Perhaps they can use processor affinity to put each user in one core ( https://en.wikipedia.org/wiki/Processor_affinity ).

Sometimes this is useful to make single thread programs fasters, because it avoids the necessity to "transfer" the cache form the old core to the new core. ( http://mailinator.blogspot.com.ar/2010/02/how-i-sped-up-my-s... the article is old (2010) but interesting)

I never tried this in a situation like yours, so I don't know if this is useful in your case.