Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2507.20063

Racing to Idle: Energy Efficiency of Matrix Multiplication on Heterogeneous CPU and GPU Architectures

Mufakir Qamar Ansari·Mudabir Qamar Ansari
View original source ↗Read source document ↗Export metadata ↓

Abstract

The paradigm shift towards multi-core and heterogeneous computing, driven by the fundamental power and thermal limits of single-core processors, has established energy efficiency as a first-class design constraint in high-performance computing (HPC). Heterogeneous systems, integrating traditional multi-core CPUs with specialized accelerators like discrete (dGPU) and integrated (iGPU) graphics processing units, offer a compelling path to navigating the trade-offs between performance and power. However, quantifying these trade-offs on widely accessible hardware remains a critical area of study. This paper presents a direct, empirical measurement of the performance and energy-to-solution of a canonical HPC workload -- a 4096x4096 matrix-matrix multiplication -- on three distinct compute architectures within a single consumer-grade laptop: a multi-core AMD Ryzen 7 5800H CPU, a discrete NVIDIA GeForce GTX 1650 GPU, and an integrated AMD Radeon Vega GPU. Using standard, validated, and minimally intrusive tools such as Linux perf and nvidia-smi, we find that the discrete GPU is not only the performance leader, achieving a 93.5x speedup over the CPU, but is also the most energy-efficient, consuming only 2% of the energy used by the CPU, resulting in a 50-fold improvement in energy efficiency. These findings provide a practical demonstration of the "race to idle" principle and offer clear, quantitative guidance on architectural choices for energy-aware software development.

Explore related subjects

cs.DCcs.CC

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Mufakir Qamar Ansari, Mudabir Qamar Ansari. 2025-07-26. Racing to Idle: Energy Efficiency of Matrix Multiplication on Heterogeneous CPU and GPU Architectures. https://arxiv.org/abs/2507.20063

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Online Treasure Hunt in Vertex-Permuted Dynamic Rings

We study the problem of treasure hunt by a group of $k \geq 1$ agents in vertex-permuted dynamic rings (VP). In this model, the $n$ vertices remain on a ring but are permuted at each time step. We first show that treasure hunt is impossible for any $k \leq n-3$ agents, if there are no restrictions on the sequence of permutations used in the dynamic ring. We then study the $VP(\delta)$ setting, in which for every pair $i, j$ of vertices, the edge $(i, j)$ is guaranteed to appear within $\delta$ steps. We show that the class $VP(\delta)$ is feasible only for $\delta \geq \left\lceil \frac{n-1}{2}\right\rceil$. For the one-agent case, we show a tight bound of $\Theta(\delta n)$ on the worst-case search time as well as competitive ratio of any online algorithm for treasure hunt, provided $\delta \geq 2n$. We then give an optimal algorithm for $k$ agents, thereby showing that $k$ agents can obtain a speedup of $k$ on the worst-case search time. Finally, in the R-VP setting, in which in every step, the vertices are arranged as a ring according to a random permutation, we show that treasure hunt takes expected $\Theta(n)$ steps against an oblivious adversary and $\Theta(n \log n)$ steps against an adaptive adversary.

cs.DC↗
arXiv2026

The Computing Channel: How Modulation Programs the Airwaves

Distributed computing and distributed artificial intelligence require frequent exchanges of intermediate results, although many applications need only an aggregate rather than messages from individual devices. Conventional systems recover each message before computing the aggregate, whereas over-the-air computation (OAC) exploits simultaneous transmission to obtain it directly. However, dominant OAC implementations rely on analog signaling, creating a mismatch with finite-precision data and digital communication procedures. This article presents digital function-oriented communication, in which finite-alphabet symbol representations and receiver decisions are jointly designed so that multiple-access superposition encodes the desired function without recovering individual inputs. We introduce its computational-constellation principle, main design approaches, extensions, and implementation challenges. Federated edge learning illustrates how the framework can reduce user-dependent data-bearing resources while operating directly on quantized model updates.

cs.DC↗
arXiv2026

Can AI Remediate Backend Failures Safely? GuardedAct with Blast-Radius-Aware Sandboxing

Large Language Models (LLMs) have shown promising capabilities in generating remediation actions for microservice failures. However, directly executing AI-generated repair actions in production risks cascading collateral damage. We propose GuardedAct, a sandbox-first remediation framework that interposes a blast-radius-aware verification layer between the LLM action generator and the production environment. GuardedAct operates in four phases: (1) ingesting a diagnosis report together with the live system topology and recent telemetry, (2) prompting an LLM to produce a ranked list of candidate remediation actions, (3) simulating each action in a lightweight digital-twin sandbox that estimates the blast radius and assigns a risk label, and (4) enforcing a rollback-confidence gate that auto-executes only low-risk actions while escalating high-risk ones for human review. We evaluate GuardedAct on five fault scenarios injected into the DeathStarBench social-network application. Experimental results show that GuardedAct achieves an overall recovery rate of 87.4% while reducing collateral damage by 79.7% relative to direct LLM execution (from 25.6% to 5.2%), at the cost of a modest sandbox-induced increase in mean time to recovery (approximately 8 s). Ablation studies confirm that each component contributes meaningfully to the safety-speed trade-off.

cs.DC↗