| Features/ Benefits
Designed for HPC and Built to
Scale
• Managed, reliable, scalable
performance
• Delivering 4TF to 1000TF
of computing power
• InfiniBand, PCI-Express,
Mellanox ConnectX adapters, Intel
fiber optic interconnects between
racks and multi-core processor
technologies to deliver outstanding
performance and scalability to
1000s of nodes.
• High performance and high
density combined into an innovative
new board design, delivering 64
nodes/128 CPUs/512 Quad-Core AMD
Barcelona quad core processors
and up to 4TF of computing power
in a single 44U equipment rack.
• Complete lights-out management
for the entire system including
the interconnect networks, servers,
clusters, resource management
and scheduling. It supports
diskless operation and multi-rail
networks to achieve maximum reliability,
performance and high availability
with load balancing and nonstop
network failover.
Reduces TCO - Energy-Efficient
(Green) Architecture
• Directed Airflow™
cooling configuration reduces
datacenter floor space by 30%
while maximizing power and cooling
efficiency
• “Power-Factor-Corrected”
power supplies that are 90% efficient
compared to the 70% to 80% efficiency
typical of power supplies used
in current equipment.
• Delivers High Value –
Reliability and availability at
an impressive price/performance
ratio
• Redundant (Dual-Rail)
Infiniband network fabric minimizes
node to node latency
• Low latency Mellanox ConnectX
host channel adapters minimizes
node to node latency
• Redundant (Dual-Rail)
Ethernet Management Fabric
• Improved reliability architecture
with redundant power supplies,
fans and switches. A critical
components are easily accessible,
hot-swappable for reliability
and serviceability.
• Meets production system
availability requirements
Simplifies IT Management and Administration
• Complete lights-out management
system
• Supports root file systems
offering instant provisioning
for rapid, standard Linux installs
on large diskless systems allowing
them to boot 100 to 1,000 blades
in about the same time.
Appro Cluster Engine™ Management
Sofware
• Network Management
• Server Management
• Cluster Management
• Storage Management
Appro Cluster Management Architecture
• Scalable to very large
configurations
• Hierarchical architecture
• Supports multiple interconnect
topologies
• Diskless operation without
lightweight kernel
• Reliable, available, serviceable
(RAS)
- Multi-Rail Fabric
with load-balancing
- Run-through fault
tolerance for network failures
Support and Services
• Closed-loop solution management,
up-front consulting, pre-integration
and testing with pre-sales and
post-sales support.
• Support for Linux operating
environments, with configurations
that provide exceptional performance
across a wide range of HPC applications.
AMD Opteron™ Processors
• Outstanding Performance
- Exceptional performance and
performance-per-watt solutions
for today’s most demanding
business applications
• Investment Protection
- AMD64 technology allows you
to minimize the cost of transitions
and maximize past investments
in hardware, software, and personnel
• Enhanced Power Efficiency
- Designed to help reduce your
power and cooling costs by lowering
the energy consumption of your
IT infrastructure
• Optimal Virtualization
- Helps simplify consolidation
and improve computing flexibility
with the industry’s best
x86 architecture and platform
support
• AMD PowerNow! Technology
with Optimized Power Management
- Delivers increased power savings
by optimizing performance-per-watt
with power on demand capabilities
• Direct Connect Architecture
- Memory is directly connected
to the CPU optimizing memory performance
and reducing latencies during
memory access over traditional
front-side bus-based memory controllers.
• AMD Virtualization (AMD-V)
Technology - Designed to dramatically
increase the performance of virtualized
applications while enabling faster
switching between virtual machines
for a powerful, flexible, secure
server solution.
|