IBM eServer p5 575 (Cluster note)


|
The IBM eServer p5 575 p5 575 cluster nodes are designed as high-powered building blocks for "extreme" performance computing applications. Featuring either innovative dual-core 1.9GHz IBM POWER5 modules with only a single core active or conventional 1.5GHz dual-core modules packed into a highly dense enclosure, the p5-575 delivers very high memory bandwidth. The 8-way node offers twice the per processor cache of any POWER5 SMP while the 16-way node offers the ability to achieve a new level of price/performance for compute intensive applications.

| Processor |
POWER5 8-way or 16-way |
| Clock rates (Min/Max) |
1.90GHz (8-way), 1.50GHz (16-way) |
| System memory (Std/Max) |
1GB / 256GB |
| Internal storage (Std/Max) |
73.4GB / 2.9TB |
| Linpack Performance (GFLOPS)1 |
56.67 (8-way), 87.3 (16-way) |
|
|

|

|

|
IBM eServer p5 590(Frame-mount)


|
The IBM eServer p5 590 server uses fifth-generation 64-bit IBM POWER5 processors in up to 32-way symmetric multiprocessing (SMP) configurations to provide the power, flexibility, scalability and security needed for mission-critical applications at a very affordable price.

| Processor |
POWER5 8-way to 32-way |
| Clock rates (Min/Max) |
1.65GHz |
| System memory (Std/Max) |
8GB / 1TB |
| Internal storage (Std/Max) |
72.8GB / 9.3TB (using optional I/O drawers) |
| Performance (rPerf range)*** |
41.68 to 151.72 |
|
|

|

|

|
IBM eServer p5 595(Frame-mount)


|
The eServer p5 595 server uses fifth-generation 64-bit IBM POWER5 processors in up to 64-way symmetric multiprocessing configurations. p5-595 is IBM's most powerful UNIX/Linux system with the power, flexibility, scalability and security needed to run mission-critical enterprise applications.

| Processor |
POWER5 16-way to 64-way |
| Clock rates (Min/Max) |
1.65GHz / 1.90GHz |
| System memory (Std/Max) |
8GB / 2TB |
| Internal storage (Std/Max) |
72.8GB to 14.0TB (using optional I/O drawers) |
| Performance (rPerf range)*** |
80.86 to 306.21 |
|
|


*** The benchmarks and values shown herein were derived using particular, well configured, development-level computer systems. Actual system performance may vary and is dependent upon many factors including system hardware configuration and software design and configuration.
1 Linpack is a collection of Fortran subroutines that analyse and solve linear equations and linear least-squares problems. For details on the latest Linpack results, visit: http://www.netlib.org/benchmark/performance.ps.
|
 |
|