Chapter 5 Robust Performance Tailoring with Tuning - SSL - MIT

configurations until one works. The fact that such a large number of tests are required
for such a simple problem (only two tuning and two uncertainty parameters) indicates
that this approach is fairly ill-suited for this application. The barrier steepest-descent
real-time optimization performs much better than SA, requiring an average of 46
hardware tests to tune successfully. The maximum value of tests is still pretty high at
231, however, and is expected to increase with the number of tuning and uncertainty
parameters in the problem.
The isoperformance tuning algorithm, a hybrid of model and hardware tuning
techniques, stands out as the clear winner. Like the hardware-only methods it suc-
cessfully tunes each and every one of the hardware realizations, but does so with only
a few hardware tests. The maximum number of tests required by the isoperformance
tuning method is 4 and the average across the sample space is 2.5. These statistics
are a factor of ten less than BSD and a hundred times less than SA. It is true that
the iso-lines in this problem are particularly well-suited for this tuning methodology
as discussed previously, but the additional example considered shows that even if
the isoperformance lines are less favorable the method performs very well. As the
complexity of the model and tuning problem increases, it is expected that the model-
only methods will fail more often and that the hardware tuning methods will require
even more tests to find a successful tuning configuration. Therefore, although the
model-only methods are attractive for their simplicity and low-cost, and the hard-
ware methods are attractive for their success rate, neither method is really a practical
solution as there is always the chance of failure or prohibitive expense. In contrast,
the isoperformance tuning method is able to consistently provide successful tuning
solutions with only a small number of hardware tests.
4.4 Summary
In this chapter, the concept of dynamic tuning is defined as the process of adjusting
hardware to bring the system performance within requirements. The tuning process is
formalized as an optimization, and guidelines for choosing appropriate tuning param-
148