Fu Lam Florian Diep

dblp:352/3263 · DBLP profile ↗
← Back
1ranked-venue papers
0as first author
1since 2021 · last 2023
0000-0002-5516-5502ORCID · reported

Domains — the database's venue-derived domains; a paper can count in several

Systems, architecture and hardware · 1 · 1 since 2021

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Computer architecture, parallel and distributed computing, and storage systems
1 paper
Electronic design automation · 77% GPUs and heterogeneous computing · 12% Hardware accelerators and domain-specific architectures · 12%

Topics — the 4 heaviest of 4, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Electronic design automation
circuit simulation
0.712023
Massively Parallel Circuit Setup in GPU-SPICE · IEEE Trans. Computers 2023
Electronic design automation › circuit simulation › analog circuit simulation
SPICE simulation
0.712023
Massively Parallel Circuit Setup in GPU-SPICE · IEEE Trans. Computers 2023
Hardware accelerators and domain-specific architectures › scientific computing accelerator
circuit simulation acceleration
0.212023
Massively Parallel Circuit Setup in GPU-SPICE · IEEE Trans. Computers 2023
GPUs and heterogeneous computing
GPU computing
0.212023
Massively Parallel Circuit Setup in GPU-SPICE · IEEE Trans. Computers 2023

Methods — techniques the papers use, named apart from their topics

device linearization · 0.7LU factorization · 0.7
YearPublicationVenuePosition
2023 Massively Parallel Circuit Setup in GPU-SPICE
abstract
SPICE simulations are the industry standard to analyze circuits for decades. However, they are computationally complex as each circuit is simulated at the transistor-level where individual transistor is modeled with dozens of sophisticated equations. This limits the practicality of SPICE simulations to relatively small circuits. However, this is in a direct conflict with the ever-increasing demands of circuit designers in which SPICE simulations for large circuits (e.g., DSPs, AES, etc.) at full accuracy are inevitably required to fulfill new industrial standards like automotive safety ISO 26262 with tool confidence level 1. To accelerate SPICE simulation without sacrificing accuracy, state-of-the-art approaches have started to employ GPUs to parallelize the LU-factorization and device linearization phases. Instead of focusing on these phases, this article demonstrates for the first time that when large circuits come into play, a new and equally important performance bottleneck emerges at the circuit setup phase. Speeding up the circuit setup phase in SPICE is our key focus in this paper. Our two implementations demonstrate that our GPU-based circuit setup reduces the analysis time from 4.5 days to merely 89 seconds for a 256-bit multiplier, which consists of more than 1M transistors. Our achieved speedup is 4396x compared to the baseline (open-source NGSPICE) and more than 2x compared to commercial (HSPICE and Spectre) SPICE circuit setup.
Victor M. van Santen, Fu Lam Florian Diep, Jörg Henkel, Hussam Amrouch
IEEE Trans. Computers2