FIG. 11 illustrates a block diagram 1100 that shows an example for adjustment of weights via back propagation by computing a margin of error during training of the machine learning module 106 based on local queuing 1004, in accordance with certain embodiments.

In FIG. 11, in one example, the following are the values of certain parameters:

(i) The actual output of the machine learning module=N (reference numeral 1102);

(ii) The number of I/O operations queued in the local queue of port=M (reference numeral 1104);

(iii) Number of free TCBs=0 (reference numeral 1106).

For this example, the expected output 1108 of the machine learning module 106 is calculated as N+M (as N+M allocated TCBs would remove the I/O operations from the local queue).

In FIG. 11, in another example, the following are the values of certain parameters:

(i) The actual output of the machine learning module=N (reference numeral 1110);

(ii) The number of I/O operations queued in the local queue of port=0 (reference numeral 1112);

(iii) Number of free TCBs=M (reference numeral 1114).

For this example, the expected output 1116 of the machine learning module 106 is calculated as N?M (as there are M excess TCBs in the local queue).