• On EC2, I am told that it may take ~5 minutes to start 32 nodes (depending on image size), so with high probability we will finish the LU benchmark within 100 + 300 = 400 secs.
• On the supercomputer, we can use Rich Wolksi’s QBETS queue time estimation service to get a bound on the queue time. When I tried this in June, QBETS told me that if I wanted 32 nodes for 20 seconds, the probability of me getting those nodes within 400 secs was only 34%–not good odds.

ianfoster.typepad.com, Ian Foster, Aug 2009

These comparisons depends from my point of view always on experienced data and mainly on your various scenario Neutral declarations are hard to find. I asked Thijs Metsch for some experienced data of RESERVOIR project he mentoined in Twitter. Looking forward for an answer from him…
Does anyone else have some own experienced data for starting EC2 images in different environments and with various scenarios?