We typically run something along the lines of an Agilent Bioanalyzer and for cancer samples we run a CGH or SNP array to verify sample purity.

Though we do not do it routinely, I have heard of some groups running only small quantities of a library in a multiplexed lane (with several other test samples) to verify the integrity of the DNA. I would guess that this technique might be applicable to only a subset of possible sequencing technologies or applications.