Topic
Defragmentation
About: Defragmentation is a research topic. Over the lifetime, 639 publications have been published within this topic receiving 10503 citations. The topic is also known as: Disk defragmentation.
Papers published on a yearly basis
Papers
Patent•
IBM1
TL;DR: In this paper, a data storage subsystem employs managed files, each comprising one or an aggregation of multiple constituent user files, to reduce file management overhead costs, and a reclamation is triggered when the amount of deleted-file space in a prescribed storage area satisfies certain criteria, e.g. poor data storage efficiency.
Abstract: A data storage subsystem employs managed files, each comprising one or an aggregation of multiple constituent user files, to reduce file management overhead costs. After receiving user files from a client station, the subsystem creates a contiguous managed file by aggregating selected ones of the received user files according to certain predetermined criteria. Managed file creation and use are transparent to the client stations. Eventually, unused or "deleted-file space" arises as individual user files are deleted from their respective managed files. "Reclamation" is triggered when the amount of deleted-file space in a prescribed storage area (e.g. device, volume, managed file, etc.) satisfies certain criteria, e.g. poor data storage efficiency. Reclamation is performed to regain wasted space between managed files, and also space that occurs within managed files due to deletion of some but not all constituent user files. Reclamation is applied to the prescribed storage area one managed file at a time. Each managed file is reviewed to determine whether it contains any deleted-file space. Managed files without any deleted-file space are simply copied intact to the target storage area. If a managed file contains deleted-file space, however, a "reconstruction" process is performed for that file. Reconstruction identifies any contiguous regions of user files within the managed file, then copies the contiguous regions to adjacent locations in a target storage area. Reclamation is complete after all managed files in the prescribed data storage area have been copied or reconstructed.
434 citations
TL;DR: The impacts of big data applications on underlying network infrastructure and the concept of flexible-grid elastic optical inter-DC networks are described and the data migration in such networks as dynamic anycast is model and several efficient algorithms are proposed.
Abstract: This article discusses the technologies for realizing highly efficient data migration and backup for big data applications in elastic optical inter-data-center (inter-DC) networks. We first describe the impacts of big data applications on underlying network infrastructure and introduce the concept of flexible-grid elastic optical inter-DC networks. Then we model the data migration in such networks as dynamic anycast and propose several efficient algorithms. Joint resource defragmentation is also discussed to further improve network performance. For efficient data backup, we leverage a mutual backup model and investigate how to avoid the prolonged negative impacts on DCs’ normal operation by minimizing the DC backup window.
417 citations
Patent•
IBM1
TL;DR: The critical file management system of the present invention includes a critical file manager, which can be accessed by programs through an application programming interface (API), and a user interface is also provided as discussed by the authors.
Abstract: A system and method for managing critical files in an information handling system. Critical files are those files which are difficult to recover after a system failure, hard disk reformat, or user error. The critical file management system of the present invention includes a critical file manager, which can be accessed by programs through an application programming interface (API). Programs call the API to register and unregister critical files. A user interface is also provided. The user interface allows users to register and unregister critical files, and also allows users to view a list of all critical files and versions, along with the name of the program or user which added each critical file to the list. Users may also request a backup or restore of one, several, or all critical files at any time. A database of critical files, including a version for each file, is maintained. Backup copies of each critical file are stored in a critical file storage. At predetermined times, such as system shutdown or boot, the critical files are backed up. The critical file management system may also automatically restore all critical files at predetermined times, such as every time the information handling system is started.
359 citations
IBM1
TL;DR: A mostly non-moving, dynamically defragmenting collector is presented that is able to obtain mutator utilization rates of 45% with only 1.6--2.5 times the actual space required by the application, a factor of 4 improvement in utilization over the best previously published results.
Abstract: Now that the use of garbage collection in languages like Java is becoming widely accepted due to the safety and software engineering benefits it provides, there is significant interest in applying garbage collection to hard real-time systems. Past approaches have generally suffered from one of two major flaws: either they were not provably real-time, or they imposed large space overheads to meet the real-time bounds. We present a mostly non-moving, dynamically defragmenting collector that overcomes both of these limitations: by avoiding copying in most cases, space requirements are kept low; and by fully incrementalizing the collector we are able to meet real-time bounds. We implemented our algorithm in the Jikes RVM and show that at real-time resolution we are able to obtain mutator utilization rates of 45% with only 1.6--2.5 times the actual space required by the application, a factor of 4 improvement in utilization over the best previously published results. Defragmentation causes no more than 4% of the traced data to be copied.
348 citations
Patent•
27 Apr 1998TL;DR: In this paper, a method and apparatus for reclaiming space on a flash memory device is provided in which valid data is copied from an individually erasable sector on the flash device to a designated memory location.
Abstract: A method and apparatus for reclaiming space on a flash memory device is provided in which valid data is copied from an individually erasable sector on the flash device to a designated memory location. The sector is then formatted or erased and the valid data is returned to the flash device at the next available free space contiguously. A pointer is updated to keep track of the location of the next available recently formatted free space on the flash device. The process is then repeated on each consecutive sector on the flash memory device until all valid data becomes contiguous on one contiguous segment of the device and all free space becomes contiguous on a second contiguous segment of the device. In this manner, the flash memory is defragmented and the largest possible contiguous file can be written to the device. Files can be saved on the flash memory device in contiguous file formats and blocked file formats. During the reclamation and defragmentation routine, the file format of the valid data is preserved. Also, file blocks on a sector which are originally non-contiguous become contiguous after the routine is performed.
210 citations
Related Topics (5)
Performance Metrics
| Year | Papers |
|---|---|
| 2021 | 20 |
| 2020 | 31 |
| 2019 | 32 |
| 2018 | 39 |
| 2017 | 51 |
| 2016 | 46 |