If the matrix subtraction reveals multiple microservice changes, system 100 can reference the analytics provided by the relative frequency matrix to determine the most likely source of the problem as being the changed microservice that is most frequently called, as described above. If the problem is not resolved by roll back of that microservice, system 100 can roll back the microservice next most likely to be the source of the problem (the microservice corresponding to the next highest relative frequency). System 100 can continue to roll back microservices iteratively beginning with the microservice having the highest relative frequency until the problem is resolved.

Matrix subtraction of one matrix from another requires that both matrices have identical dimensions. Thus, in the event that two matrices are not of the same dimension due to the introduction of one or more new processes or microservices (the dimensions of the updated matrix increases with each newly introduced process or microserve), the identification of a faulty/erroneous element can be determined differently. Specifically, the identification of a faulty/erroneous element can be traced back to the new process or microservice (newly introduced into an updated matrix) by determining post-updated row and column counts of each microservice and comparing these to the counts before the introduction of the new process or microservice. The procedure will provide an indicative analysis of how the dynamic behavior of the prior network changed with introduction of the new process or microservice.