Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data framing
Reexamination Certificate
1998-07-29
2001-05-22
Coleman, Eric (Department: 2183)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data framing
C712S200000
Reexamination Certificate
active
06237038
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a method of data communication and a system for carrying out the same. More particularly, the invention relates to a method of high-speed data communication and a high-speed data communication system carrying out the method in conjunction with a local area network (LAN).
FIG. 2
is a view of a program structure for use by a data communication system according to the invention. This system permits personal computers and workstations to perform data communication over high-speed transmission routes such as LAN's. This applicant has already filed U.S. Ser. No. 08/079,872 now U.S. Pat. No. 5,797,041 in connection with the above data communication system.
FIG. 2
highlights a buffer constitution showing how data to be communicated is related to buffers that accommodate it in the data communication system. A transmitter
6
and a receiver
7
are connected to a transmission line
13
through which data is exchanged. The transmitter
6
comprises an application program
60
, a transmitting program
63
and a communication controller
68
. The program
60
is assigned an application buffer
61
that accommodates application data
62
. The transmitting program
63
is assigned a data buffer
64
that temporarily holds part
65
of the application data
62
, and a head buffer
66
that temporarily stores communication protocol header information
67
. The communication controller
68
is assigned a transmitting buffer
69
that temporarily contains frame data
610
made of header information and application data. The application program
60
and transmitting program
63
are placed in a system memory of the transmitter
6
and processed by a system processor thereof. The system memory and system processor of the transmitter
6
are not shown in FIG.
2
. The receiver
7
is composed of an application program
70
, a receiving program
73
and a communication controller
76
. The application program
70
is assigned an application buffer
71
for holding the application data
72
received. The receiving program
73
is assigned a protocol buffer
74
that temporarily stores part
75
of the application data
72
. The communication controller
76
is assigned a receiving buffer
77
that temporarily accommodates frame data
78
received. The application program
70
and receiving program
73
are placed in a system memory of the receiver
7
and processed by a system processor thereof. The system memory and system processor of the receiver
7
are not shown in FIG.
2
. In
FIG. 2
, the transmitter
6
divides the application data
62
in the application buffer
61
into pieces and sends them onto the transmission line
13
via the data buffer
64
and transmitting buffer
69
. The receiver
7
receives the pieces of data via the receiving buffer
77
and protocol buffer
74
into the application buffer
71
where the pieces are reassembled.
One drawback of this inventor's previous invention is the absence of consideration for the amount to be divided of the application data
62
. Also not considered are the locations in which to store data in the buffers involved. If the data starting address boundaries fail to coincide with each other upon data copy between two buffers, it takes time to do the copying therebetween. The result is a drop in performance.
In
FIG. 2
, the system processor of the transmitter
6
and that of the receiver
7
are a 32-bit processor each, the processors handling data in increments of four bytes. The starting address in each buffer is located so that the address will be an integer multiple of the minimum byte length. That location is called a word boundary.
FIG. 3
shows how data is typically moved between buffers. The buffers in
FIG. 3
each have the same constitution as that depicted in FIG.
2
and identified by the same reference numerals. Each division in a buffer represents a one-byte address, the addresses getting higher from left to right. Each of the inverted triangles points to a word boundary. The location one byte higher than a given word boundary is called a one-byte boundary; two bytes higher, a two-byte boundary; three bytes higher, a three-byte boundary. In general, the 32-bit processor can read data four bytes at a time from word boundaries in the system memory but cannot gain access to that memory across any word boundary. This means that upon data copy, data may be moved from a copy source to a copy destination simply in increments of four bytes if the word boundaries coincide therebetween; if the word boundaries do not match between the source and the destination, data needs to be moved from one to the other in increments or one or two bytes. In the latter case, the increased number of times the memory is accessed means more time required for the copy operation.
FIG. 3
shows an example in which 10 bytes of application data
621
are transmitted first, followed by another 10 bytes of application data
622
. The header information is composed of five bytes. Because the starting address of the application data
622
is stored on a two-byte boundary, data is copied from the application buffer
61
to the data buffer
64
in five increments of two bytes each. The header information is copied from the data buffer
64
to the transmitting buffer
69
in two divisions, i.e., four bytes and one byte. Meanwhile, the data is copied from the data buffer
64
to the transmitting buffer
69
in 12 increments of one byte each. The frame data received from the starting address of the receiving buffer
77
via the transmission line
13
is copied in three increments of four bytes each and in three increments of one byte each to the protocol buffer
74
. Between the application buffer
71
and the protocol buffer
74
, starting address boundaries do not coincide with each other. This requires copying the data from the protocol buffer
74
to the application buffer
71
in 10 increments of one byte each.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a method of high-speed data communication and a system for carrying out the same in copying data between buffers at high speed.
In carrying out the invention and according to one aspect thereof, there is provided a method of data communication comprising the steps of having a processor process communication protocols by a program, using a local memory through which to transmit and receive communication data, having a system memory store the program and the communication data, and adjusting increments in which to transmit application data for data communication, wherein the processor is arranged to handle data in increments of a word width, and wherein the increments are set for an integer multiple of the word width in transmitting and receiving the communication data.
When embodied as outlined, the invention allows data to be transmitted and received in increments of the word width. This boosts the rate of data copy between buffers and thereby provides high-speed data communication.
These and other objects, features and advantages of the invention will become more apparent upon a reading of the following description and appended drawings.
REFERENCES:
patent: 3942155 (1976-03-01), Lawlor
patent: 4543654 (1985-09-01), Jones
patent: 4975915 (1990-12-01), Sako
patent: 4987572 (1991-01-01), Scott
patent: 4992931 (1991-02-01), Hirasawa
patent: 5168561 (1992-12-01), Vo
patent: 5191653 (1993-03-01), Banks et al.
patent: 5293379 (1994-03-01), Carr
patent: 5307351 (1994-04-01), Webster
patent: 5317719 (1994-05-01), Rotmovits
patent: 5392406 (1995-02-01), Petersen et al.
patent: 5408613 (1995-04-01), Okabayashi
patent: 5416907 (1995-05-01), Polzin et al.
patent: 5511166 (1996-04-01), Travaglio et al.
patent: 5655140 (1997-08-01), Haddock
U.S. application No. 08/079,872, Title Communication Control System and Method, copy of specification, claims and drawings.
Dalton, Chris, et al. “Afterburner—A Network-Independent Card Provides Architectural Support for High-Performance Protocols,” IEEE Network, Jul. 1993 pp.
Higuchi Hidemitsu
Horimoto Toru
Miyamoto Takahisa
Shimamoto Yukio
Yasue Toshikazu
Coleman Eric
Hitachi , Ltd.
Mattingly Stanger & Malur, P.C.
LandOfFree
Method of data communication and system for carrying out the... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of data communication and system for carrying out the..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of data communication and system for carrying out the... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2522897