Table of Contents
Advertisement
INTRODUCTION
Although data groups are fixed-length and always contain 22 data frames, the DDS format is designed such
that variable-length computer records can be stored in the fixed-length data groups.
The native transfer rate of 11 million bytes/minute (183 KB/sec) is a characteristic of the DDS format for
computer OAT technology. At this transfer rate, a full 1.3 GB of information can be written in 120 minutes,
and a full 2.0 GB of information can be written in 180 minutes. Another important speed consideration is that
once the information is written, specific files or data sets can be located at up to 200 times the speed at which
they were originally written. The resulting average time to randomly access any file on a 60-meter OAT
cassette is 20 seconds.
DDS-DC Tape Format
A superset of the basic DDS OAT format, DDS-DC drives can write compressed and uncompressed data to
the same cassette.
Thus, because DDS-DC is based on the DDS format, backward compatibility is
maintained.
Introduced by the DDS Manufacturers Group and approved by ANSI and ECMA, DDS-DC is a
record
compression
industry-standard format that provides support for
loss less
compression algorithms based on
substitution---such as those of the Lempel-Ziv family.
This format supports compressed and uncompressed records.
A recorded OAT cassette may contain
compressed records, uncompressed records, filemarks, and setmarks.
Compressed records exist within
recorded objects called
entities.
Entities and uncompressed records are collected into groups.
Many aspects of the DDS-DC format are identical to those of the DDS format:
o
The series of transformations (randomizing, interleaving, generation and inclusion of two Reed-
Solomon error-correcting codes, and etc.) applied to a group before recording
o
The tape layout
o
The third group-based level of Reed-Solomon error-correcting codes (C3)
The only differences between the DDS and DDS-DC formats are in the contents of the groups.
The combination of OAT technology and the DDS-DC format provides a solid core around which computer
OAT drives with exceptional performance and reliability can be designed, such as the Python data
compression prodUCts.
1.5.3 Data Compression---General
Data compression is based on reducing the redundancy that occurs naturally in data streams of text, graphics,
code, and other data. Reducing or eliminating such redundancy prior to recording the information to tape
significantly increases the amount of data that can be recorded on a given amount of tape.
Data compression causes repeated strings of data to be recognized and replaced by symbols or codewords
that encode the strings or point back to the original occurrence of the string. In this way, data compression
uses fewer characters to represent the original data.
1-9
Advertisement
Table of Contents
