| |
 IBM Systems > Deep Computing > Blue Gene >
IBM® System Blue Gene®/P SolutionComparison Chart
|
| |
IBM® System Blue Gene® Solution and IBM System Blue Gene®/P Solution
| Property |
Blue Gene/L |
Blue Gene/P |
| Node Properties |
Node Processors
Processor Frequency
Coherency
L1 Cache (private)
L2 Cache (private)
L3 Cache size (shared)
Main Store
Main Store Bandwidth (1:2 pclk)
Peak Performance |
2* 440 PowerPC®
0.7GHz
Software managed
32KB/processor
7 stream prefetching
2 line buffers/stream
4MB
512MB/1GB
5.6GB/s (16B wide)
5.6GF/node
|
4* 450 PowerPC®
0.85GHz
SMP
32KB/processor
7 stream prefetching
2 line buffers/stream
8MB
2GB
13.6 GB/s (2*16B wide)
13.6 GF/node
|
| Torus Network |
Bandwidth
Hardware Latency (Nearest Neighbor)
Hardware Latency (Worst Case 72 racks)
|
Core/inj/rec packets
6*2*175MB/s=2.1GB/s
200ns (32B packet)
1.6us (256B packet)
7us (68 hops)
|
DMA inj/rec packets
6*2*425MB/s=5.1GB/s
160ns (32B packet)
1.3us (256B packet)
5.0us(68 hops)
|
| Collective Network |
Bandwidth
Hardware Latency (round trip worst case) |
2*350MB/s=700MB/s
6.0us |
2*0.85GB/s=1.7GB/s
5.0us |
System Properties
Example: 72 racks |
Area (square meters)
Peak Performance
Total Power |
150
410TF
1.9MW |
200
1PF
2.9MW |
|
