Patent US5574931A successfully refreshed from Google Patents.
US5574931A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Interconnection process and system for the control of messages in an array of processors having a parallel structure
Assignee: Airbus Group SAS
•
Filing: 1994-09-26
•
Granted: 1996-11-12
•
Expires: 2014-09-26
Abstract
Abstract The invention relates to a process for controlling the circulation of messages in a ring network, in which the data are consecutively supplied, which ensures a consistency of the circulating messages. It also relates to an interconnection system for performing this process. This system comprises a plurality of processors (P0, . . . , Pn) connected on a ring network by means of in each case a cell (C0, . . . , Cn) ensuring the transmission and reception of messages circulating in the ring, a reception module (R0, . . . , Rn) able to store the messages intended for the processor and a transmission module (E0, . . . En) able to store the messages to be supplied to the network.
Inventors
- 👤 Laurent Letellier
- 👤 Hassane Essafi
- 👤 Thierry Collette
- 👤 Jean-François Larue
- 👤 Jean-Pierre Guerin
- 👤 Jean Rebillat
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-10 01:45
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/16; G06F15/163; G06F15/173; G06F15/17337
Claims Analysis
Total Claims: 7
Claim 1
Independent Claim
382 words
Process for the control of messages circulating in a ring network of elementary processors (Po, . . . , Pn) organized in accordance with a parallel structure and able to process, per cycle, messages of the same size, each of these processors being associated with a respective interconnection module and connected to the network by the interconnection module, wherein a question of priorities of interconnection modules for transmission and for reception of messages is established by circulation of tokens in the network and one size of messages (TMESS) to be transmitted on the network is specified prior to a first transmission, the process comprising the steps of: each elementary processor transmits a token if it verifies the equations: (NPE modulo TMESS)=0 and (N-NPE)>TMESS, with NPE between 0 and N-1 and in which TMESS is the size of the messages to be transmitted, NPE is an elementary processor number in the network and N is the total number of elementary processors connected to the network; when one of the interconnection modules receives a token, the interconnection module receives and acknowledges a first data item of a message to be received circulating in the network if the message is intended for the elementary processor associated with one of the interconnection modules; determining whether the interconnection module has sufficient memory space for storing the message: using the interconnection module to return the message to the network if the memory space is inadequate for storing the message, using the interconnection module to store the first data item of the message and then all successive data items of the message until the interconnection module detects an end of the message if the memory space is adequate for storing the message; and using the elementary processor to check that the interconnection module has an adequate memory space for completely storing a message to be sent: storing the message to be sent in the interconnection module if there is memory space for the message; using the interconnection module to supply a first data item of the stored message to the network when the token is received and then all successive data items of the stored message if at least one message has been stored; waiting until the interconnection module receives a following token if no message has been stored.
Claim 2
Dependent Claim
382 words
Process for the control of messages circulating in a ring network of elementary processors (Po, . . . , Pn) organized in accordance with a parallel structure and able to process, per cycle, messages of the same size, each of these processors being associated with a respective interconnection module and connected to the network by the interconnection module, wherein a question of priorities of interconnection modules for transmission and for reception of messages is established by circulation of tokens in the network and one size of messages (TMESS) to be transmitted on the network is specified prior to a first transmission, the process comprising the steps of: each elementary processor transmits a token if it verifies the equations: (NPE modulo TMESS)=0 and (N-NPE)>TMESS, with NPE between 0 and N-1 and in which TMESS is the size of the messages to be transmitted, NPE is an elementary processor number in the network and N is the total number of elementary processors connected to the network; when one of the interconnection modules receives a token, the interconnection module receives and acknowledges a first data item of a message to be received circulating in the network if the message is intended for the elementary processor associated with one of the interconnection modules; determining whether the interconnection module has sufficient memory space for storing the message: using the interconnection module to return the message to the network if the memory space is inadequate for storing the message, using the interconnection module to store the first data item of the message and then all successive data items of the message until the interconnection module detects an end of the message if the memory space is adequate for storing the message; and using the elementary processor to check that the interconnection module has an adequate memory space for completely storing a message to be sent: storing the message to be sent in the interconnection module if there is memory space for the message; using the interconnection module to supply a first data item of the stored message to the network when the token is received and then all successive data items of the stored message if at least one message has been stored; waiting until the interconnection module receives a following token if no message has been stored.
Claim 3
Dependent Claim
204 words
Interconnection system implementing the process according to claim 1, the system comprising: a plurality of elementary processors (P0, . . . , Pn) able to perform synchronous processing operations, organized in accordance with a parallel structure and communicating with one another in a synchronous manner; and a plurality of interconnection modules, each interconnection module being associated with one of the elementary processors in order to ensure connection of the elementary processor to the network, each interconnection module comprising: a cell (C0, . . . , Cn) connected to cells of adjacent modules and ensuring, when a token is present, reception of the first data item of the message circulating in the network and intended for the elementary processor with which the cell is associated and transmission of the message which the processor wishes to send to the network; a reception module (R0, . . . , Rn) connected between the cell and the elementary processor and ensuring storage of complete messages received by the cell and intended for the elementary processor; a transmission module (E0, . . . , En) connected between the elementary processor and the cell and ensuring storage of complete messages which the elementary processor wishes to send to the network.
Claim 4
Dependent Claim
41 words
Interconnection system according to claim 2, wherein each transmission module incorporates a memory able to store each of the messages to be transmitted and a logic circuit able to determine whether the memory is empty or not, or full or not.
Claim 5
Dependent Claim
39 words
Interconnection system according to claim 2, wherein each reception module incorporates a memory able to store each of the messages received and a logic circuit able to determine whether the memory is full or not, or empty or not.
Claim 6
Dependent Claim
31 words
Interconnection system according to claim 2, further comprising a central control module (MC) directly connected to each of the cells of the network in order to manage sequencing of the network.
Claim 7
Dependent Claim
39 words
Interconnection system according to claim 2, further comprising a central control module (MC) connected to each of the cells and connected to the network by means of a supplementary cell which initializes tokens at the start of a cycle.
References Cited (Backward Citations)
US5335277A
Cited
EP0186150A2
Cited
US4816993A
Cited
EP0281757A2
Cited
US4930122A
Cited
US5151900A
Cited
FR2680026A1
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-11-12 | 2000-05-13 | 2000-11-12 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-11-12 | 2004-05-13 | 2004-11-12 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-11-12 | 2008-05-13 | 2008-11-12 | $7700.00 | Window Closed / Lapsed |