Methods for estimating the performance of database management systems
can aid the design of database systems by identifying potential perfor
mance bottle-necks or by predicting the relative performance of differ
ent designs. performance estimation is critical in parallel database s
ystems with distributed memory, where an effective overall performance
depends on a good choice among a wide range of ways of placing data.
An approach is described for performance estimation for shared-nothing
parallel database systems. It estimates system throughput for a given
benchmark or set of queries, and can exercise different data placemen
t schemes to determine the data layout that provides the best throughp
ut value.