Electrical computers and digital processing systems: multicomput – Multicomputer data transferring via shared memory
Reexamination Certificate
1998-10-14
2001-05-29
Harrell, Robert B. (Department: 2758)
Electrical computers and digital processing systems: multicomput
Multicomputer data transferring via shared memory
Reexamination Certificate
active
06240446
ABSTRACT:
FIELD OF INVENTION
This invention relates in general to transferring information between a main storage and input/output devices in a data processing system, and in particular to a method for increasing the information transferring capacity of the data processing system including techniques for reducing the number of operations required to perform certain I/O functions and for increasing the amounts and types of information that may be transferred during operations.
BACKGROUND OF THE INVENTION
In a network computing environment, multitudes of commands and requests for retrieval and storage of data are processed every second. To properly address the complexity of routing these commands and requests, a number of different resolutions have been implemented. In some data processing architectures, such as International Business Machines Enterprise System Architecture/390 (Enterprise System Architecture/390 is a registered trademark of International Business Machines Corporation), a channel subsystem is utilized to pass information between the main storage and input/output (I/O) devices. The channel subsystem includes one or more channel paths, each including one or more channels and one or more control units. Recently developed technologies such as the International Business Machines ESCON switch (ESCON is a registered trademark of International Business Machines Corporation), connect the I/O devices to the main memory through the control units using legacy channels to support the data transfer there between.
But as the technology improves, the performance of new system processors will require many more legacy channels than are presently in use to support the resulting increase of information transfer in the data processing systems. Current architectural constraints make the addition of such legacy channels an expensive proposition. A further challenge is to provide the link data rate required to support the data rates of new I/O devices such as DASDs and Tapes. In this case, simply adding more legacy channels does not adequately address the problem. A new architecture is needed that can scale up to the higher link speeds needed for normal transaction processing. Therefore, any new proposed architecture, must include a capacity to accommodate higher bandwidth channel links such as Fibre Channel links while providing better data rates and higher link speeds.
This application is being filed at the same time as related application, U.S. patent application Ser. No. 09/172,696 filed Oct. 14, 1998 by Casper et al., still pending; U.S. patent application Ser. No. 09/172,695 filed Oct. 14, 1998 by Casper et al., still pending; and U.S. patent Ser. No. 09/172,462 filed Oct. 14, 1998 by Casper et al., still pending, all assigned to the assignee of the present invention.
This application incorporates by reference the following patents and publications:
1) Fibre Channel Single Byte-2(FC-SB-2) Architecture (AR-6865-00-POK)
2) Fibre Channel (FC-PH) REV 4.3 ANSI X3.230-199x
3) U.S. Pat. No. 5,526,484 to Casper et. al.
SUMMARY OF THE INVENTION
The present invention provides for a method of transferring information in a network computing system environment. At first multiple packets of data are concurrently received data from an information transfer interface mechanism capable of coupling to a plurality of input-output devices. At the same time data is stored concurrently including multiple packets of data in a main storage element in processing communication to the interface mechanism. The interface mechanism in one embodiment of the present invention comprises a plurality of control units capable of coupling to the plurality of input output devices and channel subsystem having a plurality of channels in processing communication with the control units and the main storage. In this embodiment, at least one control unit is dedicated to each input/output device. The interface mechanism can also comprise a fabric or a plurality of optical fiber and switches. The network environment can also include a storage control element in processing communication with the main storage and channel subsystem and coupled to at least one central processing unit.
REFERENCES:
patent: 5526484 (1996-06-01), Casper et al.
Casper Daniel F.
Elliott Joseph C.
Gonzalez Floyd A.
Harrell Robert B.
International Business Machines - Corporation
Neff Lily
LandOfFree
Multiplexing of multiple data packets for multiple... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multiplexing of multiple data packets for multiple..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiplexing of multiple data packets for multiple... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2564616