,----[ http://sc-2002.org/paperpdfs/pap.pap207.pdf ]
|
| Abstract: This paper gives an overview of the
| BlueGene/L Supercomputer.
| [ . . . ]
|
| This massively parallel system of 65,536 nodes is
| based on a new architecture that exploits
| system-on-a-chip technology to deliver target peak
| processing power of 360 teraFLOPS (trillion
| floating-point operations per second). The machine is
| scheduled to be operational in the 2004-2005 time
| frame, at price/performance and power
| consumption/performance targets unobtainable with
| conventional architectures.
| [ . . . ]
|
| Our goal in developing the system software for BG/L
| has been to create an environment which looks
| familiar and also delivers high levels of application
| performance. The applications get a feel of executing
| in a Unix-like environment.
|
| The approach we have adopted is to split the
| operating system functionality between compute and
| I/O nodes.
| [ . . . ]
|
| The compute node operating system, also called the
| BlueGene/L compute node kernel, is a simple,
| lightweight, single-user operating system that
| supports execution of a single dual-threaded
| application compute process. Each thread of the
| compute process is bound to one of the processors in
| the compute node. The compute node kernel...
| [ . . . ]
|
| I/O nodes are expected to run the Linux operating
| system, supporting the execution of multiple
| processes. Only system software executes on the I/O
| nodes, no application code. The purpose of the I/O
| nodes during application execution is to complement
| the compute node partition with services that are not
| provided by the compute node software. I/O nodes
| provide an actual file system to the running
| applications.
`----