性能

Mixture-of-Experts (MoE) has become a practical architecture for scaling LLM capacity while keeping per-token compute modest, but deploying MoE models on a single, memory-limited GPU remains difficult because expert weights dominate the HBM…

WebAssembly (Wasm) has become a key compilation target for portable and efficient execution across diverse platforms. Benchmarking its performance, however, is a multi-dimensional challenge: it depends not only on the choice of runtime…

The increasing demand for generative AI as Large Language Models (LLMs) services has driven the need for specialized hardware architectures that optimize computational efficiency and energy consumption. This paper evaluates the performance…

As computer systems continue to diversify across technologies, architectures, applications, and beyond, the relevant design space has become larger and more complex. Given such trends, design space exploration (DSE) at early stages is…

We introduce WritePolicyBench, a benchmark for evaluating memory write policies: decision rules that choose what to store, merge, and evict under a strict byte budget while processing a stream with document/API drift. The benchmark provides…

This work explores the use of the AMD Xilinx Versal Adaptable Intelligent Engine (AIE) to accelerate Gated Recurrent Unit (GRU) inference for latency constrained applications. We present a custom workload distribution framework across the…

We study the Multiserver-Job Queuing Model (MJQM) with general independent arrivals and service times under FCFS scheduling, using stochastic recurrence equations (SREs) and ergodic theory. We prove the monotonicity and separability…

Markov-modulated fluid queues (MMFQs) are a powerful modeling framework for analyzing the performance of computer and communication systems. Their distinguishing feature is that the underlying Markov process evolves on a continuous state…

This paper presents a systematic theoretical performance analysis of the Real-Valued root-MUSIC (RV-root-MUSIC) algorithm under non-asymptotic conditions. A well-known limitation of RV-root-MUSIC is the estimation ambiguity caused by mirror…

High-performance attention kernels are essential for Large Language Models. This paper presents analysis of CuTile-based Flash Attention memory behavior and a technique to improve its cache performance. In particular, our analysis on the…

Zero-knowledge proofs (ZKPs) are the cornerstone of programmable cryptography. They enable (1) privacy-preserving and verifiable computation across blockchains, and (2) an expanding range of off-chain applications such as credential…

This paper highlights first steps towards enabling graphics processing unit (GPU) acceleration of the task-parallel smoothed particle hydrodynamics (SPH) solver SWIFT. Novel combinations of algorithms are presented, enabling SWIFT to…

We study d-way balanced allocation, which assigns each incoming job to the lightest loaded among d randomly chosen servers. While prior work has extensively studied the performance of the basic scheme, there has been less published work on…

This paper presents our experience to understand latency variance caused by kernel and hardware events, which are often invisible at the application level. For this purpose, we have built VarMRI, a tool chain to monitor and analyze those…

Emergency department (ED) overcrowding and patient boarding represent critical systemic challenges that compromise care quality. We propose a threshold-based admission policy that redirects non-urgent patients to alternative care pathways,…

In practice, the cost of delaying a job can grow as the job waits. Such behavior is modeled by the Time-Varying Holding Cost (TVHC) problem, where each job's instantaneous holding cost increases with its current age (a job's age is the time…

Different from traditional Large Language Model (LLM) serving that colocates the prefill and decode stages on the same GPU, disaggregated serving dedicates distinct GPUs to prefill and decode workload. Once the prefill GPU completes its…

The rapid growth of AI has fueled the expansion of accelerator- or GPU-based data centers. However, the rising operational energy consumption has emerged as a critical bottleneck and a major sustainability concern. Dynamic Voltage and…

Edge computing provides a cloud-like architecture where small-scale resources are distributed near the network edge, enabling applications on resource-constrained devices to offload latency-critical computations to these resources. While…

Modern computing workloads are often composed of parallelizable jobs. A parallelizable job can be completed more quickly when run on additional servers. However, each job can only use a limited number of servers, known as its…