<HTML><BODY>I worked always w/very small HPC clusters and built them manually (each server). <br>But what is reasonable to do for clusters containing some tens or hundred of nodes ?<br>Of course w/modern Xeon (or Xeon Phi KNL) and IB EDR, during the next year for example. <br>There are some automatic systems like OSCAR or even ROCKS. <br><br>But it looks that ROCKS don't support modern interconnects, and there may be problems <br>w/OSCAR versions for support of systemd-based distributives like CentOS 7. For next year -<br>is it reasonable to wait new OSCAR version or something else ?<br><br>Mikhail Kuzminsky,<br>Zelinsky Institute of Organic Chemistry RAS,<br>Moscow <br><br><br><br></BODY></HTML>