Recent observations have shown that there is a tight correlation between star formation rate (SFR) and stellar mass (M*) of star-forming galaxies. This SFR-M* relation - the so-calld "star formation main sequence (SFMS)" - is one of the most fundamental scaling relations for star-forming galaxies. At least in the local universe, it is reported that the SFMS is universal everywhere: i.e. the SFMS is independent of environment. In my talk, I will present our recent efforts to investigate the SFMS and its environmental variation out to z>2. By compiling our large H-alpha selected galaxy samples in distant clusters and those in general field environments, we find that there is no detectable environmental dependence in the SFMS out to z~2, as far as we rely on the simple H-alpha based SFRs. However, in contrast to this apparently simple picture, we also find an interesting evidence that SF galaxies in cluster environments tend to be more massive, and perhaps more highly obscured by dust. We argue that such massive/dusty galaxies surviving in distant cluster environments should hold the key for understanding environmental effects on galaxy evolution in the early universe.