Starting November 26, 2012, the nucleotide collection (nt) will be the default nucleotide search database. The nucleotide collection consists of GenBank+EMBL+DDBJ+PDB+RefSeq sequences, but excludes EST, STS, GSS, WGS, TSA, patent sequences as well as phase 0, 1, and 2 HTGS sequences.
The nt database recently joined a list of other fast, indexed database searches offered by the NCBI that include the human G+T (genome plus transcript) and mouse G+T databases, as well as the human and mouse reference genome databases. The indexed databases are available when using megaBLAST. Indexed megaBLAST, at the NCBI BLAST web page, can search queries of a couple thousand bases against the 43 billion base nt database in a few seconds.
Indexed searches at the NCBI use an in-memory index described by Morgulis et al. and the megaBLAST algorithm.