Abstract: A computer system includes a main memory, at least one processor, and at least one level of cache. The system contains at least one segment table having multiple segment entries recording the assignment of segments in an address space. At least some segment table entries include pre-fetch data indicating which portions of a segment should be pre-fetched. Preferably, a segment table entry contains a list of pages for pre-fetching. Preferably, pre-fetching a listed page causes address translation data for the page to be cached in at least one address translation caching structure. Pre-fetching may also cause data contents of the page to be loaded into a cache.

Abstract: In one embodiment of the present invention, a method includes observing disk requests for a drive associated with a memory device; and mapping the memory device based on observing the disk requests.

Abstract: A method is described comprising: scheduling jobs for a program build to execute in parallel across a plurality of nodes; predicting the files required to complete each of the jobs; and preloading the files to each node prior to the execution of each job.

Abstract: In one embodiment, a multilevel segmented memory device may be used to store persistent data in a first memory level and dynamic data in a second memory level. In the first level, data fragments may grow in an ascending order, and sequence tables may grow in a descending order. In the second level, object pointers may grow in a descending order, and data units grow in an ascending order.

Abstract: A method and apparatus for coordinating dynamic memory page deallocation with a redundant bit line steering mechanism are provided. With the method and apparatus, memory scrubbing and redundant bit line steering operations are performed in parallel with handling of notifications of runtime correctable errors. When a correctable error is encountered during runtime, and the correctable error is determined to be persistent, then dynamic memory page deallocation is requested of a hypervisor. The determination of persistence is based on a history CE table that is populated by the operation of the memory scrubbing and redundant bit line steering mechanism of a service processor. Thus, only those correctable errors that persist for longer than one memory scrubbing cycle are subject to memory page deallocation.

Abstract: A method includes a configuration definition creation step of writing configuration information on a primary site into a storage subsystem; a data transfer step of copying the configuration information, which is written into a storage device, to a storage subsystem in a secondary site over a network; a data reception step of receiving the transferred configuration information and storing it in the storage subsystem in the secondary site; and a configuration definition step of reading the stored configuration information and settings up a server in the secondary site.

Abstract: A journal write unit writes journal data into a third storage device. The journal data includes an identifier of a logical volume in a first storage device into which data has been written, information of a location in which the data is stored in the logical volume, update time which is current time acquired from a timing mechanism, and the data. A second write unit refers to update time of the journal data stored in the third storage device, selects journal data for which a difference between current time acquired from the timing mechanism and the update time is longer than a detection time stored in the third storage device, and writes the data into a place indicated by the location information, in a logical volume in the second storage device in the order of update time in the selected journal data.

Abstract: Duplicate data obtained by copying and a volume containing it are flexibly managed according to the attribute of the file to be copied by a storage apparatus as is required by a user. When receiving a backup instruction for the file to be backed up, an instruction for a method managing the attribute of the object to be backed up and the duplicate data obtained by the backup of the object to be backed up is also received. In this case, the duplicate data and the volume name to store it are related, thereby managing the specified attribute.