You'll have to do benchmarks on your own.
As a rule, we don't release benchmarks because there are just too many variables and too much risk. Understand that if we were to release benchmark statistics, they would only be used against us... our customers would constantly point back to the benchmarks as the gold standard for what they absolutely must get in their own environments. Unfortunately, it then becomes Symantec's problem when their misconfigured/overloaded/un-tested/ancient environment doesn't meet the mark. The best we could hope for in those cases is a finger-pointing stand-off where we each point at the other. We simply can't set ourselves up for those negative scenarios.
All that said, if you can take my word for it, I can tell you about a pre-beta internal report on SEP 12.1's performance vs SEP 11.0 that said the dev team's goal was to meet or exceed 11.0's performance. The keyword is "goal" and I don't have anything showing whether or not they met it. However, based on my own experience with the product, it is faster/lighter in almost every case.
Finally, performance is important, but I'd encourage you to investigate the dramatic improvments in SEP 12.1's security improvements. At the end of the day, the performance will be close but the difference in security technology is almost uncomparable. Even if SEP 12.1 scored worse in performance, the gain in security is worth it and will pay-off in the long run.