Data recovery algorithms are the basis on which the software and hardware components function in three different modes of operation, namely diagnosis, error detection and data recovery. The fourth aspect of algorithm is meant for rectification of physical and logical errors in the device which can be achieved by the effective combination of hardware tools and software utilities. In this session you will be given an overview of the principles and practices related to some of the most prominently used algorithms in network server data restoration system. It will give you basic ideas about the way in which these algorithms are incorporated within the tools in a logical manner.
Algorithm for diagnosis
Optical flow algorithm: – This is one of the basic types used for diagnosing the disk failure symptoms by drilling down to the bit level of data present in the tracks and sectors of the server disk drive. The bitwise scanning operation is conducted in all the platters of the disk from zero till the last platter. While the recursive search is in progress, an image of the disk is created by the probing software based on the bitwise data collected. This image can be used to determine the probable causes and effects of disk failure.
The flow of algorithm needs a starting set of points which are standardized. Grid-data is one such function which is used to create a liner path which passes through all the sectors and tracks in the failed disk and determines the data points. These data points are compared with the standard values of normal data storage. The resulting line will be straight as long as the existing data points are matching with the standard values. Once an error is found in the data storage pattern (It may start from boot sector, master file table or the data areas), the straight line experiences a gradient. The angle and intensity of gradient will depend on the nature and intensity of the error encountered during the scanning operation. Once the complete disk diagnosis is complete, the graph is submitted to the analyzing software. The application will create a diagrammatic view of the errors and based on that it will determine the nature and intensity of errors in the disk drive. Various values are allocated for logical failures and physical damages. Based on these values, the exact nature of damage is determined and displayed on the user interface screen.
Pattern matching: – This type of algorithm is used for matching the file patterns based on bit comparison method. It can be used to determine the errors in file allocation tables and NTFS partitions, apart from comparing many other types of errors in the HFS, HFS plus and other file system patterns. It will also determine the nature and intensity of bad sectors within the tracks scanned by the software application. Once the variations in patterns with respect to the standard patterns are listed out, this will be fed to the diagnosis part of the software application. The system derives at a report which shows all the errors and displays them on the user interface screen. The technician will be able to generate a diagnosis report based on the listing. This report will be highly useful for repair and restoration planning.
This algorithm is primarily used for hardware probing methods for determining the extent of physical damages sustained by the components and parts of hard drives, SSD, flash cards and memory cards. Mostly this algorithm is built into the OEM software application that accompanies the hardware diagnosis and restoration tool kit. One of the primary types used is the scanning probe function for generating a 3-D topographic-image of the data storage surfaces. The same concept is implemented in its other derivatives named magnetic-Force-Microscopy and tunneling microscopy.
Scanning probe microscopy: – This is one of the methods which are common. This method uses a 3-D topographical imaging of the disk surface at the minuscule levels. With this type of probing the technician will be able to determine most types of surface defects in the disk.
Magnetic-Force-Microscopy: – This is another method of deep probing into the platters of disk drive. By using this method, the tool will be able to detect the complete set of old data written on the tracks and sectors of the disk drive, even after they have been overwritten with new files. When this hardware tool is sued along with OEM software from the product manufacturer, it is possible to recover most of the lost, corrupt and overwritten data from each sector in a bitwise manner. This is said to be one of the specialties of Linux based data recovery solutions.
The layers of data in the core disk areas that contain images of all the previously written data get gradually diminished with time. This is the reason for the data recovery experts to recommend immediate action when data is lost or gets corrupt. The images will be large enough for the hardware and software tools to detect them easily and increase the probability of maximum data recovery with ease.
Scanning-Tunneling-Microscopy: – This method of probing also goes deeper into the tracks and sectors of the disk drive, but it explores the sectors with the help of a probing-tip that is made of nickel metal. The diagnosing method goes into the bit level of tracks and sectors in the disk platters, starting from track zero of platter-zero till the last track of the last platter in the disk.
Data recovery functions
Recursive reading and restoring algorithm used in data recovery methods is implemented in most of the OEM and enterprise software tools. Direct-copy, reverse-copy, bisect-copy are some of the other methods used in data restoration software. Cloning of data bits from the damaged sectors is the basic principle on which these algorithms work. Links between various bits of data is established based on the master file table data structure. After the cloning is complete, the software will club the similar sets of clones to restore the files and folders.