Bringing N + 1 CPUs out of reset is generally no harder than bringing N CPUs out of reset, when N > 1. Fast, efficient scheduling across that many CPUs is diff

Bringing N + 1 CPUs out of reset is generally no harder than bringing N CPUs out of reset, when N > 1. Fast, efficient scheduling across that many CPUs is difficult, but that isn't even considered in the article. Perhaps I missed something, but why is this news?