arXiv subjects

Md Arafat Hossain

Publications and source records attributed to Md Arafat Hossain.

6 recordsLinked to original sources

arXiv2026

Accelerating Transfer-Learning-Based Autotuning with Predictive LLVM IR Performance Ranking

As the complexity of High Performance Computing (HPC) ecosys- tems continually increases, achieving optimal performance becomes a challenge. Traditional performance autotuning techniques pro- vide promising means to navigate this complexity, these techniques remain computationally intensive and require many evaluations to find optimal configurations. This work proposes an autotuning framework that designs a machine learning-based ensemble LLVM Intermediate Representa- tion (IR) ranker, Neural Configuration Scorer (NCS). NCS ranks the performance of IRs sampled by a transfer-learning-based autotuner, improving the efficiency of the tuning process by reducing tuning overheads and circumventing subpar evaluations. By leveraging knowledge from related tasks, we are able to effectively exploit the transfer relationship to access high-performing configurations in fewer samples than traditional techniques that rely upon itera- tive refinement. Our framework can achieve similar performance improvements as state-of-the-art autotuning techniques with up to 61.67% fewer evaluations, averaging 27.85% fewer evaluations across various HPC benchmarks.

cs.PF↗

arXiv2026

Generalizing Scaling Laws for Dense and Sparse Large Language Models

Despite recent advancements of large language models (LLMs), optimally predicting the model size for LLM pretraining or allocating optimal resources still remains a challenge. Several efforts have addressed the challenge by proposing different empirical scaling laws, but almost all of them are architecture-specific (dense or sparse). In this work we revisit existing empirical scaling laws and propose a generalized scaling law to provide a unified framework that is applicable to both dense and sparse large language models. We evaluate and compare our proposed scaling law with existing scaling laws and demonstrate that our proposed scaling law captures the scaling behavior of existing scaling laws. Further, we show an IsoFLOP comparison between our proposed scaling law and the state-of-the-art scaling law to illustrate the effectiveness of our proposed scaling law for Mixture-of-Expert (MoE)-based very large LLMs like DeepSeek-V3. Our proposed scaling law can be used to estimate the best model hyperparameters (Model size, Tokens and Compute) for a given sparsity or to identify the optimal sparsity for the given model hyperparameters.

cs.LG↗

arXiv2025

Mining Service Behavior for Stateful Service Emulation

Enterprise software systems are increasingly integrating with diverse services to meet expanding business demands. Testing these highly interconnected systems presents a challenge due to the need for access to the connected services. Service virtualization has emerged as a widely used technique to derive service models from recorded interactions, for service response generation during system testing. Various methods have been proposed to emulate actual service behavior based on these interactions, but most fail to account for the service's state, which reduces the accuracy of service emulation and the realism of the testing environment, especially when dealing with stateful services. This paper proposes an approach to deriving service models from service interactions, which enhance the accuracy of response generation by considering service state. This is achieved by uncovering contextual dependencies among interaction messages and analyzing the relationships between message data values. The approach is evaluated using interaction traces collected from both stateful and stateless services, and the results reveal notable enhancements in accuracy and efficiency over existing approaches in service response generation.

cs.SE↗

arXiv2017

Time-Resolved and Temperature Tuneable Measurements of Fluorescent Intensity using a Smartphone Fluorimeter

A smartphone fluorimeter capable of time-based fluorescence intensity measurements at various temperatures is reported. Excitation is provided by an integrated UV LED (370 nm) and detection obtained using the in-built CMOS camera. A Peltier is integrated to allow measurements of the intensity over T = 10 to 40 C with a maximum temperature resolution of DELTA T ~ 0.1 C. All components are controlled using a smartphone battery powered Arduino microcontroller and a customised Android application that allows sequential fluorescence imaging and quantification every DELTA t = 4 seconds. The temperature dependence of fluorescence intensity for four emitters (Rhodamine B, Rhodamine 6G, 5,10,15,20-tetraphenylporphyrin and 6-(1,4,8,11-tetraazacyclotetradecane)2-ethyl-naphthalimide) are characterised. The normalised fluorescence intensity over time of the latter chemosensor dye complex in the presence of Zn ion is observed to accelerate with an increasing rate constant, k = 1.94 min-1 at T = 15 C and k = 3.64 min-1 at T = 30 C, approaching a factor of ~ 2 with only a change in temperature of DELTA T = 15 C. Thermally tuning these twist and bend associated rates to optimise sensor approaches and device applications is proposed.

physics.ed-ph↗

arXiv2016

Step-index optical fibre drawn from 3D printed preforms

Optical fibre is drawn from a dual-head 3D printer fabricated preform made of two optically transparent plastics with a high index core (NA ~ 0.25, V > 60). The asymmetry observed in the fibre arises from asymmetry in the 3D printing process. The highly multi-mode optical fibre has losses measured by cut-back as low as α ~ 0.44 dB/cm in the near IR.

physics.ins-det↗

arXiv2016

Air-structured optical fibre drawn from a 3D-printed preform

A structured optical fibre is drawn from a 3D-printed structured preform. Preforms containing a single ring of holes around the core are fabricated using filament made from a modified butadiene polymer. More broadly, 3D printers capable of processing soft glasses, silica and other materials are likely to come on line in the not-so distant future. 3D printing of optical preforms signals a new milestone in optical fibre manufacture.

physics.ins-det↗