2011 International Conference on High Performance Computing &Amp; Simulation 2011
DOI: 10.1109/hpcsim.2011.5999914
|View full text |Cite
|
Sign up to set email alerts
|

Self-related traces: An alternative to full-system simulation for NoCs

Abstract: The network-on-chip (NoC)

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2011
2011
2019
2019

Publication Types

Select...
4
4

Relationship

1
7

Authors

Journals

citations
Cited by 13 publications
(4 citation statements)
references
References 7 publications
0
4
0
Order By: Relevance
“…For architecture simulations, a Replay mechanism is required to process the trace and generate architectural events, as is standard in event-driven simulation [14,15,29,32,34]. The Replay mechanism dynamically generates the appropriate actions for all events during simulation while providing lightweight thread scheduling and management.…”
Section: Event-trace Replay Frameworkmentioning
confidence: 99%
See 1 more Smart Citation
“…For architecture simulations, a Replay mechanism is required to process the trace and generate architectural events, as is standard in event-driven simulation [14,15,29,32,34]. The Replay mechanism dynamically generates the appropriate actions for all events during simulation while providing lightweight thread scheduling and management.…”
Section: Event-trace Replay Frameworkmentioning
confidence: 99%
“…Trace-based approaches have also been employed to specifically explore the CMP design space [14,15,29,34]. Most work in this space has recognized the need to establish causation between network messages to model the associated delays correctly.…”
Section: Other Efficient Modeling Solutionsmentioning
confidence: 99%
“…The simulation platform is based on the Simics-GEMS simulators [14]. The NoC is modeled with a cycle-accurate flit-level network simulator.…”
Section: Performance Evaluationmentioning
confidence: 99%
“…Existing works on trace capturing involve both single node and distributed cluster. Trivino et al [42] propose a self-related trace model to capture the trace for a network-on-chip (NoC) simulator, whereas other works [4,7,33] present the mechanisms to get communication traces of MPI-parallel applications. Miguel-Alonso et al [33] make use of the multi-processing environment (MPE) [25], which is a set of libraries and tools used for generating and analyzing traces of parallel applications, to get the trace files.…”
Section: Introductionmentioning
confidence: 99%