Rémy Evard

Designing the future of collaborative science: Argonne's Futures Laboratory

Argonne National Laboratory created its Computing and Communications Infrastructure Futures Laboratory to develop next-generation computing and communications infrastructure systems. An important goal of the Futures Lab project is to understand how to incorporate advanced display and media server systems into scientific computing environments. The objective is to create new collaborative environment technologies that combine advanced networking, virtual-space technology, and high-end virtual environments to construct virtual teams for scientific research. We hope to add new capabilities to scientific computing systems and to accelerate development of virtual organizations to conduct experimental, computational, and theoretical science. Our research focuses largely on developing new technology and integrating it with selected applications projects to demonstrate the technology. We want to drive large-scale supercomputer systems architecture toward more balanced (and human-oriented) systems that support a variety of human interfaces and media systems. Large-scale parallel computing and immersive technology are naturally complementary, so we are striving to develop improved mechanisms for linking parallel supercomputers and immersive displays. The collaborative environments we envision are logical extensions of the classic Internet. The Futures Lab and its associated research projects are a new research focus for Argonnes Mathematics and Computer Science Division. Our approach is both interdisciplinary and collaborative. Two primary projects-the multimedia and collaborative space project and the virtual environments project-are under way, both supported by activities in advanced networking. >

Scalable cluster administration - Chiba City I approach and lessons learned

Systems administrators of large clusters often need to perform the same administrative task hundreds or thousands of times. Administrators have traditionally performed some time-consuming tasks, such as operating system installation, configuration, and maintenance, manually. By combining network services such as DHCP, TFTP, FTP, HTTP, and NFS with remote hardware control and scripted installation, configuration, and maintenance techniques, cluster administrators can automate these administrative tasks. Scalable cluster administration addresses this challenge: What hardware and software design techniques can cluster builders use to automate cluster administration on very large clusters? We describe the approach used in the Mathematics and Computer Science Division of Argonne National Laboratory on Chiba City I, a 314-node Linux cluster; and we analyze the scalability, flexibility, performance and reliability benefits and limitations from that approach.