Electrical computers and digital processing systems: memory – Addressing combined with specific memory configuration or... – Dynamic-type storage device
Reexamination Certificate
2000-02-29
2002-10-15
Yoo, Do Hyun (Department: 2187)
Electrical computers and digital processing systems: memory
Addressing combined with specific memory configuration or...
Dynamic-type storage device
C711S170000, C711S202000, C711S209000, C711S112000, C711S220000, C711S221000
Reexamination Certificate
active
06467014
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to large capacity data storage devices. More specifically, the present invention relates to a system and method for automatic mapping and address translation for data storage systems having multi-surface, multi-zone storage devices.
Today's computers have many different data storage needs. These needs can be satisfied by a large number of storage devices such as disk storage systems, hard drivers, tape storage systems, etc. These systems communicate with the host computer and manage the actual data storage task. Typically, data is transferred to and from the storage devices as a contiguous stream of logical blocks, wherein each block has a unique logical block address (LBA). This information, identified by the logical block address, must then be stored at some physical location in the data storage device.
In the case of a disk storage device, the storage area is arranged in surface-track-sector addresses (STSA). Each LBA referred to by the host computer must be mapped to a unique STSA on the storage device in order to accommodate easy identification and retrieval. As a further complication to this mapping requirement, the storage devices will often employ multiple surfaces, and multiple zones on the storage media. Efficient use of these multiple surfaces and zones must be considered when devising the mapping and subsequent translations methodology.
Once a data storage and retrieval methodology is established, it is extremely difficult to make adjustments. If a new disk configuration is introduced, the established mapping methodology must be reworked. As would be expected a new configuration requires the development of new address mapping and translation formulations, including further testing and validation.
When using storage devices with multiple surfaces, it is often beneficial to read data from or write data to both sides of the disk simultaneously. This helps to increase the overall data rate of the storage device. However, this complicates the mapping and data storage methodologies. Specifically, the data storage methodology is much more complicated because two read/write heads must be continually utilized, and the addressing must be maintained for both those surfaces.
As mentioned above, certain data storage devices have multiple zones on a surface thereof. Each zone is typically an area on the surface wherein each track within the zone has similar characteristics. More specifically, a zone is typically identified as a group of tracks which all have the same number of sectors on each track. For example, a disk may be broken into four different zones wherein the tracks in the first zone have two sectors per track, the second zone have four sectors per track, the third zone have six sectors per track, and the last zone have eight sectors per track. The number of zones contained on a disk, and the number of sectors per track within that zone will depend upon the physical set up and configuration of the disk, as dictated by the manufacturer.
In order to solve the mapping problems, prior systems have utilized a “hardcoded” algorithm. That is, the data was mapped to the disk using a predetermined set of instructions for the particular disk configuration. While this solution was efficient, it was not capable of providing for varying disk configurations. For example, if the number of sectors per track in a given zone changed, or the number of tracks changed, the code had to be rewritten. This is an undesired activity as it is extremely costly and time consuming.
SUMMARY OF THE INVENTION
In order to maintain effective data transfer rates using a data storage device with multiple surfaces and multiple zones, and to automatically react to changes in the disk configuration, the present invention provides an automated mapping technique and system for a general STSA configuration. In order to map from an LBA to an STSA, a configuration table (CT) is used which specifies the physical configuration of the storage device. This configuration table is utilized to automatically generate address translation formulations. Consequently, a modification of the disk configuration for the particular storage device, will result in a modified configuration table. This will then automatically change the address translation formulation.
An address translation structure, is used to accommodate efficient address translation formulations. This address translation structure utilizes a Physical Block Address (PBA) as an intermediate tool in the translation. Utilizing the above mentioned configuration table, the physical block address is correlated to each available sector on the disk. This PBA helps to simplify the address translation formulations.
As a starting point, information about the disk layout is input by the user. This includes providing range definitions for the disk surfaces which identifies those areas available for user data storages, and those areas designated for other purposes. Collection of all range definitions is assembled in the configuration table which identifies the layout of the disk. Specifically, the configuration table will identify which tracks on the disk are available for user data storage and which tracks are not. Also, the user will provide the number of sectors per track in the particular zones, and any desired pairing information. Based on this information, an address translation structure is developed. This address translation structure, along with appropriate translation formulas, will then allow fast and efficient address mapping.
As the host computer provides directions regarding data storage and retrieval, the address translation structure provides a useful mechanism for accomplishing the necessary mapping. When information is provided by host computer and identified by a logical block address (LBA), this information will be assigned and stored at one or more sectors, each sector identified by a physical block address. The LBA and its associated physical block addresses are easily calculated using the address translation structure and associated formulas. These formulations allow for easy retrieval and identification of stored information.
As can be easily recognized, this methodology allows the disk configuration to be easily modified, without detrimentally affecting the remainder of the addressing scheme. More specifically, a change in the disk configuration, will show up as a change to the configuration table. Since all other mapping steps are dependent upon the configuration table, changing its contents will automatically change the steps and information used by the remainder of the addressing scheme. Consequently, these additional changes are automatically passed down.
It is an object of the present invention to provide an address mapping and translation system which can easily and quickly react to changes in the disk configuration. Consequently, various disk configurations can easily be changed without requiring large amounts of reprogramming.
It is a further object of the present invention to provide an address mapping and translation system which utilizes an easily modifiable configuration table. The configuration table itself defines the various ranges and configurations for the disk itself.
It is a further object of the present invention to provide a data mapping system which easily converts a logical block address, provided by a host computer, to a surface-track-sector address. This surface-track-sector address identifying a specific data sector on the data storage media.
REFERENCES:
patent: 5978336 (1999-11-01), Mine et al.
patent: 6084731 (2000-07-01), Uchida et al.
patent: 6181497 (2001-01-01), Malone, Sr.
Namazi Mehdi
Oppenheimer Wolff & Donnelly LLP
Plasmon LMS Inc.
Yoo Do Hyun
LandOfFree
Automatic mapping and efficient address translation for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic mapping and efficient address translation for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic mapping and efficient address translation for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2939099