Fault tolerant systems and methods incorporating a minimum checkpoint interval
View Patent ↗In part, disclosure relates to a method of regulating checkpointing in an active active fault tolerant system. The method includes receiving a request from a client through a network at a primary computer; copying, by the primary computer, the request from the client to a secondary computer; processing the request from the client, using the primary computer, to generate a primary computer result; processing the copy of the request from the client, using the secondary computer, to generate a secondary computer result; comparing the primary computer result and the secondary computer result to obtain a comparison metric; determining whether a minimum checkpoint interval has been met or exceeded; and if the minimum checkpoint interval has not been met or exceeded, delay initiating a checkpoint process from primary computer to secondary computer.
1. A method of regulating checkpointing in an active active fault tolerant system, the method comprising:
receiving a request from a client through a network at a primary computer of the active active fault tolerant system;
copying, by the primary computer, the request from the client to a secondary computer of the active active fault tolerant system;
processing the request from the client, using the primary computer, to generate a primary computer result;
processing the copy of the request from the client, using the secondary computer, to generate a secondary computer result;
comparing the primary computer result and the secondary computer result to obtain a comparison metric;
determining a minimum checkpoint interval by running one or more simulations under various conditions and for different use cases and tracking which minimum checkpoint threshold results in performance gains for each of the foregoing;
determining whether the minimum checkpoint interval has been met or exceeded; and
if the minimum checkpoint interval has not been met or exceeded, delay initiating a checkpoint process from primary computer to secondary computer, wherein checkpoint processes are performed to maintain fault tolerance when managing a plurality of client requests.
2. The method of claim 1 further comprising initiating a checkpoint process if the comparison metric indicates that primary computer and secondary computer are diverging or differing with regard to one or more comparable parameters.
3. The method of claim 1 , wherein the minimum checkpoint interval ranges from about 30 ms to about 70 ms.
4. The method of claim 1 , wherein the minimum checkpoint threshold time ranges from about 50 ms to about 100 ms.
5. The method of claim 1 , wherein if the primary computer result and the secondary computer result diverge, holding the primary computer result until completion of a next checkpoint process.
6. The method of claim 1 , wherein if the primary computer result and the secondary computer result match, sending the primary computer result to the client.
7. The method of claim 1 , further comprising:
creating a first thread, at the primary computer, for processing the request from the client;
creating a second thread, at the secondary computer, for processing the copy of the request from the client; and queuing network packets from first thread and second thread until next checkpoint.
8. The method of claim 1 , wherein request from client comprises input/output (I/O) request and/or a packet.
9. The method of claim 1 , wherein the minimum checkpoint interval ranges from about 20 ms to about 60 ms.
10. The method of claim 1 , wherein the minimum checkpoint threshold time ranges from about 60 ms to about 120 ms.
11. The method of claim 1 further comprising queuing mismatched network packets while delaying initiating a checkpoint process to improve overall active active fault tolerant system performance.