. The index key assigned at commit time comprises an attribute (such as a message ID or correlation ID) specified by the sending application program, which placed the message on the queue. This deferred assignment of an index key until commit time means that the index key can be used to search for messages having the particular attribute without any possibility of identifying messages for retrieval before the messages have been committed. This maintains transactional requirements of a transaction-oriented messaging system which requires a message to only be made available for retrieval by receiver application programs after the sender application's put operation has committed.
摘要:
Resource contention monitoring and analysis are provided for a single system or multi-system environment. Resource contention is tracked at a task level when contention occurs by listing each owner of a resource in a blocker list and each waiter for the resource in a waiter list. Each list is ordered oldest to newest so each instance of contention is added to the end of the respective list. The resource request and contention lists are cross-linked to provide easy coordination between the resource queues and the contention lists. Techniques for analyzing contention using the lists, including deadlock analysis, are described.
摘要:
A method of assigning and releasing locks, in a network computing environment having a plurality of central processing units coupled to a high performance locking facility. The method comprises of the steps of first processing multiple requests for locking operations simultaneously and then recording lock names and lock states in entry records of a lock table based on the requests. When a request is being processed, providing high-speed searching to search any particular lock entry in the table and altering and modifying the table accordingly. Finally an appropriate response and the status of the lock state will be generated and sent to other requestors requesting the lock.
摘要:
The present invention provides for a computer program product and device including instructions executable by a digital processing apparatus for conducting a high-performance locking facility. At first multiple requests for locking operations are processed and received simultaneously and their lock names and lock states are recorded in entry records of a lock table based on the requests. Next high-speed searching of any particular lock entry in this table is performed followed. The table can be altered and modified at this time as well accordingly depending upon lock status that is being processed. Finally an appropriate response is generated and the status of the lock state is provided to those requesting the lock.
摘要:
A LOCK PAGE instruction is provided for locking a page of real storage using a virtual address. The LOCK PAGE instruction includes an operation code which specifies the operation to be performed, a first operand which contains the value of the real address obtained during execution of the LOCK PAGE instruction and a second operand which contains the value of the virtual address of the page to be locked during execution of LOCK PAGE. LOCK PAGE enables an address page or a data page to be locked without requiring the entire address space to be locked. A page is locked when a lock control bit in general register 0 is zero, the page is valid in main storage and a lock bit located in a page table entry is zero. In addition to the above, the technique used by an operating system service for reclaiming a page of real storage is altered to intersect with the LOCK PAGE facility.
摘要:
A data processing system, having virtual addressing capability, has a real storage manager to associate virtual storage locations with real storage by accessing page tables to determine the locations of "backed" virtual storage pages in central, expanded, or auxiliary storage. The real storage manager accepts requests to copy ranges of virtual storage from one virtual storage range to another, and, in so doing, uses the page tables to effectively perform the copying by reassigning backed pages from the source range to the target range. Having been reassigned to the target range, the backed pages are artificially aged by increasing the unreferenced interval count (UIC), so that the backed pages in the target range will be likely candidates for page stealing.
摘要:
An application is included in a virtual machine sent to a cloud computing server. The cloud computing server has a remote access layer that fetches data blocks of the private dataset of the application from private data storage as the data blocks are requested by the application, so that the application in the public cloud begins execution without waiting for the entire application dataset to be transferred to the public cloud, and the data blocks are transferred from the private dataset to the public cloud only when the data blocks are accessed by the application. The application's private data is kept in the public cloud only when it is currently being used. If there are security concerns, the application's private data is transferred over the public network in an encrypted form and stored in the public cloud in an encrypted form.
摘要:
A high-speed cache is shared by a plurality of independently-operating data systems in a multi-system data sharing complex. Each data system has access both to the high-speed cache and to lower-speed, upper-level storage for obtaining and storing data. Management logic in the shared high-speed cache is provided to meet the serialization and data coherency requirements of the data systems when sharing the high speed cache as a store-multiple cache in a multi-system environment.
摘要:
A Structured External Storage (SES) Facility is coupled to one or more Central Electronic Complex (CEC) Processors, each linked to each other in a System Processing Complex (Sysplex). Applications within a CEC communicate with the SES via messages over a communication means. The SES comprises one or more List Structures, each structure comprising one or more Lists, each List comprising a list control and none or more List Entries. List Entries may be keyed or unkeyed, have a List Entry Identifiers (LEID), created within the SES, and unique within a particular List Structure. List Entries may also have a unique (within a List Structure) Name, created and managed by the application. A plurality of List commands are provided for creating, locating, moving and deleting List Entries--with location reference possible by LEID, Name, Key, Location within List, or a combination of Key and Location within Key. Name and LEID are preserved between moves, while Key may or may not be changed, and will be used to determine relative location.
摘要:
The configuration of the logical processors of a logical partition is managed dynamically. A logical partition is initially configured with one or more logical processors. Thereafter, the configuration can be dynamically adjusted. This dynamic adjustment may be in response to workload of the logical partition.