One of the concerns I had with moving away from a microserver to a full Haswell desktop build for ESXi was a jump in power consumption. With around 4 days of data collected since the new system was installed, I thought that I would do a quick comparison for this month.
The graph is a daily consumption of my entire server rack, the aim was to see the overall impact of the new box. Early indications are that with similar usage patterns, my overall power consumption appears to have dropped. The microserver was removed from the rack on the 7th and the Haswell box put into the rack on the 13th. A little early to draw any firm conclusions but it certainly looks promising.