Silicon Graphics UV 2000, Руководство пользователя системы

Страница: 59 / 116

System Architecture
007-5832-002 39
System Architecture
The SGI UV 2000 computer system is based on a distributed shared memory (DSM) architecture.
The system uses a global-address-space, cache-coherent multiprocessor that scales up to 512
processor cores in a single rack. Because it is modular, the DSM combines the advantages of lower
entry cost with the ability to scale processor count, memory, and I/O independently in each rack.
Note that a maximum of 2,048 cores are supported on a single-system image (SSI). Larger SSI
configurations may be offered in the future, contact your SGI sales or service representative for
additional information.
The system architecture for the SGI UV 2000 system is a sixth-generation NUMAflex DSM
architecture known as NUMAlink 6 or NL6. In the NUMAlink 6 architecture, all processors and
memory can be tied together into a single logical system. This combination of processors,
memory, and internal switches constitute the interconnect fabric called NUMAlink within and
between each 10U IRU enclosure.
The basic expansion building block for the NUMAlink interconnect is the processor node; each
processor node consists of a dual-Hub ASIC (also known as a HARP) and two eight-core
processors with on-chip secondary caches. The Intel processors are connected to the dual-Hub
ASIC via quick path interconnects (QPIs). Each dual-HUB ASIC is also connected to the system’s
NUMAlink interconnect fabric through one of sixteen NL6 ports.
The dual-Hub ASIC is the heart of the processor and memory node blade technology. This
specialized ASIC acts as a crossbar between the processors and the network interface. The Hub
ASIC enables any processor in the SSI to access the memory of all processors in the SSI.
Figure 3-3 on page 40 shows a functional block diagram of the SGI UV 2000 series system IRU.
System configurations of up to eight IRUs can be constructed without the use of external routers.
Routerless systems can have any number of blades up to a maximum of 64. Routerless system
topologies reduce the number of external NUMAlink cables required to interconnect a system.
External optional routers are needed to support multi-rack systems with more than four IRUs, see
Chapter 5, “Optional Octal Router Chassis Information” for more information.