Performance measurements using several commercial applications
and performance benchmarks (SpecJBB, SpecWeb,
TPC-C, SpecIntRate, SpecFPRate, etc.) confirm that Niagara2
has achieved its goal of doubling the throughput performance
Fig. 8. Key statistical highlights.
and performance/watt as compared to UltraSPARC T1. Most of
the gain comes from doubling the thread count and the number
of execution units. Some of the gain comes from a higher
operating frequency. Similarly, performance measurements
using commonly used Floating Point benchmarks confirm that
Niagara2’s Floating Point throughput performance is more than
an order of magnitude higher compared to UltraSPARC T1.
Niagara2 has eight Floating Point units (FPUs), each occupying
only 1.3 mm , against only one FPU for UltraSPARC T1.
Also, the Niagara2 FPUs are within the SPCs as compared
to UltraSPARC T1 where the SPCs had to access the FPU
through the Crossbar. Another factor that helps performance is
the higher memory bandwidth on Niagara2