next up previous contents
Next: The NEC SX-4. Up: Shared-memory MIMD systems Previous: The HP/Convex C4 series.

The Digital Equipment Corp. AlphaServer.

Machine type RISC-based distributed-memory multi-processor
Models AlphaServer 8200 5/300, 8400 5/300
Operating system Digital Unix (DEC's flavour of Unix)
Connection structure Crossbar
Compilers Fortran 77, HPF, C, C++
Vendors information Web page

System parameters:

Model 8200 8400
Clock cycle 3.3 ns 3.3 ns
Theor. peak performance
Per proc. (64-bit) 600 Mflop/s 600 Mflop/s
Maximal (64-bit) 3.6 Gflop/s 7.2 Gflop/s
Main memory <=6 GB <=14 GB
Memory bandwidth
Processor/memory 1.6 GB/s 1.6 GB/s
No. of processors 6 12


The AlphaServers are symmetric multi-processing systems which are based on the Alpha 21164 processor. The 8200 model is a somewhat smaller copy of the 8400 model: in the 8200 a maximum of 6 CPUs can be accomodated while this number is 12 for the 8400 model. Also, there is room for at most 6 GB of memory in the 8200 while the 8400 can house 14 GB. However, the amount of CPUs and memory is not independent. For instance, the 8400 has 9 system slots. One of these is reserved for I/O and one will have to contain at least one CPU module which can contain 1 or 2 CPUs. From the remaining slots 6 can be used either for memory or for a CPU module. So, one has to choose for either higher computational power or for more memory. This can potentially be a problem for large applications that require both.

As with the SGI PowerChallenge (see 3.3.6), AlphaServers can be clustered using a PCI bus Memory channel for interconnection of the systems. The systems need not be of the same model. The bandwidth of this interconnect is 100 MB/s. Eight systems can be coupled in this way. To support this kind of cluster computing, HPF and optimised versions of PVM and MPI are available.

Measured Performances: As yet, only single system results for the AlphaServer 8400 are available. No cluster results are known. In [#linpackbm##1#] a speed of 5.0 Gflop/s for an 8-processor system are reported for the solution of a dense linear system of order 9548. In [#nasbm##1#] results for the NAS parallel benchmarks are given. For the class B EP benchmark a time of 78.43 seconds was measured for 8 processors. For the class B LU, SP, and BT benchmarks times of 296.19, 364.54, and 458.21 seconds were found, respectively.

next up previous contents
Next: The NEC SX-4. Up: Shared-memory MIMD systems Previous: The HP/Convex C4 series.

Jack Dongarra
Sat Feb 10 15:12:38 EST 1996