Searcharxiv⌕ Search

arXiv subjects

Peihao Huang

Publications and source records attributed to Peihao Huang.

At least 19 recordsLinked to original sources

MTRDMA: Lightweight User-Space Control-Plane Isolation for Multi-Tenant RDMA

In managed multi-tenant clouds, RDMA's lack of control-plane (CP) resource isolation lets a noisy tenant exhaust shared NIC microarchitectural caches, degrading victim throughput by up to 96.4\%. Data-path schedulers leave CP resources unprotected, and hardware approaches require FPGAs or SmartNICs rather than the commodity RNICs operators already deploy. We present MTRDMA, a lightweight user-space CP isolation mechanism for commodity RNICs, applicable to managed runtimes such as containers and sandboxed processes. MTRDMA intercepts library-level CP operations to enforce hierarchical quotas while preserving the zero-copy data path, using atomic shared-memory quota state with per-process cached admission (${\sim}$0.13\,$μ$s), a hybrid garbage collector that reclaims orphaned quota within seconds, and rdma-cgroup caps with ledger--kernel reconciliation to bound and detect residual tampering. On ConnectX-6 RNICs, MTRDMA cuts victim degradation under Queue Pair (QP) cache exhaustion from 94.5\% to 0.08\%, extends protection to Memory Region (MR) cache exhaustion, and scales to 256 concurrent tenants with negligible data-plane overhead.

cs.NI↗

Electric-dipole-induced nuclear electric resonance in silicon

The nuclear spin of a phosphorus atom in silicon is considered as a promising quantum information carrier for quantum computing due to its long coherence time. Coherent control of the nuclear spin is typically realized using nuclear magnetic resonance (NMR). In this work, we theoretically propose that with the aid of hyperfine interactions and magnetic field gradients, fast electric-dipole-induced nuclear electric resonance (EDNER) can be achieved. The stronger EDNER (compared with NMR) arises from the combined effects of hyperfine interaction between the nuclear and electron spins and coupling between the electron spin and charge degrees of freedom mediated by magnetic field gradients in this scheme. By operating at specific working points, the nuclear-spin splitting becomes less sensitive to electrical noise, enabling long coherence times and high-fidelity quantum operations. All elctrical, addressable control of nuclear spins paves the way for scalable silicon-based quantum computing.

cond-mat.mes-hall↗

Micromagnet engineering for enhanced addressability of donor-cluster arrays in silicon

Phosphorus donors in silicon are one of the promising platforms for fault-tolerant quantum computing, owing to their long coherence times and compatibility with complementary metal-oxide-semiconductor (CMOS) technology. However, a key bottleneck for scaling this platform lies in the independent addressability of individual spin qubits. Here, we present a scalable quantum processor architecture that integrates scanning tunneling microscope (STM)-defined phosphorus donor-cluster arrays with on-chip micromagnets. By generating magnetic field gradients, the micromagnets enable a broader addressable frequency range compared to the addressing scheme based solely on variations in hyperfine (HF) interactions, which arise from the randomness of intra-cluster configurations. Through micromagnetic simulations to optimize micromagnet geometry, we demonstrate that these structures are compatible with customized array designs, facilitating the integration of over 120 atomic spin qubits and enabling frequency multiplexing with a single microwave antenna, which greatly reduces the overhead of fan-out addressing lines. This design is further validated via magnetic measurements, which generally align with our micromagnetic simulations. These results establish a critical foundation for the development of scalable silicon quantum computing with atomic qubits.

cond-mat.mes-hall↗

Cross-Model Autoscaling for Shared LLM Serving

Multi-model LLM serving is moving toward shared MaaS clusters, where co-hosted models compete for a fixed GPU budget while each model experiences time-varying demand and must satisfy its own latency SLO. Existing LLM autoscalers remain largely model-local: their signals expose local runtime activity or delayed latency outcomes, and their scale-up or provisioning decisions do not directly determine how shared capacity should be allocated across competing models. We present the Token-service-share Rebalancing Engine (TRE), a control-plane framework for hot-switched multi-model LLM serving. TRE introduces Token Service Share (TSS), a calibrated, demand-normalized signal that estimates effective token service per active or queued request and yields a comparable health score across heterogeneous models and SLO classes. Guided by TSS, TRE coordinates bounded receiver--donor capacity movement under a fixed GPU budget: it separates fast rescue from slower rebalancing and incrementally reallocates active replicas toward models with the largest calibrated service deficits. We implement TRE on a Kubernetes-based hot-switch serving stack without modifying the inference scheduler. Across seven LLM serving traces, TRE reduces P95 end-to-end latency by 11.9--79.0\% and P99 latency by 12.5--72.6\% compared with a state-of-the-art KV-cache-based reactive autoscaler running on the same hot-switch runtime. The gains hold on both targeted stress probes and production-derived conversation/code traces, where TRE reduces P95/P99 latency by 50.8/63.7\% and 79.0/72.6\%, respectively. These results show that effective hot-switched autoscaling requires not only fast replica actuation, but also calibrated service-deficit signals and coordinated cross-model capacity arbitration. Our code and artifacts are available at https://github.com/zxzx9898/Token-service-share_Rebalancing_Engine.

cs.DC↗

Breaking the Duplex Barrier: Lane-Granularity OCS Scheduling for LLM Training

Optical circuit switch (OCS) can reconfigure physical connectivity to match the predictable communication schedules of large language model (LLM) training. Although each OCS light path is physically simplex, existing demand-aware OCS schedulers allocate capacity in duplex-port pairs, forcing equal bandwidth in both directions and stranding capacity under asymmetric node-pair traffic. This paper present LACE, the first offline OCS schedule compiler that independently allocates transmit (TX) and receive (RX) lanes for LLM training. Without changing the selected collective algorithms, operation order, or rank placement, LACE reconstructs directed node-level demand, jointly determines which consecutive operations share a configuration and how many simplex circuits serve each direction, and realizes these allocations as physical lane bindings and optical paths under per-node lane-inventory and multi-OCS fabric constraints. Software acknowledgments carry feedback over independently provisioned return paths, while coordinated link configuration and recovery verify each configuration before communication resumes. On a separate three-server testbed using fixed topologies and matched per-port rate limits, LACE's asymmetric connectivity achieves $1.80\times$ speedup for communication replay and $1.27\times$ for GPT-2 training over a symmetric-topology baseline. At larger scale, simulations of LLaMA-3.1 70B and 405B schedules with sixteen 400-Gb/s ports per server show that LACE achieves $1.21$--$2.04\times$ communication speedup over the latest duplex OCS scheduler.

cs.NI↗

Scalable Spin Qubit Architecture with Donor-Cluster Arrays in Silicon

Spin qubits in silicon donors offer a promising platform for quantum computing due to their long coherence times and semiconductor compatibility. However, scaling donor-based spin qubits in silicon is fundamentally challenged by frequency crowding, crosstalk, and the tight tolerances on donor placement in conventional single-donor architectures.To overcome this, we introduce a paradigm based on a two-dimensional array of phosphorus-donor clusters, in which multiple donors share a bound electron. The natural hyperfine distribution within each cluster enables individual addressability of the electron and nuclear spins, while tunable exchange interactions between clusters mediate local all-to-all connectivity. We present a universal control protocol achieving gate fidelities exceeding 99% for both intra-cluster and inter-cluster multi-qubit operations, with crosstalk effectively suppressed. The architecture natively supports efficient quantum error correction, including bias-tailored codes that exploit the intrinsic noise bias of spin qubits. Furthermore, its modular design is compatible with long-range coupling via electron shuttling for large-scale integration. This donor-cluster array architecture establishes a robust and hardware-efficient pathway towards scalable, fault-tolerant quantum computing in silicon.

quant-ph↗

MultiPath Memory Access: Breaking Host-GPU Bandwidth Bottlenecks in LLM Services

Host-GPU data movement has become a latency-critical bottleneck in LLM serving, surfacing in common paths such as model-weight movement and KV cache offload/fetch. Today, each host-GPU copy is effectively confined to the PCIe path of the target GPU, even though modern multi-GPU servers contain additional PCIe links on peer GPUs and high bandwidth GPU interconnects. This leaves substantial intra-server I/O capacity unused. To address this issue, we present Multipath Memory Access (MMA), a software-defined multipath memory access system for host--GPU data transfer. To the best of our knowledge, MMA is the first software-defined system to enable efficient multipath host--GPU data transfer within a single multi-GPU server. MMA expands a single host--GPU copy across available direct and relay paths without hardware, driver, or application changes. It preserves CUDA stream semantics with a dependency-preserving Dummy Task, coordinates distributed micro-transfer completion through a lightweight synchronization mechanism, and uses queue backpressure to route traffic without explicit link-state feedback. On an 8-GPU NVIDIA H20 server, MMA achieves 245 GB/s peak host-to-GPU bandwidth, a 4.62x improvement over native CUDA copies, and reduces TTFT for KV cache fetching by 1.14-2.38x and model wake-up/switching latency by 1.12-2.48x.

cs.DC↗

Demonstration of quantum error detection in a silicon quantum processor

Quantum error detection is essential in realizing large-scale universal quantum computation, especially for quantum error correction (QEC). However, key elements for FTQC have yet to be realized in silicon qubits. Here, we demonstrate quantum error detection on a donor-based silicon quantum processor comprising four-nuclear spin qubits and one electron spin as an auxiliary qubit. The entanglement capability of this system is validated through the establishment of two-qubit Bell state entanglement between the nuclear spins and the generation of a four-qubit Greenberger-Horne-Zeilinger (GHZ) state, achieving a GHZ state fidelity of 88.5(2.3)%. Furthermore, by executing a four-qubit error detection circuit with the stabilizers, we successfully detect arbitrary single-qubit errors. The encoded Bell state entanglement information is recovered by performing the Pauli-frame update (PFU) via postprocessing. Based on the detected errors, we identify strongly biased noise in our system. Our results mark a significant advance toward FTQC in silicon spin qubits.

quant-ph↗

Two-electron edge states in a double SSH-chain of quantum dots

We study an interacting two-body model with adjustable spin tunneling in the context of the double SSH chains for a quantum dot system. We discovered that varying interaction strengths and spin tunneling significantly influence the properties of correlated edge states in the energy spectrum obtained through exact diagonalization. We observe that stronger interactions lead to longer decay lengths of these states. Conversely, the decay length decreases as the difference in intracell or intercell tunneling increases. Importantly, the decay length is strongly correlated with the dynamical behavior of two particles; specifically, an increase in decay length corresponds to a decrease in motion frequency. This conclusion is supported by the observation of the expectation value of coordinate operators of the particles.

cond-mat.mes-hall↗

Scalable Robust Quantum Control for Semiconductor Spin Qubits with Always-on Couplings

We demonstrate a robust quantum control framework that enables high-fidelity gate operations in semiconductor spin qubit systems with always-on couplings. Always-on interactions between qubits pose a fundamental challenge for quantum processors by inducing correlated errors that can trigger chaotic dynamics. Our approach suppresses both static coupling noise and time-dependent crosstalk without requiring high on/off ratio tunable couplers. Significantly, these pulses also prevent the emergence of chaotic entanglement growth in deep quantum circuits, preserving coherence in large multi-qubit systems. By relaxing hardware constraints on coupling control, our method provides a practical path toward scaling semiconductor quantum processors within existing fabrication capabilities, with particular relevance for silicon spin qubit architectures where high-contrast coupling modulation remains challenging.

quant-ph↗

Quantum squeezing amplification with a weak Kerr nonlinear oscillator

Quantum squeezed states, with reduced quantum noise, have been widely utilized in quantum sensing and quantum error correction applications. However, generating and manipulating these nonclassical states with a large squeezing degree typically requires strong nonlinearity, which inevitably induces additional decoherence that diminishes the overall performance. Here, we demonstrate the generation and amplification of squeezed states in a superconducting microwave cavity with weak Kerr nonlinearity. By subtly engineering an off-resonant microwave drive, we observe cyclic dynamics of the quantum squeezing evolution for various Fock states |N> with N up to 6 in displaced frame of the cavity. Furthermore, we deterministically realize quantum squeezing amplification by alternately displacing the Kerr oscillator using the Trotterization technique, achieving a maximum squeezing degree of 14.6 dB and squeezing rate of 0.28 MHz. Our hardware-efficient displacement-enhanced squeezing operations provide an alternative pathway for generating large squeezed states, promising potential applications in quantum-enhanced sensing and quantum information processing.

quant-ph↗

An Addressable and Tunable Module for Donor-based Scalable Silicon Quantum Computing

Donor-based spin qubit offers a promising silicon quantum computing route for building large-scale qubit arrays, attributed to its long coherence time and advancements in nanoscale donor placement. However, the state-of-the-art device designs face scalability challenges, notably in achieving tunable two-qubit coupling and ensuring qubit addressability. Here, we propose a surface-code-compatible architecture, where each module has both tunable two-qubit gates and addressable single-qubit gates by introducing only a single extra donor in a pair of donors. We found that to compromise between the requirement of tunability and that of addressability, an asymmetric scheme is necessary. In this scheme, the introduced extra donor is strongly tunnel-coupled to one of the donor spin qubits for addressable single-qubit operation, while being more weakly coupled to the other to ensure the turning on and off of the two-qubit operation. The fidelity of single-qubit and two-qubit gates can exceed the fault-tolerant threshold in our design. Additionally, the asymmetric scheme effectively mitigates valley oscillations, allowing for engineering precision tolerances up to a few nanometers. Thus, our proposed scheme presents a promising prototype for large-scale, fault-tolerant, donor-based spin quantum processors.

cond-mat.mes-hall↗

Universal control of superexchange in linear triple quantum dots with an empty mediator

Superexchange is one of the vital resources to realize long-range interaction between distant spins for large-scale quantum computing. Recent experiments have demonstrated coherent oscillations between logical states defined by remote spins whose coupling is given by the superexchange interaction mediated by central spins. Excavating the potential of superexchange requires a full understanding of the interaction in terms of control parameters, which is still lacking in literature. Here, using full configuration interaction calculations, we study a two-electron system in a linear triple-quantum-dot device in which the left and right dots are occupied by a single electron each, whose spin states are defined as qubits. The numerical nature of the full configuration interaction calculations allows access to the microscopic details of the quantum-dot confining potential and electronic wavefunctions, some of which are overlooked in the celebrated Hubbard model but turn out to be critical for the behavior of superexchange. We focus on the detuning regime where the charge ground state yields an empty middle dot. We have found that, when the detunings at the left and right dots are leveled, the superexchange can exhibit a non-monotonic behavior which ranges from positive to negative values as a function of the middle-dot detuning. We further show that a larger relative detuning between the left and right dots causes the magnitude of the superexchange to increase (decrease) for an originally positive (negative) superexchange. We then proceed to show the results for a much larger left-right dot detuning. Our results suggest that even a simple configuration of delocalized two-electron states in a linear triple-quantum-dot device exhibits superexchange energy with non-trivial behaviors, which could have important applications in spin-based quantum computing.

cond-mat.mes-hall↗

Ultrafast and Electrically Tunable Rabi Frequency in a Germanium Hut Wire Hole Spin Qubit

Hole spin qubits based on germanium (Ge) have strong tunable spin orbit interaction (SOI) and ultrafast qubit operation speed. Here we report that the Rabi frequency (f_Rabi) of a hole spin qubit in a Ge hut wire (HW) double quantum dot (DQD) is electrically tuned through the detuning energy and middle gate voltage (V_M). f_Rabi gradually decreases with increasing detuning energy; on the contrary, f_Rabi is positively correlated with V_M. We attribute our results to the change of electric field on SOI and the contribution of the excited state in quantum dots to f_Rabi. We further demonstrate an ultrafast f_Rabi exceeding 1.2 GHz, which evidences the strong SOI in our device. The discovery of an ultrafast and electrically tunable f_Rabi in a hole spin qubit has potential applications in semiconductor quantum computing.

cond-mat.mes-hall↗

Noisy intermediate-scale quantum computers

Quantum computers have made extraordinary progress over the past decade, and significant milestones have been achieved along the path of pursuing universal fault-tolerant quantum computers. Quantum advantage, the tipping point heralding the quantum era, has been accomplished along with several waves of breakthroughs. Quantum hardware has become more integrated and architectural compared to its toddler days. The controlling precision of various physical systems is pushed beyond the fault-tolerant threshold. Meanwhile, quantum computation research has established a new norm by embracing industrialization and commercialization. The joint power of governments, private investors, and tech companies has significantly shaped a new vibrant environment that accelerates the development of this field, now at the beginning of the noisy intermediate-scale quantum era. Here, we first discuss the progress achieved in the field of quantum computation by reviewing the most important algorithms and advances in the most promising technical routes, and then summarizing the next-stage challenges. Furthermore, we illustrate our confidence that solid foundations have been built for the fault-tolerant quantum computer and our optimism that the emergence of quantum killer applications essential for human society shall happen in the future.

quant-ph↗

Pressured-induced superconductivity extending across the topological phase transition in thallium-based topological materials TlBi(S1-xSex)2

The coexistence of superconductivity and topology holds the potential to realize exotic quantum states of matter. Here we report that superconductivity induced by high pressure in three thallium-based materials, covering the phase transition from a normal insulator (TlBiS2) to a topological insulator (TlBiSe2) through a Dirac semimetal (TlBiSeS). By increasing the pressure up to 60 GPa, we observe superconductivity phase diagrams with maximal Tc values at 6.0-8.1 K. Our density-functional theory calculations reveal topological surface states in superconductivity phases for all three compounds. Our study paves the path to explore topological superconductivity and topological phase transitions.

cond-mat.supr-con↗

Acceptor-based qubit in silicon with tunable strain

Long coherence time and compatibility with semiconductor fabrication make spin qubits in silicon an attractive platform for quantum computing. In recent years, hole spin qubits are being developed as they have the advantages of weak coupling to nuclear spin noise and strong spin-orbit coupling (SOC), in constructing high-fidelity quantum gates. However, there are relatively few studies on the hole spin qubits in a single acceptor, which requires only low density of the metallic gates. In particular, the investigation of flexible tunability using controllable strain for fault-tolerant quantum gates of acceptor-based qubits is still lacking. Here, we study the tunability of electric dipole spin resonance (EDSR) of acceptor-based hole spin qubits with controllable strain. The flexible tunability of LH-HH splitting and spin-hole coupling (SHC) with the two kinds of strain can avoid high electric field at the "sweet spot", and the operation performance of the acceptor qubits could be optimized. Longer relaxation time or stronger EDSR coupling at low electric field can be obtained. Moreover, with asymmetric strain, two "sweet spots" are induced and may merge together, and form a second-order "sweet spot". As a result, the quality factor $Q$ can reach $10^{4}$ for single-qubit operation, with high tolerance for the electric field variation. Furthermore, the two-qubit operation of acceptor qubits based on dipole-dipole interaction is discussed for high-fidelity two-qubit gates. The tunability of spin qubit properties in acceptor via strain could provide promising routes for spin-based quantum computing.

cond-mat.mes-hall↗

Sampling Is All You Need on Modeling Long-Term User Behaviors for CTR Prediction

Rich user behavior data has been proven to be of great value for Click-Through Rate (CTR) prediction applications, especially in industrial recommender, search, or advertising systems. However, it's non-trivial for real-world systems to make full use of long-term user behaviors due to the strict requirements of online serving time. Most previous works adopt the retrieval-based strategy, where a small number of user behaviors are retrieved first for subsequent attention. However, the retrieval-based methods are sub-optimal and would cause more or less information losses, and it's difficult to balance the effectiveness and efficiency of the retrieval algorithm. In this paper, we propose SDIM (Sampling-based Deep Interest Modeling), a simple yet effective sampling-based end-to-end approach for modeling long-term user behaviors. We sample from multiple hash functions to generate hash signatures of the candidate item and each item in the user behavior sequence, and obtain the user interest by directly gathering behavior items associated with the candidate item with the same hash signature. We show theoretically and experimentally that the proposed method performs on par with standard attention-based models on modeling long-term user behaviors, while being sizable times faster. We also introduce the deployment of SDIM in our system. Specifically, we decouple the behavior sequence hashing, which is the most time-consuming part, from the CTR model by designing a separate module named BSE (behavior Sequence Encoding). BSE is latency-free for the CTR server, enabling us to model extremely long user behaviors. Both offline and online experiments are conducted to demonstrate the effectiveness of SDIM. SDIM now has been deployed online in the search system of Meituan APP.

cs.IR↗