1994
DOI: 10.1007/3-540-58021-2_12
|View full text |Cite
|
Sign up to set email alerts
|

A model for performance estimation in a multistreamed superscalar processor

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
11
0

Year Published

2000
2000
2014
2014

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 12 publications
(11 citation statements)
references
References 9 publications
0
11
0
Order By: Relevance
“…Performance models including the Markov chain based ones have also been adopted for concurrent tasks modeling on the CPUs. Serrano et al [36], [37] developed a model to estimate the instruction throughput of super-scalar processors executing multiple instruction streams. Chen et al [6] proposed an analytical model for estimating throughput of multi-threaded CPUs.…”
Section: Cpu Scheduling and Performance Modelingmentioning
confidence: 99%
“…Performance models including the Markov chain based ones have also been adopted for concurrent tasks modeling on the CPUs. Serrano et al [36], [37] developed a model to estimate the instruction throughput of super-scalar processors executing multiple instruction streams. Chen et al [6] proposed an analytical model for estimating throughput of multi-threaded CPUs.…”
Section: Cpu Scheduling and Performance Modelingmentioning
confidence: 99%
“…Serrano et al [68,69] at the University of Santa Barbara, CA, extended the interleaved multithreading (then called multistreaming) technique to a general purpose superscalar processor architecture and presented an analytical model of multithreaded superscalar performance, backed up by simulation.…”
Section: Multistreamed Superscalarmentioning
confidence: 99%
“…The core idea of SMT aware scheduling is to co-schedule threads with complementary resource requirements. Performance models including the Markov chain-based ones have also been adopted for concurrent tasks modeling on the CPU [114][115][116]. Due to the architectural difference between the CPU and the GPU, their models are not applicable to the GPU.…”
Section: Scheduling On Smt Cpusmentioning
confidence: 99%
“…Several mechanisms have been proposed to select the threads to be executed on the same SMT core, such as hardware counters feedback [108,111], pre-execution of all combinations [109] and probabilistic job symbiosis model [113]. Serrano et al [114,115] and develops a performance model to guide the scheduling decision. Wende et al [122] exploited space sharing for concurrent execution of kernels with a small number of thread blocks.…”
Section: Scheduling On Smt Cpusmentioning
confidence: 99%