arXiv subjects

Qingkai Meng

Publications and source records attributed to Qingkai Meng.

7 recordsLinked to original sources

arXiv2025

STAR: Decode-Phase Rescheduling for LLM Inference

Large Language Model (LLM) inference has emerged as a fundamental paradigm, however, variations in output length cause severe workload imbalance in the decode phase, particularly for long-output reasoning tasks. Existing systems, such as PD disaggregation architectures, rely on static prefill-to-decode scheduling, which often results in SLO violations and OOM failures under evolving decode workloads. In this paper, we propose STAR, a decode rescheduling system powered by length prediction to anticipate future workloads. Our core contributions include: (1) A lightweight and continuous LLM-native prediction method that leverages LLM hidden state to model remaining generation length with high precision (reducing MAE by 49.42%) and low overhead (cutting predictor parameters by 93.28%); (2) A rescheduling solution in decode phase with a dynamic balancing mechanism that integrates current and predicted workloads, reducing P99 TPOT by 75.1% and achieving 2.63 times higher goodput.

cs.DC↗

arXiv2025

Incremental Policy Iteration for Unknown Nonlinear Systems with Stability and Performance Guarantees

This paper proposes a general incremental policy iteration adaptive dynamic programming (ADP) algorithm for model-free robust optimal control of unknown nonlinear systems. The approach integrates recursive least squares estimation with linear ADP principles, which greatly simplifies the implementation while preserving adaptive learning capabilities. In particular, we develop a sufficient condition for selecting a discount factor such that it allows learning the optimal policy starting with an initial policy that is not necessarily stabilizing. Moreover, we characterize the robust stability of the closed-loop system and the near-optimality of iterative policies. Finally, we perform numerical simulations to demonstrate the effectiveness of the proposed method.

math.OC↗

arXiv2024

Link-identified Routing Architecture in Space

Low earth orbit (LEO) satellite networks have the potential to provide low-latency communication with global coverage. To unleash this potential, it is crucial to achieve efficient packet delivery. In this paper, we propose a Link-identified Routing (LiR) architecture for LEO satellite networks. The LiR architecture leverages the deterministic neighbor relation of LEO constellations, and identifies each inter-satellite link (ISL). Moreover, LiR architecture adopts source-route-style forwarding based on in-packet bloom filter (BF). Each satellite could efficiently encode multiple ISL identifiers via an in-packet BF to specify the end-to-end path for the packets. Due to false positives caused by BF, the more ISLs are encoded at a time, the more redundant forwarding cases emerge. Based on the topology characteristics, we derive the expected forwarding overhead in a closed-form and propose the optimal encoding policy. To accommodate link-state changes in LEO satellite networks, we propose the on-demand rerouting scheme and the on-demand detouring scheme to address the intermittent ISLs. We also elaborate how to take advantage of LiR architecture to achieve seamless handover for ground-satellite links (GSLs). Finally, we conduct extensive numerical experiments and packet-level simulations to verify our analytical results and evaluate the performance of the LiR architecture.

cs.NI↗

arXiv2024

Thermodynamic origin of the phonon Hall effect in a honeycomb antiferromagnet

The underlying mechanism of the thermal Hall effect (THE) generated by phonons in a variety of insulators is yet to be identified. Here, we report on a sizeable thermal Hall conductivity in NiPS$_3$, a van der Waals stack of honeycomb layers with a zigzag antiferromagnetic order below $T_N$ = 155 K. The longitudinal ($\kappa_{aa}$) and the transverse ($\kappa_{ab}$) thermal conductivities peak at the same temperature and the thermal Hall angle, at this peak, respects a previously identified bound. The amplitude of $\kappa_{ab}$ is extremely sensitive to the amplitude of magnetization along the $b$-axis, in contrast to the phonon mean free path, which is not at all. We show that the magnon and acoustic phonon bands cross each other along the $b^\ast$ orientation in the momentum space. The relevance of a thermodynamic property, combined with the irrelevance of the mean free path, points to an intrinsic origin.

cond-mat.str-el↗

arXiv2023

SFC: Near-Source Congestion Signaling and Flow Control

State-of-the-art congestion control algorithms for data centers alone do not cope well with transient congestion and high traffic bursts. To help with these, we revisit the concept of direct \emph{backward} feedback from switches and propose Back-to-Sender (BTS) signaling to many concurrent incast senders. Combining it with our novel approach to in-network caching, we achieve near-source sub-RTT congestion signaling. Source Flow Control (SFC) combines these two simple signaling mechanisms to instantly pause traffic sources, hence avoiding the head-of-line blocking problem of conventional hop-by-hop flow control. Our prototype system and scale simulations demonstrate that near-source signaling can significantly reduce the message completion time of various workloads in the presence of incast, complementing existing congestion control algorithms. Our results show that SFC can reduce the $99^{th}$-percentile flow completion times by $1.2-6\times$ and the peak switch buffer usage by $2-3\times$ compared to the recent incast solutions.

cs.NI↗

arXiv2023

Magnetostriction, piezomagnetism and domain nucleation in a kagome antiferromagnet

Whenever the elastic energy of a solid depends on magnetic field, there is a magnetostrictive response. Field-linear magnetostriction implies piezomagnetism and vice versa. Here, we show that Mn$_3$Sn, a non-collinear antiferromanget with Weyl nodes, hosts a large and almost perfectly linear magnetostriction even at room temperature. The longitudinal and transverse magnetostriction, with opposite signs and similar amplitude are restricted to the kagome planes and the out-of-plane response is negligibly small. By studying four different samples with different Mn:Sn ratios, we find a clear correlation between the linear magnetostriction, the spontaneous magnetization and the concentration of Sn vacancies. The recently reported piezomagnetic data fits in our picture. We show that linear magnetostriction and piezomagnetism are both driven by the field-induced in-plane twist of spins. A quantitative account of the experimental data requires the distortion of the spin texture by Sn vacancies. We find that the field-induced domain nucleation within the hysteresis loop corresponds to a phase transition. Within the hysteresis loop, a concomitant mesoscopic modulation of local strain and spin twist angles, leading to twisto-magnetic stripes, arises as a result of the competition between elastic and magnetic energies.

cond-mat.str-el↗

arXiv2021

The free energy of twisting spins in Mn$_3$Sn

The magnetic free energy is usually quadratic in magnetic field and depends on the mutual orientation of the magnetic field and the crystalline axes. Tiny in magnitude, this magnetocrystalline anisotropy energy (MAE) is nevertheless indispensable for the existence of permanent magnets. Here, we show that in Mn$_3$Sn, a non-collinear antiferromagnet attracting much attention following the discovery of its large anomalous Hall effect, the free energy of spins has superquadratic components, which drive the MAE. We experimentally demonstrate that the thermodynamic free energy includes terms odd in magnetic field ($\mathcal{O}(H^3)+\mathcal{O}(H^5)$) and generating sixfold and twelve-fold angular oscillations in the torque response. We show that they are quantitatively explained by theory, which can be used to quantify relevant energy scales (Heisenberg, Dzyaloshinskii-Moriya, Zeeman and single-ion anisotropy) of the system. Based on the theory, we conclude that, in contrast with common magnets, what drives the MAE in Mn$_3$Sn is the field-induced deformation of the spin texture.

cond-mat.str-el↗