Abstract: Various systems and methods for inhibiting a modified block of data from being written to an underlying persistent storage device are disclosed. A cache is maintained by a file system and the cache is implemented to temporarily cache portions of application data for an application. The application data for the application is stored on the underlying persistent storage device. Metadata is accessed if it is detected that a block of data in the cache has been modified. The metadata indicates that the block of data is not needed for the integrity of the application. In response to the metadata, the modified block of data is inhibited from being written to the underlying persistent storage device.

Abstract: Techniques for implementing selective file system caching are disclosed. In one particular embodiment, the techniques may be realized as a method including receiving an I/O request from an application such that the application provides a no-cache indication to a file system to implement a direct. I/O for the I/O request, comparing the I/O request and the no-cache indication with a configurable cache map such that the configurable cache map indicates a selective caching based on an I/O type and a file type, processing the I/O request to render a caching decision based on the comparison, and executing selective caching on the I/O request in accordance with the caching decision.

Abstract: In general, a block of data in a data file is stored in a multi-tier storage system. The block of data includes multiple rows and multiple entry values per row, including values for a particular entry. The values of the particular entry in the data block can be used to determine whether to move the data block to a different tier of a multi-tier storage system. The block of data can then either be relocated in a different tier or kept in the current tier.

Abstract: A computer-implemented method for compressing database objects may include 1) identifying a database object within a relational database, 2) determining that the database object meets a compression criterion that is not met by at least one additional database object within the relational database, 3) identifying a location of the database object within a database file of the relational database, and 4) compressing a portion of the database file at the location of the database object without compressing all of the database file. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A method for implementing selective file system caching. The method includes receiving I/O requests from an application and comparing each of the I/O requests with a configurable cache map, wherein the configurable cache map controls a selective caching based on an I/O type and a file type. Each of the I/O requests are processed to render caching decisions based on the configurable cache map. Selective caching is then implemented on each of the I/O requests in accordance with the caching decisions.

Abstract: A sliding window policy is implemented to manage database objects. At a first time, a first portion of a database object is allocated to a first tier of a multi-tier storage system. The first portion corresponds to a first specified percentage of the size of the database object at the first time and is selected according to information associated with the database object. The remaining portion of the database object at the first time is allocated to one or more other tiers. At a second (later) time, a second portion of the database object is allocated to the first tier. The second portion corresponds to a second specified percentage of the size of the database object at the second time and is also selected according to information associated with the database object. The remaining portion of the database object at the second time is allocated to the other tier(s).

Abstract: The present disclosure presents a method for managing portions of files in multi-tier storage systems. The method may include identifying a file that is managed by an application and stored in a multi-tier storage system. The method may also include determining how the application stores data in the file by identifying data-management information associated with the application. The method may further include using the data-management information to identify a first portion of the file, identifying a data-placement policy, and implementing the data-placement policy by moving the first portion of the file from a first tier of the multi-tier storage system to a second tier of the multi-tier storage system. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Storage systems and methods are presented. In one embodiment, a storage reclamation method comprises performing a potential reclamation identification process, wherein the potential reclamation identification process includes determining if there is the potential for reclaiming storage resources; performing a reclamation process, wherein reclamation is performed on storage resources identified by the potential reclamation identification process as being eligible for reclamation, and wherein the storage resources correspond to free space associated with a data file; and performing a valid free space indication process, wherein a valid free space indication process includes forwarding a valid free space indication recognizable to an application as a valid free space indication when the application attempts to direct a read to the storage spaces reclaimed by the reclamation process.

Abstract: Various methods and systems can be used to provide standard file system access to data that is originally accessed only by a proprietary interface. One such method involves querying an application (e.g., a database management system) for information identifying a location of each of several data objects within several volumes. The application provides a proprietary interface to the data objects stored on the volumes. The method then creates metadata, which associates each of several files within a new file system with the location of a respective one of the data objects within the volumes. The metadata can be created by an in-place conversion module implemented on a computing device. The method then provides the application with access to the files within the new file system, subsequent to creating the metadata.

Abstract: In general, a block of data in a data file is stored in a multi-tier storage system. The block of data includes multiple rows and multiple entry values per row, including values for a particular entry. The values of the particular entry in the data block can be used to determine whether to move the data block to a different tier of a multi-tier storage system. The block of data can then either be relocated in a different tier or kept in the current tier.

Abstract: A method for implementing selective file system caching. The method includes receiving I/O requests from an application and comparing each of the I/O requests with a configurable cache map, wherein the configurable cache map controls a selective caching based on an I/O type and a file type. Each of the I/O requests are processed to render caching decisions based on the configurable cache map. Selective caching is then implemented on each of the I/O requests in accordance with the caching decisions.

Abstract: The processors in a multiprocessor computer system are grouped into nodes. The processors can request a lock, but the lock is granted to only one processor at any given time to provide exclusive processor access to the resource protected by the lock. When a processor releases the lock, the lock is made available to another processor at the same node, even though a processor at a different node may have requested the lock earlier. To maintain fairness, the lock is forced to another node after granting a certain number of consecutive requests at a node or after a certain time period. In one embodiment, a specialized data structure representing a lock request from a processor at a particular node is placed into a queue. A later requesting processor can acquire a preemptive position in the queue by spinning on a data structure already in the queue if the data structure corresponds to the processor's node.