Cricket Liu of Infoblox has posted a second article in his series on DNSSEC overhead.  He notes: 

…I’ve recommended that organizations deploying DNSSEC watch the CPU load on their recursive name servers carefully:  As the proportion of responses that are signed increases, so will the load on their recursors. Ultimately, though, the ever-increasing speed of processors and networks will trump the burden DNSSEC adds.  Years from now – assuming DNSSEC becomes widely deployed – we’ll look back at our concerns about the overhead of DNSSEC and chuckle.  I hope.