Asvie: A Timing-Agnostic SVE Optimization Methodology

Cruz, Miguel Tairum; Ruíz, Daniel; Rusitoru, Roxana

doi:10.1109/protools49597.2019.00007

Cited by 4 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our technique for model swapping was implemented in SVE-Cachesim, which is an in-order, Python-based cache simulator developed for [14]. While a simple model such as this ignores some parts of the cache system, such as the effects of pre-fetching, by focusing on a component in the middle of the system, the level one data cache, we were able to study cascading downstream effects introduced by model swapping, and pick models that minimized this disruption.…”

Section: A Methodology 1) Simulatormentioning

confidence: 99%

Online Model Swapping in Architectural Simulation

Lavin¹,

Young²,

Vuduc³

et al. 2020

Preprint

View full text Add to dashboard Cite

As systems and applications grow more complex, detailed simulation takes an ever increasing amount of time. The prospect of increased simulation time resulting in slower design iteration forces architects to use simpler models, such as spreadsheets, when they want to iterate quickly on a design. However, the task of migrating from a simple simulation to one with more detail often requires multiple executions to find where simple models could be effective, which could be more expensive than running the detailed model in the first place. Also, architects must often rely on intuition to choose these simpler models, further complicating the problem.In this work, we present a method of bridging the gap between simple and detailed simulation by monitoring simulation behavior online and automatically swapping out detailed models with simpler statistical approximations. We demonstrate the potential of our methodology by implementing it in the opensource simulator SVE-Cachesim to swap out the level one data cache (L1D) within a memory hierarchy. This proof of concept demonstrates that our technique can handle a non-trivial usecase in not just approximation of local time-invariant statistics, but also those that vary with time (e.g. the L1D is a form of a time-series function), and downstream side-effects (e.g. the L1D filters accesses for the level two cache). Our simulation swaps out the built-in cache model with only an 8% error in the simulated cycle count while using the approximated cache models for over 90% of the simulation, and our simpler models require two to eight times less computation per "execution" of the model.

show abstract

Section: A Methodology 1) Simulatormentioning

confidence: 99%

Online Model Swapping in Architectural Simulation

Lavin¹,

Young²,

Vuduc³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…For the SVE analysis, the Arm Instruction Emulator (ArmIE) is used as this tool can execute unsupported instructions on AArch64 architecture by converting them into native ones. Dynamic binary instrumentation (through DynamoRIO) is used to extract additional metrics such as the memory traces( [9]).…”

Section: A Setupmentioning

confidence: 99%

On the Usage of the Arm C Language Extensions for a High-Order Finite-Element Kernel

Jubertie

Quintin

Dupros

2020

2020 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

Physics-based three-dimensional numerical simulations are becoming more predictive and have already become essential. As an example in geophysics, simulations at scale with a very fine resolution, including uncertainty quantification procedures are crucial to provide the relevant physical parameters for forward modeling of seismic wave propagation. Consequently, the diversity of HPC architectures available (heterogeneity, high core counts or depth of the memory hierarchy) leads to increasing concerns regarding the portability of applicative performances. In this paper, we discuss the implementation of the classical spectral finite-elements method using the Arm C Language Extensions for both Neon and SVE SIMD units.

show abstract

The Effects of Wide Vector Operations on Processor Caches

Poenaru

McIntosh–Smith

2020

2020 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

Asvie: A Timing-Agnostic SVE Optimization Methodology

Cited by 4 publications

References 9 publications

Online Model Swapping in Architectural Simulation

Online Model Swapping in Architectural Simulation

On the Usage of the Arm C Language Extensions for a High-Order Finite-Element Kernel

The Effects of Wide Vector Operations on Processor Caches

Contact Info

Product

Resources

About