Arrangement and method for connecting a processing node in a distribution system

• US 7,818,613 B2
• Filed: 08/03/2004
• Issued: 10/19/2010
• Est. Priority Date: 08/05/2003
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Freescale Semiconductor, Inc. (NXP Semiconductors NV)
• Inventors: Temple, Christopher
• Primary Examiner(s): GUYTON, PHILIP A
• Application Number:
  US10/567,309
• Publication Number:
  US 20060274790A1
• Time in Patent Office: 2,268 Days
• Field of Search: 714/9, 714/47, 370/216
• US Class Current: 714/6.13
• CPC Class Codes: H04L 43/50 Testing arrangements

Abstract

An arrangement and method for interconnecting fail-uncontrolled processor nodes in a dependable distributed system. A node has a bus guardian with input switches which act in combination with a logic element as an input multiplexer under the control of a control unit.

This provides the advantage of transferring the problem of fault containment from the output interface of a potentially faulty processing node to the input interface of fault-free processing nodes. By doing so, problems encountered by spatial proximity faults or functional dependencies within a faulty processing node that may jeopardize fault containment at its output interface are mitigated.

Claims

1. An arrangement for connecting a node in a distributed system containing fail-uncontrolled nodes, the arrangement comprising:

a receiver for receiving signals from another node of the system, and

a node guardian coupled to the receiver to control selectively reception according to a predetermined TDMA schedule of a message thereat so as to reduce reception of uncontrolled transmission from another node of the system;

wherein the node guardian comprises;

a switch for receiving a plurality of input signals,

logic coupled to the switch for combining the plurality of received signals according to a predetermined logic function comprising an OR logic function, and

a controller coupled to the switch for controlling application of the plurality of received signals to the logic, the controller being arranged to control the switch according to the predetermined TDMA schedule.

2. A distributed system comprising the arrangement according to claim 1.

3. The system of claim 2, further comprising at least one node having a bus guardian.

4. The system of claim 3 wherein the system is one of A-B:

A a TTP/C system,

B a FlexRay™

system.

5. A method of operating a node in a fail-uncontrolled distributed system, the method comprising:

providing a receiver receiving signals from another node of the system, and

providing a node guardian coupled to the receiver and controlling selectively according to a predetermined TDMA schedule reception of a message thereat so as to reduce reception of uncontrolled transmission from another node of the system;

wherein the node guardian comprises;

a switch for receiving a plurality of input signals,

logic coupled to the switch for combining the plurality of received signals according to a predetermined logic function comprising an OR logic function, and

a controller coupled to the switch for controlling application of the plurality of received signals to the logic, the controller being arranged to control the switch according to the predetermined TDMA schedule.

6. A method of operating a distributed system comprising the method of operating a node according to claim 5.

7. The method of claim 6, further comprising providing at least one node having a bus guardian.

8. The method of claim 7, comprising:

operating a first node according to claim 5,

operating a second node according to claim 5,

providing a first group of nodes having respective bus guardians, and

providing a second group of nodes having respective bus guardians,

wherein the first group is coupled to the first and second nodes via a first common channel, and the second group is coupled to the first and second nodes via a second common channel,

the first group and the first node forming a first error containment region, and the second group and the second node forming a second error containment region.

9. The method of claim 8, the first group further being coupled to the first and second nodes via a third common channel, and the second group further being coupled to the first and second nodes via a fourth common channel.

10. The method of claim 8 further comprising:

operating a third node according to claim 5, and

operating a fourth node according to claim 5,

wherein the third node is coupled to the first common channel, the fourth node is coupled to the second common channel, and the first, second, third and fourth nodes are cross-coupled,

the third node being in the first error containment region, and the fourth node being in the second error containment region.

11. The method of claim 7 wherein the system is one of A-B:

A a TTP system,

B a FlexRay™

system.

12. A method of operating a distributed system comprising:

operating a first node, a second node, a third node, and a fourth node, each according to a node operating method comprising;

providing a receiver receiving signals from another node of the system, and

providing a node guardian coupled to the receiver and controlling selectively according to a predetermined TDMA schedule reception of a message thereat so as to reduce reception of uncontrolled transmission from another node of the system;

providing a first group of nodes having respective bus guardians;

providing a second group of nodes having respective bus guardians, wherein the first group is coupled to the first and second nodes via a first common channel, and the second group is coupled to the first and second nodes via a second common channel;

wherein the first group and the first node form a first error containment region, and the second group and the second node form a second error containment region;

wherein the third node is coupled to the first common channel, the fourth node is coupled to the second common channel, and the first, second, third and fourth nodes are cross-coupled;

wherein the third node is in the first error containment region, and the fourth node is in the second error containment region.