In 1993, sizes of on-chip caches on current commercial microprocessors range from 16 Kbytes to 36 Kbytes. These microprocessors can be directly usedin the design of a low cost single-bus shared memory multiprocessors with-out using any second-level cache. In this paper, we explore theviability of such a multi-microprocessor. Simulations results clearly establish that performance of such a system will be quite poor if on-chip caches are direct-mapped. On the other hand, when the on-chip caches are partially associative, the achieved level of performance is quite promising. In particular, two recently proposed innovative cache structures, the skewed associative cache organization and the semi-unified cache organization are shown to work fine
L'accroissement de la vitesse des déplacements depuis la Révolution des transports au 19ème siècle e...
International audienceFrance is subjected to several types of natural phenomena (floods, earthquake,...
Bien que le spinozisme d'Einstein ne soit pas celui d'un érudit et que l'on doive parler, plutôt que...
L'évolution de l'architecture des microprocesseurs est très rapide ; dans ce rapport nous présentons...
Ce document présente une analyse comparative de divers langages prototypes et modèles formels exista...
Projet POEMSDans ce rapport, nous effectuons une présentation mathématique détaillée de la méthode d...
We provide analytical results about the performance of various Connected Dominating Set (CDS) algori...
We consider the problem of disseminating messages in networks whose topology and size are not known ...
In this paper, we propose the auto-associative (AA) model to generalize the Principal component anal...
International audienceIn the feedback vertex set problem, the aim is to minimize, in a connected gra...
Odyssée is an automatic differentiation (AD) package developed at INRIA. This tool is able to differ...
The problem studied in this paper originates from a just-in-time environment where a new supplying m...
We present Coca, an automated debugger for C, where the breakpoint mechanism is based on events rela...
International audienceIn this paper, we study the validity of the assumption that speech source sign...
The ATerm library is a well-designed and well-known library in the term rewriting community. In this...
L'accroissement de la vitesse des déplacements depuis la Révolution des transports au 19ème siècle e...
International audienceFrance is subjected to several types of natural phenomena (floods, earthquake,...
Bien que le spinozisme d'Einstein ne soit pas celui d'un érudit et que l'on doive parler, plutôt que...
L'évolution de l'architecture des microprocesseurs est très rapide ; dans ce rapport nous présentons...
Ce document présente une analyse comparative de divers langages prototypes et modèles formels exista...
Projet POEMSDans ce rapport, nous effectuons une présentation mathématique détaillée de la méthode d...
We provide analytical results about the performance of various Connected Dominating Set (CDS) algori...
We consider the problem of disseminating messages in networks whose topology and size are not known ...
In this paper, we propose the auto-associative (AA) model to generalize the Principal component anal...
International audienceIn the feedback vertex set problem, the aim is to minimize, in a connected gra...
Odyssée is an automatic differentiation (AD) package developed at INRIA. This tool is able to differ...
The problem studied in this paper originates from a just-in-time environment where a new supplying m...
We present Coca, an automated debugger for C, where the breakpoint mechanism is based on events rela...
International audienceIn this paper, we study the validity of the assumption that speech source sign...
The ATerm library is a well-designed and well-known library in the term rewriting community. In this...
L'accroissement de la vitesse des déplacements depuis la Révolution des transports au 19ème siècle e...
International audienceFrance is subjected to several types of natural phenomena (floods, earthquake,...
Bien que le spinozisme d'Einstein ne soit pas celui d'un érudit et que l'on doive parler, plutôt que...