Reliable datagram via independent source destination resources

• US 20080177890A1
• Filed: 01/07/2008
• Published: 07/24/2008
• Est. Priority Date: 05/24/1999
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Fred B. Worley, Michael R. Krause, Shankar G. Iyer
• Inventors: Worley, Fred B., Iyer, Shankar G., Krause, Michael R.
• Granted Patent: US 7,904,576 B2
• Time in Patent Office: Days
• US Class Current: 709/230
• CPC Class Codes: H04L 1/1635 Cumulative acknowledgement, i.e. the acknowledgement message applying to all previous messages
• CPC Class Codes: H04L 1/1829 Arrangements specially adapted for the receiver end

Abstract

A method includes producing units of work with at least one source application instance (AI) at a source device and consuming units of work with at least one destination AI at a destination device. A first reliable datagram service is established, with a first source and destination resource (SDR), between the source device and the destination device. A second reliable datagram service is established, with a second SDR independent of the first SDR, between the source device and the destination device. A first unit of work stream is transmitted over a communication services/fabric with the first reliable datagram service, which guarantees strong ordering of the first unit of work stream received at the destination device. A second unit of work stream is transmitted over the communication services/fabric with the second reliable datagram service, which guarantees strong ordering of the second unit of work stream received at the destination device.

Claims

1. A data processing system comprising:

a source device having at least one source application instance (AI) configured to produce units of work;

a destination device having at least one destination AI configured to consume units of work;

communication services/fabric providing communication between the source device and the destination device;

a first source and destination resource (SDR) including first source SDR resources in the source device and first destination SDR resources in the destination device and configured to implement a first reliable datagram service between the source device and the destination device, transmit a first unit of work stream over the communication services/fabric, and guarantee strong ordering of the first unit of work stream received at the destination device; and

a second SDR independent from the first SDR and including second source SDR resources in the source device and second destination SDR resources in the destination device and configured to implement a second reliable datagram service between the source device and the destination device, transmit a second unit of work stream over the communication services/fabric, and guarantee strong ordering of the second unit of work stream received at the destination device.

2. The data processing system of claim 1 wherein the first SDR is assigned a first priority level for effecting throughput and response time of units of work transmitted by the first SDR and the second SDR is assigned a second priority level for effecting throughput and response time of units of work transmitted by the second SDR

3. The data processing system of claim 1 further comprising:

at least a third SDR independent from the first and second SDRs and configured to implement at least a third reliable datagram service between the source device and the destination device, wherein the SDRs are grouped into multiple SDR groups, wherein each of the multiple SDR groups includes at least one SDR and is assigned a unique priority level for effecting throughput and response time of units of work transmitted by the at least one SDR.

4. The data processing system of claim 3 wherein the SDRs within at least one of the multiple SDR groups having multiple SDRs are serviced based on simple scheduling policies.

5. The data processing system of claim 4 wherein the simple scheduling policies comprise at least one of round-robin scheduling policies and weighted round-robin scheduling policies.

6. The data processing system of claim 1 further comprising:

SDR scheduling heuristics configured to adjust a first scheduling rate of the first SDR and a second scheduling rate of the second SDR.

7. The data processing system of claim 6 wherein the first and second scheduling rates respectively create different quality of service (QoS) for AIs respectively bound to the first and second SDRs to transparently, from the perspective of the AIs, adjust QoS a respective AI receives relative to at least one other AI.

8. The data processing system of claim 1 wherein the first unit of work stream and the second unit of work stream are produced by one source AI and consumed by one destination AI.

9. The data processing system of claim 1 wherein the source device is configured to also function as a destination device and the destination device is configured to also function as a source device.

10. The data processing system of claim 1 wherein the first and second destination SDR resources are each configured to provide a negative acknowledgement (NAK) for a unit of work received ahead of its defined order.

11. The data processing system of claim 10 wherein the first and second source SDR resources are each configured to respond to the NAK by performing one of retransmitting all unacknowledged units of work or selectively retransmitting unacknowledged units of work.

12. The data processing system of claim 1 wherein the first and second destination SDR resources are each configured to drop a unit of work received ahead of its defined order.

13. The data processing system of claim 1 wherein the first and second destination SDR resources are each configured to provide at least one of a positive acknowledgement (ACK) for each unit of work which is successfully received and processed by the destination SDR resources and a cumulative positive ACK for a set of units of work that indicate that all units of work in the set of units of work up to and including a current unit of work have been successfully received and processed by the destination SDR resources.

14. The data processing system of claim 1 wherein:

each unit of work in the first and second serial unit of work streams transmitted respectively from the first and second source SDR resources includes a protocol header containing fields employed by the communication services/fabric to target the at least one destination AI;

the protocol header includes a SDR sequence number field indicating the defined order of its corresponding unit of work; and

the first and second destination SDR resources include an expected next sequence number value indicating an expected defined order corresponding to the next unit of work to be received.

15. The data processing system of claim 14 wherein the SDR sequence number field value being less than the expected next sequence number value stored in the destination SDR resources indicates that the unit of work is a duplicate unit of work.

16. The data processing system of claim 14 wherein the first and second destination SDR resources are each configured to, if the SDR sequence number field value matches the expected next sequence number value, verify other protocol header fields and provide a positive acknowledgement (ACK) if the current unit of work is valid from the destination device'"'"'s perspective and provide a NAK if the current unit of work is invalid from the destination device'"'"'s perspective.

17. The data processing system of claim 14 wherein the SDR sequence number field value being greater than the expected next sequence number value indicates that the unit of work is received ahead of its defined order.

18. The data processing system of claim 17 wherein the first and second destination SDR resources are each configured to provide a NAK in response to the indication that the unit of work is received ahead of its defined order contains the expected next sequence number value in the SDR sequence number field of the protocol header of the NAK to indicate to the respective source SDR resources that an intermediate unit of work corresponding to the expected next sequence number value is missing.

19. The data processing system of claim 18 wherein the first and second source SDR resources are each configured to respond to the NAK and retransmit all units of work having an assigned SDR sequence number value equal to or greater than the SDR sequence number value corresponding to the missing intermediate unit of work.

20. The data processing system of claim 17 wherein the first and second destination SDR resources are each configured to verify other protocol header fields in response to the indication that the unit of work is received ahead of its defined order and temporarily store the unit of work if the other verification checks pass.

21. The data processing system of claim 20 wherein the data processing system further comprises:

at least one middleware AI configured to perform a resynchronization operation to recover a missing intermediate unit of work corresponding to the expected next sequence number value if the other verification checks pass.

22. A method of processing data comprising:

producing units of work with at least one source application instance (AI) at a source device;

consuming units of work with at least one destination AI at a destination device;

establishing a first reliable datagram service, with a first source and destination resource (SDR), between the source device and the destination device;

establishing a second reliable datagram service, with a second SDR independent of the first SDR, between the source device and the destination device;

transmitting a first unit of work stream over a communication services/fabric with the first reliable datagram service;

guaranteeing strong ordering of the first unit of work stream received at the destination device with the first reliable datagram service;

transmitting a second unit of work stream over the communication services/fabric with the second reliable datagram service; and

guaranteeing strong ordering of the second unit of work stream received at the destination device with the second reliable datagram service.

23. The method of claim 22 further comprising:

assigning a first priority level to the first SDR for effecting throughput and response time of units of work transmitted by the first SDR; and

assigning a second priority level to the second SDR for effecting throughput and response time of units of work transmitted by the second SDR.

24. The method of claim 22 further comprising:

establishing at least a third reliable datagram service, with at least a third SDR independent of the first and second SDRs, between the source device and the destination device;

grouping the SDRs into multiple SDR groups, wherein each of the multiple SDR groups includes at least SDR; and

assigning a unique priority level to each SDR group for effecting throughput and response time of units of work transmitted by the at least one SDR in each SDR group.

25. The method of claim 22 further comprising:

storing scheduling heuristics;

adjusting a first scheduling rate of the first SDR based on the stored scheduling heuristics; and

adjusting a second scheduling rate of the second SDR based on the stored scheduling heuristics.