Home | Links | Contact Us | More About Intellectual Property | Bookmark
Search patents:
Home I/O Systems Dual-bus-system-with-multiple-processors-having-data-coherency-maintenance

 Low inductance circuit apparatus with controlled impedance cross-unders and connector for connecting to backpanels
The present invention is directed to a rigid/flex printed wiring board comprising an integral ...


 Cover arrangement for electrolytic capacitors
The combinations of the features recited by the Patent claims 1 and 3 are provided in order to ...


 Ejection mechanism
In its first preferred embodiment, the present invention comprises a housing for up to two cards ...


 Computer system having a structure for easy assembling/disassembling of peripheral equipment
Accordingly, it is an object of the present invention to provide an improved portable computer. It ...


 Media eject mechanism
It has been discovered that a media eject mechanism which includes an L shaped eject member having ...


 Media module locking and ejecting mechanism
In accordance with the present invention, an apparatus for locking and ejecting a media module in a ...


 Device for mounting fan in a portable computer
Accordingly, it is an object of the present invention to provide an improved device for mounting a ...


 Personal digital assistant and associated computer host device bay structure
In carrying out principles of the present invention, in accordance with a preferred embodiment ...


 Portable computer having latching hooks
In one embodiment, as shown in FIG. 1, computer system 10 includes a microprocessor 12, which is ...


 Data set network diagnostic system
FIG. 1 depicts a communication system, or network, in which the present invention is used. In ...


 Dual bus system with multiple processors having data coherency maintenance

Details
Inventors: Tran, Dan Trong; Ricci, Paul Bernard; Sheth, Jayesh Vrajlal; White, Theodore Curt; Cowgill, Richard Allen;
Assignee: Unisys Corporation (Blue Bell, PA)
Primary Examiner: Swann; Tod R.
Assistant Examiner: Tran; Denise
Attorney, Agent or Firm: Kozak; Alfred W., Starr; Mark T., Petersen; Steven R.

A multi-module network having dual system busses using a common bus protocol which supports processor modules using a store-through cache memory and processor modules using a non-store-through cache memory which includes interface modules including a spy unit for maintaining cache coherency and arbiter modules so that any particular module is not starved out. A maintenance processor organizes the network as a joined system where both the store-through and non-store through processor units can utilize either one of the dual system busses or a split system where one bus is dedicated to the store-through processor units and one bus is dedicated to the non-store through processor units.

DETAILED DESCRIPTION The present invention provides a specialized dual system bus capability by which different types of processor modules having different types of cache functions can be connected and integrated for efficient data transfer operations and computer processing.
Each one of the specialized dual system busses provides separate data lines for each of the functions of control, status, arbitration, clock signals, and address/command/data information transfer.
Multiple choice of bus pathways are provided for each module when it becomes a module requesting bus access.
Additionally multiple pathways are provided via the dual system busses for receipt of data when a module becomes a receiving module.
Additionally, there can be simultaneous operations wherein a single module can be both transmitting information on one system bus and receiving information on the other bus line, thus enhancing throughput.
A primary benefit of the dual bus system is the capability of operational activity for processors having Store-Through cache memory units and at the same time having the capability of using the same data transfer protocol for operating with processing units having Non-Store-Through (NST) cache capability.
The data transfer commands and protocol described herein enable the same bus to be used for one configuration involving Store-Through cache memory units in a processor or alternatively being available for operating with processing units having Non-Store-Through capability without any change in the bus structure or the bus protocol of the system.



Related patents
  Heat sink
The object of the present invention is to solve the above-described drawbacks presented in conventional heat sink. It is another object of the present invention to ...
  Missing clock detection circuit
Missing clock pulses are detected in accordance with the present invention by a detection circuit for reading encoded data including a timer for measuring a first time ...
  Method and apparatus for qualifying data
OF PREFERRED EMBODIMENTS Referring to FIGS. 1-4, the hardware components of an embodiment of the present invention are illustrated and the following discussion is ...
  Self-synchronizing optical imaging system
A self-synchronizing optical radiation imaging device is provided having optics for receiving optical radiation and an optical image delay system for delaying at least a ...
  Pulsed light signal receiver
I claim: 1. A receiver for pulsed light signals comprising: a pulsed light detecting means including a phototransistor having base and collector and emitter electrodes, ...
  Convolution filter arrangement for digital multifrequency receiver
In the illustrative digital switching system of FIG. 1, a digital switching network 10 operates under control of a program controlled facility 11 and provides for the ...
  Delay time checking arrangement
Accordingly, it is an object of the present invention, to provide a delay time checking arrangement for checking the delay time in a signal propagated between first and ...
  High-speed multi-port FIFO buffer circuit
OF THE INVENTION FIG. 1 shows a data packet switch which consists of n input lines 5a-n, each of which is connected to a separate router 10a-n. All of the routers 10a-n ...
  Reducing memory access in a multi-cache multiprocessing environment with each cache mapped into different areas of main memory to avoid contention
This invention discloses a method for screening the memory accesses and thereby freeing the cache to service their processor in multiple caches and a circuit which ...
  Flex circuit connector assembly and method for manufacturing the same
The present invention recognizes the need for an improved flex circuit connector assembly that can be used to interconnect a flex circuit to a rigid circuit board or to ...

0.454

Archive: All patents - Links

Copyright (c)2006 Eipa-patents.org - All rights reserved