Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data streaming
Reexamination Certificate
1997-12-31
2001-02-06
Maung, Zarni (Department: 2758)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data streaming
C348S086000, C711S161000
Reexamination Certificate
active
06185621
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to data archiving and restoring and relates most closely to the field of multimedia data servers.
BACKGROUND
The invention herein involves computer systems. A computer system normally includes a central processing unit (CPU), a permanent memory unit, a fast memory unit, input and/or output (I/O) units, storage units, and a user interface. Programs contained in the memory, control the operation of the CPU to operate the computer. Whenever the computer is started, the CPU is initially controlled by a small program contained in the permanent memory. This startup program loads an operating system and any other necessary programs and data from files in the storage units into the fast memory. The CPU is able to alter any data contained in fast memory and control the I/O units to transfer data between the fast memory and peripherals such as the storage units, printers and user interface components. During operation, all or parts of data files are loaded into the fast memory. The contents of the data files are then changed, and the new versions of the data files are saved back into storage.
Memory generally refers to electronic information storage. When discussing memory or storage, both programs and data files are referred to collectively as either “data” or “information”. There are two main type of memory: read only memory (ROM) and random access memory (RAM). RAM tends to be much faster than ROM, but also tends to be highly volatile, meaning that it must be regularly written to in order to prevent data loss. ROM, on the other hand, is permanent, so that any data stored in ROM is available when the computer is first turned on.
There are many different types of ROM. Some kinds of ROM are created with an unchangeable program built in. Programmable ROM (PROM), on the other hand, can be written to, but this usually requires high voltage, and may be difficult to erase (requiring, for example, ultraviolet light). Permanent memory may also be rewrittable memory such as electrically erasable ROM (EEPROM), battery backed SRAM, or flash memory.
Faster, more volatile types of memory are known as random access memory (RAM). Types of RAM include static RAM (SRAM), in which data is stored using circuits with several transistors, and dynamic RAM (DRAM), in which information is stored as charges in capacitors. The capacitors of DRAM must be regularly recharged to prevent data loss. The data stored in fast memory is usually lost when the system is turned off. Random access refers to the fact that any word i.e. 2 or 4 bytes (1 byte equals 8 bits of data), of the memory can be arbitrarily selected and immediately read or written to in one operation at any time without reading through the rest of memory. Note that permanent memory (ROM) is also random memory in this sense.
ROM is normally built into a motherboard of the computer. Fast memory is commonly provided on small boards that are mounted in memory slots on the motherboard of the computer. Flash memory PC cards are available that insert into PCMCIA slots commonly found in laptop computers, and small computer systems often use ROM cartridges especially for computer games.
Storage usually refers to units which contain computer media onto which the data and programs are stored. The computer media may be magnetic media in which a layer of plastic is coated with a layer of metal alloy oxide (e.g. iron rust) which is magnetic but not electrically conducting. Bits of data may be written to such magnetic media by magnetizing the particles at a storage point on the layer in a particular direction by controlling current to a magnetic write head moving closely over the point. Data may be read from magnetic media by detecting the effect of the magnetization on electrical current output from a magnetic read head. Alternately the media may be optical media in which bits may be written by directing a higher power laser beam at a point in a layer of the media to change the reflectivity of the layer at that point and read by detecting the amount of reflected light when a lower power laser is directed at the point on the layer. Some storage devices, such as hard disk drives, include a fixed media which can not be removed. In other devices, such as floppy disk drives, tape drives, and optical disk drives, the media can easily be removed and replaced by other media containing different programs or data files.
Some digital storage devices, such as hard disk drives and some types of optical disk drives, provide high speed access to any arbitrary individual block of data in the media of the device. This capability is referred to as random access. Unlike memory, no storage device allows random access to individual words (e.g. 2 or 4 bytes), but random access storage devices do allow relatively fast access to individual blocks (e.g. 512 bytes).
Other types of digital storage devices, such as digital tape drives, only allow access to blocks of data in a sequence. For example, a tape may have to be wound for hundreds of meters in order to reach a particular block of data. Thus, these types of storage devices are only useful for accessing large sequential sections of data, since random access to individual blocks is either not available or too slow for practical use. Such devices are referred to as sequential access storage. Common sequential access devices have removable media which is relatively inexpensive, allowing the storage of large amounts of data at relatively low cost.
Some storage drives are built-in or internal and require opening the computer to install or remove. Some other storage devices are external and plug into ports of the computer to be easily installed and removed. High performance systems often have hard drives installed in enclosures that slide into internally mounted frames for convenient replacement.
The unreliability of high speed, random access storage devices is a major problem in data storage. Certain mechanical parts of a storage drives, such as positioning motors and bearings, are not nearly as reliable as solid state devices. In addition, because the data on a hard disk is so easily accessible, computer viruses, software failures, and operation errors can easily damage stored data. For this reason, important data is often copied onto removable media which is removed from the system so that no failure of the system can directly damage the data. This process is referred to as backing-up the data. If there is a failure in the computer system, the data on the system storage device can then be restored by copying the backed-up data from the removable media.
Also, in order to increase reliability of hard disk storage, computer systems used for critical, changing information commonly use a system known as a redundant array of inexpensive disks (RAID). In a RAID system, instead of being stored on a single hard disk, each data file is about evenly spread out across several data disks by a RAID controller card. In addition, parity information is written to a parity disk, so that if any single disk drive fails, there will be no loss of data or access to the data. Access to the disks is cycled across the data disks by the RAID disk controller and parts of each disk file is read or written in turn to each data drive. This allows a large number of smaller inexpensive disks to operate as though they were one large disk drive. This process of spreading the data across multiple devices is known as striping. Typically in a RAID system, the disk drives are networked to a disk controller card using a small computer interface (SCSI) peripherals network. Commonly available SCSI type RAID disk controllers access up to 13 data disk drives and one parity disk drive and inexpensive SCSI disk drives holding 23 GB (gigabytes=1 billion bytes) are available thus providing up to 299 GB of highly reliable storage in one hard disk storage system (HDSS).
The cost of high speed, random access, storage devices is one of the major costs of computer systems. Because the cost of sequential access storage (including the removab
Belk Michael E.
Caldwell Andrew
Maung Zarni
Philips Electronics N.A. Corp.
LandOfFree
Direct copying between disk blocks in memory onto a network... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Direct copying between disk blocks in memory onto a network..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Direct copying between disk blocks in memory onto a network... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2609086