Cray-1 Computer System Hardware Reference Manual 2240004, 1979, Cray Research Inc., Mendota Hts., MN, 55120. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Parallel Programming Languages - Pingali (1998) (Correct)
....in these languages contain a combination of scalar and vector operations. On array processors like CM 2 [41] the scalar operations are usually performed by a front end high performance work station, while the vector operations are performed on the array processor. Vector processors like the CRAY [11] can execute both scalar and vector instructions. Therefore, the key problem in designing a SIMD language is to design constructs that expose as many vector operations as possible to the compiler. 2.1 Shared Memory SIMD Languages The simplest vector operations involve the application of an ....
....if control vectors are used. The operands and results of vector operations are usually stored in memory, so it is usually not worth performing an arithmetic operation in vector mode unless the loads and stores can also be done in vector mode. However, some computers (such as the CRAY 1 and CRAY 2 [11]) permit only constant stride loads and stores from memory. Unless the compiler can determine that the true entries in a control vector occur with a fixed stride, it is forced to generate scalar loads and stores. The IBM VECTRAN language [36] addressed this problem by introducing array triplets ....
CRAY Research Inc., Bllomington, Minn. CRAY-1 Computer System Hardware Reference Manual, 1978.
Why Do We Need a Floating-Point Arithmetic Standard? - Kahan (1981) (2 citations) (Correct)
No context found.
Cray-1 Computer System Hardware Reference Manual 2240004, 1979, Cray Research Inc., Mendota Hts., MN, 55120.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC