For computer science in general, and database systems in particular, papers in the best conferences are considered equivalent to high-quality journal papers. To understand why, see this document, authored by highly eminent professors of computer science [David Patterson (Univ. of California, Berkeley), Lawrence Snyder (Univ. of Washington, Seattle), and Jeffrey Ullman (Stanford University)]
To quantify the above, the typical paper acceptance ratios in the best database conferences (such as ACM SIGMOD, VLDB, IEEE ICDE and EDBT) are around 1:7, selected from a few hundred international submissions. A recent example is ICDE 2005, where 67 papers were accepted from 521 submissions, corresponding to a 1:8 selectivity. Further, even in second-tier conferences, the acceptance ratio is usually less than 1:4.