In this paper, we present I-PACK, a deterministic layout algorithm for embedding a data set X in 2D provided that distances (duv)u,v?X between data items are given or can be computed. The layout reflects well similarities and dissimilarities between items and it is computed in quasi-linear time. Experimental comparisons with other multidimensional scaling algorithms show that : i) our algorithmhas similar performance when the aspect ratio A = \frac{{\max _{u,v} (\delta uv)}} {{\min _{u,v} (\delta uv)}} is small (i.e. log2A \lt 10) and ii) the larger the aspect ratio, the better I-PACK performs with respect to otherMDS algorithms. This is also true when data can be ?naturally? clustered.