problem-constrained scaling does the same problem but on more processors and focuses on trying to compute/solve the problem faster.
Speedup = time it takes to get job done on 1 processor/time it takes for P processors. (ideal speedup is P)
time-constrained scaling focuses on doing something more advanced (completing more work) in a fixed amount of time. Speedup = work done by P processors/work done by 1 processor
memory-constrained scaling focuses on how much faster we can work on bigger problems given that we scale up from using 1 processor to P processors.
Speedup = work per unit time on P processors/work per unit on 1 processor (ie: transactions per sec on P processors vs on 1 processor)
There are three resource-oriented scaling models:
problem-constrained scaling does the same problem but on more processors and focuses on trying to compute/solve the problem faster. Speedup = time it takes to get job done on 1 processor/time it takes for P processors. (ideal speedup is P)
time-constrained scaling focuses on doing something more advanced (completing more work) in a fixed amount of time. Speedup = work done by P processors/work done by 1 processor
memory-constrained scaling focuses on how much faster we can work on bigger problems given that we scale up from using 1 processor to P processors. Speedup = work per unit time on P processors/work per unit on 1 processor (ie: transactions per sec on P processors vs on 1 processor)