SOLID-STATE RAM DATA STORAGE FOR VIRTUAL MEMORY COMPUTER USING FIXED-SIZED SWAP PAGES WITH SELECTIVE COMPRESSED/UNCOMPRESSED DATA STORE ACCORDING TO EACH DATA SIZE

Patent №

US 5,490,260

Granted

1996-02-06

Filed 1991

Owner

CERAM, INC.

+3 more

Lab

AI components

1

hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

07679530

A computer using virtual memory management employs a random-access type storage device such as a semiconductor memory for page swapping. The semiconductor memory is formatted to provide multiple partitions of varying block size, e.g., two block sizes, for compressed pages, and another block size for uncompressed original-sized pages. The data to be stored is in pages of fixed size, and these pages are compressed for storage if the compressed size fits in the block size of one of the small-block partitions in the memory. If a data page is not compressible to one of the small block sizes, it is stored uncompressed in the other full-size partition. The operating system maintains a table storing the locations of the pages in the partitions, so upon recall the page (if compressed) is retrieved from its location found using the table, decompressed and sent to the CPU. The relative number of blocks in the partitioned memory (e.g., the physical storage capacity of the memory) is set (or dynamically allocated) at the average ratio of compressible pages to uncompressible pages for the compression algorithm used. For example, an algorithm may compress 90% of the pages to either 50% or 70% of their original size, so a ratio of the number of locations in the compressed partitions of the semiconductor memory to the number of locations in the uncompressed partition is selected as 90:10. The compression mechanism operates on bytes in bit-parallel format, and uses a lookahead buffer which is compared with bytes in a window to produce 9-bit symbols. The stream of 9-bit symbols passes through an ECC generator, also operating in bit-parallel.

AI hardwareG11B 20/00007G06F 3/0601G06F 12/023G06F 12/08G06T 9/005H03M 7/3084H03M 7/3086H03M 13/151+1 more

AI classification

AI hardware0.64
Natural language0.00
Vision0.00
Speech0.00
Machine learning0.00
Knowledge representation0.00
Evolutionary computation0.00
Planning0.00

Ownership

CERAM, INC.

assignment · 57560699

GNADINGER, ALFRED P.

assignment · 76240171

MICROTRONIX CORPORATION

assignment · 157470940

DIGITAL DATA FUNDING LLC

assignment · 162560156

Assignors

MILLER, WILLIAM D., HARRINGTON, GARY L., BELLMAN, CHRISTOPHER M.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC