System and method for fail-over data transport

A method for fail-over connection between a first processing unit and a second processing unit within a computer system, the method comprising the acts of:

• queuing a plurality of transactions, the plurality of transactions for communication between the first processing unit and the second processing unit across a first communications link;
• detecting a failure condition interrupting communication of the plurality of transactions across the first communications link;
• establishing communication, after detection of the failure condition, between the first and second processing units across a second communications link connecting the first and second processing units, the second communications link for communicating the plurality of queued transactions;
• and
• if the act of establishing communication fails to occur within a given timeout period, purging the plurality of queued transactions.

A method for fail-over connection between a first processing unit and a second processing unit within a computer system, the method comprising the acts of:

• queuing a plurality of transactions for communication between the first processing unit and the second processing unit across a first communications link;
• detecting a failure condition interrupting communication of the plurality of transactions across the first communications link;
• selecting, upon the failure condition detection, a second communications link between the first and second processing units;
• establishing communication between the first and second processing units across the second communications link to communicate the plurality of queued transactions;
• and
• if the act of selecting the second communications link fails to occur within a given selection timeout period, then purging the plurality of queued transactions.

A computer system comprising:

• a first input/output processing unit;
• a second input/output processing unit, the first and second processing units communicating across a first communications link therebetween;
• and
• fail-over logic to maintain communications between the first and second input/output processing units after detecting a failure condition across the first communication link, the fail-over logic comprising:
• a data transaction queue to queue, upon detection of the failure condition, a plurality of transactions for communication between the first and second input/output processing units,
• the fail-over logic configured to:
• establish communication between the first and second input/output processing units across a secondary communications link, and
• submit the plurality of queued transactions for forwarding across the secondary communications link.

A computer system comprising:

• a first processing unit;
• a second processing unit, the first and second processing units communicating across a first communications link established therebetween;
• fail-over logic to maintain communications between the first and second input/output processing units after detecting a failure condition across the first communication link, the fail-over logic configured to establish communication between the first and second input/output processing units across a secondary communications link;
• and
• a data transaction queue to store a plurality of data transactions for communication between the first and second processing units, wherein, upon the failure condition detection, the plurality of data transactions are stored within the data transaction queue until the fail-over logic establishes communication across the secondary communications link;
• wherein, if the fail-over logic does not establish communication across the secondary communications link before expiration of a timeout period, the data transaction queue is purged.