The performance of a vector processor accessing vectors is strongly dependent on the conflicts produced in the memory subsystem. These conflicts delay the work of the functional units. The concurrent memory access of several vector streams causes inter-conflicts between the references of the different vector streams. In a memory system where several memory modules are mapped in every bus (complex memory system) the number of conflicts increases because the bus must be shared by vector streams. Conflicts named linked conflicts and complex conflicts decrease the performance of the steady-state phase in complex memory systems (CRAY X-MP). This paper proposes a method that allows a concurrent access to several vector streams reducing the averag...
The synchronized and simultaneous access to several vectors that form a single stream occurs in SIMD...
This paper presents mathematical foundations for the design of a memory controller subcomponent that...
Address transformation schemes, such as skewing and linear transformations, have been proposed to ac...
The high latency of memory accesses is one of the factors that most contribute to reduce the perform...
Most existing analytical models for memory interference generally assume random bank selection for e...
When accessing streams in vector multiprocessor machines, degradation in the interconnection network...
The poor bandwidth obtained from memory when conflicts arise in the modules or in the interconnectio...
Vector supercomputers, which can process large amounts of vector data efficiently, are among the fas...
SIGLECNRS 14802 E / INIST-CNRS - Institut de l'Information Scientifique et TechniqueFRFranc
Abstract- The partitioning of shared memory into a number of memory modules is an approach to achiev...
Memory access time is a key factor limiting the performance of large-scale, shared-memory multiproce...
Presently, the highest performance computer systems are the vector processors which are mainly emplo...
On many commercial supercomputers, several vector register processors share a global highly interlea...
Address transformation schemes, such as skewing and linear transformations, have been proposed to ac...
The purpose of this paper is to show that multi-threading techniques can be applied to a vector proc...
The synchronized and simultaneous access to several vectors that form a single stream occurs in SIMD...
This paper presents mathematical foundations for the design of a memory controller subcomponent that...
Address transformation schemes, such as skewing and linear transformations, have been proposed to ac...
The high latency of memory accesses is one of the factors that most contribute to reduce the perform...
Most existing analytical models for memory interference generally assume random bank selection for e...
When accessing streams in vector multiprocessor machines, degradation in the interconnection network...
The poor bandwidth obtained from memory when conflicts arise in the modules or in the interconnectio...
Vector supercomputers, which can process large amounts of vector data efficiently, are among the fas...
SIGLECNRS 14802 E / INIST-CNRS - Institut de l'Information Scientifique et TechniqueFRFranc
Abstract- The partitioning of shared memory into a number of memory modules is an approach to achiev...
Memory access time is a key factor limiting the performance of large-scale, shared-memory multiproce...
Presently, the highest performance computer systems are the vector processors which are mainly emplo...
On many commercial supercomputers, several vector register processors share a global highly interlea...
Address transformation schemes, such as skewing and linear transformations, have been proposed to ac...
The purpose of this paper is to show that multi-threading techniques can be applied to a vector proc...
The synchronized and simultaneous access to several vectors that form a single stream occurs in SIMD...
This paper presents mathematical foundations for the design of a memory controller subcomponent that...
Address transformation schemes, such as skewing and linear transformations, have been proposed to ac...