arXiv subjects

Ayse K. Coskun

Publications and source records attributed to Ayse K. Coskun.

11 recordsLinked to original sources

arXiv2026

PHAT: PHotonic Accelerator for TFHE

Fully Homomorphic Encryption (FHE) enables secure computation on encrypted data, making it a promising solution for privacy-preserving applications in the cloud. Among various FHE schemes, FHE over the Torus (TFHE) stands out due to its support for arbitrary operations. However, its high computation and communication overhead, particularly in the Fast Fourier Transform (FFT) operations required during bootstrapping, limits its practicality for real-world applications. Conventional electronic accelerators struggle to achieve sufficient throughput due to the limitations of technology scaling and the memory-wall problem. To address these challenges, we propose PHAT, a PHotonic Accelerator for TFHE leveraging Optically-addressed Phase-Change Memory (OPCM). OPCM-based processing-in-memory systems offer high computation and communication throughput, making them well-suited for accelerating FFT operations in TFHE. However, directly mapping FFT to OPCM presents challenges such as high-precision analog computation and the high latency and energy cost of programming OPCM cells. To overcome these challenges, we introduce a novel electro-photonic accelerator architecture optimized for TFHE, featuring OPCM-based FFT units, a twiddle-stationary dataflow tailored for OPCM, and a scheduling mechanism to maximize the utilization of the FFT units. PHAT delivers $2.14\times$--$5.10\times$ speedup across four real-world TFHE workloads against the state-of-the-art ASIC accelerator. Our approach significantly enhances the performance of TFHE applications, paving the way for practical and efficient homomorphic encryption in cloud computing.

cs.CR↗

arXiv2026

Praxium: Diagnosing Cloud Anomalies with AI-based Telemetry and Dependency Analysis

As the modern microservice architecture for cloud applications grows in popularity, cloud services are becoming increasingly complex and more vulnerable to misconfiguration and software bugs. Traditional approaches rely on expert input to diagnose and fix microservice anomalies, which lacks scalability in the face of the continuous integration and continuous deployment (CI/CD) paradigm. Microservice rollouts, containing new software installations, have complex interactions with the components of an application. Consequently, this added difficulty in attributing anomalous behavior to any specific installation or rollout results in potentially slower resolution times. To address the gaps in current diagnostic methods, this paper introduces Praxium, a framework for anomaly detection and root cause inference. Praxium aids administrators in evaluating target metric performance in the context of dependency installation information provided by a software discovery tool, PraxiPaaS. Praxium continuously monitors telemetry data to identify anomalies, then conducts root cause analysis via causal impact on recent software installations, in order to provide site reliability engineers (SRE) relevant information about an observed anomaly. In this paper, we demonstrate that Praxium is capable of effective anomaly detection and root cause inference, and we provide an analysis on effective anomaly detection hyperparameter tuning as needed in a practical setting. Across 75 total trials using four synthetic anomalies, anomaly detection consistently performs at >0.97 macro-F1. In addition, we show that causal impact analysis reliably infers the correct root cause of anomalies, even as package installations occur at increasingly shorter intervals.

cs.SE↗

arXiv2026

PALS: Power-Aware LLM Serving for Mixture-of-Experts Models

Large language model (LLM) inference has become a dominant workload in modern data centers, driving significant GPU utilization and energy consumption. While prior systems optimize throughput and latency by batching, scheduling, and parallelism, they largely treat GPU power as a static constraint rather than a controllable resource. In this paper, we present a power-aware runtime for LLM serving, PALS, that treats GPU power caps as a first-class control knob and jointly optimizes them with software parameters such as batch size. The system combines lightweight offline power-performance models with a feedback-driven controller to select configurations that satisfy throughput targets while maximizing energy efficiency. We implement PALS within an existing LLM serving framework, vLLM, demonstrating that it requires no model retraining or API changes. Across multi-GPU systems and both dense and mixture-of-experts (MoE) models, PALS improves energy efficiency by up to 26.3%, reduces QoS violations by 4x to 7x under power constraints, and tracks dynamic power budgets. These results highlight the potential of integrating power control directly into LLM inference runtimes, enabling energy-proportional and grid-interactive AI systems.

cs.AI↗

arXiv2025

UniCoMTE: A Universal Counterfactual Framework for Explaining Time-Series Classifiers on ECG Data

Machine learning models, particularly deep neural networks, have demonstrated strong performance in classifying complex time series data. However, their black-box nature limits trust and adoption, especially in high-stakes domains such as healthcare. To address this challenge, we introduce UniCoMTE, a model-agnostic framework for generating counterfactual explanations for multivariate time series classifiers. The framework identifies temporal features that most heavily influence a model's prediction by modifying the input sample and assessing its impact on the model's prediction. UniCoMTE is compatible with a wide range of model architectures and operates directly on raw time series inputs. In this study, we evaluate UniCoMTE's explanations on a time series ECG classifier. We quantify explanation quality by comparing our explanations' comprehensibility to comprehensibility of established techniques (LIME and SHAP) and assessing their generalizability to similar samples. Furthermore, clinical utility is assessed through a questionnaire completed by medical experts who review counterfactual explanations presented alongside original ECG samples. Results show that our approach produces concise, stable, and human-aligned explanations that outperform existing methods in both clarity and applicability. By linking model predictions to meaningful signal patterns, the framework advances the interpretability of deep learning models for real-world time series applications.

cs.LG↗

arXiv2025

Turning AI Data Centers into Grid-Interactive Assets: Results from a Field Demonstration in Phoenix, Arizona

Artificial intelligence (AI) is fueling exponential electricity demand growth, threatening grid reliability, raising prices for communities paying for new energy infrastructure, and stunting AI innovation as data centers wait for interconnection to constrained grids. This paper presents the first field demonstration, in collaboration with major corporate partners, of a software-only approach--Emerald Conductor--that transforms AI data centers into flexible grid resources that can efficiently and immediately harness existing power systems without massive infrastructure buildout. Conducted at a 256-GPU cluster running representative AI workloads within a commercial, hyperscale cloud data center in Phoenix, Arizona, the trial achieved a 25% reduction in cluster power usage for three hours during peak grid events while maintaining AI quality of service (QoS) guarantees. By orchestrating AI workloads based on real-time grid signals without hardware modifications or energy storage, this platform reimagines data centers as grid-interactive assets that enhance grid reliability, advance affordability, and accelerate AI's development.

cs.DC↗

arXiv2024

A New Dataflow Implementation to Improve Energy Efficiency of Monolithic 3D Systolic Arrays

Systolic arrays are popular for executing deep neural networks (DNNs) at the edge. Low latency and energy efficiency are key requirements in edge devices such as drones and autonomous vehicles. Monolithic 3D (MONO3D) is an emerging 3D integration technique that offers ultra-high bandwidth among processing and memory elements with a negligible area overhead. Such high bandwidth can help meet the ever-growing latency and energy efficiency demands for DNNs. This paper presents a novel implementation for weight stationary (WS) dataflow in MONO3D systolic arrays, called WS-MONO3D. WS-MONO3D utilizes multiple resistive RAM layers and SRAM with high-density vertical interconnects to multicast inputs and perform high-bandwidth weight pre-loading while maintaining the same order of multiply-and-accumulate operations as in native WS dataflow. Consequently, WS-MONO3D eliminates input and weight forwarding cycles and, thus, provides up to 40% improvement in energy-delay-product (EDP) over the native WS implementation in 2D at iso-configuration. WS-MONO3D also provides 10X improvement in inference per second per watt per footprint due to multiple vertical tiers. Finally, we also show that temperature impacts the energy efficiency benefits in WS-MONO3D.

cs.ET↗

arXiv2022

Temperature-Aware Monolithic 3D DNN Accelerators for Biomedical Applications

In this paper, we focus on temperature-aware Monolithic 3D (Mono3D) deep neural network (DNN) inference accelerators for biomedical applications. We develop an optimizer that tunes aspect ratios and footprint of the accelerator under user-defined performance and thermal constraints, and generates near-optimal configurations. Using the proposed Mono3D optimizer, we demonstrate up to 61% improvement in energy efficiency for biomedical applications over a performance-optimized accelerator.

cs.ET↗

arXiv2021

Architecting Optically-Controlled Phase Change Memory

Phase Change Memory (PCM) is an attractive candidate for main memory as it offers non-volatility and zero leakage power, while providing higher cell densities, longer data retention time, and higher capacity scaling compared to DRAM. In PCM, data is stored in the crystalline or amorphous state of the phase change material. The typical electrically-controlled PCM (EPCM), however, suffers from longer write latency and higher write energy compared to DRAM and limited multi-level cell (MLC) capacities. These challenges limit the performance of data-intensive applications running on computing systems with EPCMs. Recently, researchers demonstrated optically-controlled PCM (OPCM) cells, with support for 5 bits/cell in contrast to 2 bits/cell in EPCM. These OPCM cells can be accessed directly with optical signals that are multiplexed in high-bandwidth-density silicon-photonic links. The higher MLC capacity in OPCM and the direct cell access using optical signals enable an increased read/write throughput and lower energy per access than EPCM. However, due to the direct cell access using optical signals, OPCM systems cannot be designed using conventional memory architecture. We need a complete redesign of the memory architecture that is tailored to the properties of OPCM technology. This paper presents the design of a unified network and main memory system called COSMOS that combines OPCM and silicon-photonic links to achieve high memory throughput. COSMOS is composed of a hierarchical multi-banked OPCM array with novel read and write access protocols, and uses an Electrical-Optical-Electrical (E-O-E) control unit to interface with the processor. Our evaluation of a 2.5D-integrated system containing a processor and COSMOS demonstrates 2.14x average speedup compared to an EPCM system. COSMOS consumes 3.8x lower read energy-per-bit and 5.97x lower write energy-per-bit compared to EPCM.

cs.AR↗

arXiv2020

ConfEx: A Framework for Automating Text-based Software Configuration Analysis in the Cloud

Modern cloud services have complex architectures, often comprising many software components, and depend on hundreds of configurations parameters to function correctly, securely, and with high performance. Due to the prevalence of open-source software, developers can easily deploy services using third-party software without mastering the configurations of that software. As a result, configuration errors (i.e., misconfigurations) are among the leading causes of service disruptions and outages. While existing cloud automation tools ease the process of service deployment and management, support for detecting misconfigurations in the cloud has not been addressed thoroughly, likely due to the lack of frameworks suitable for consistent parsing of unstandardized configuration files. This paper introduces ConfEx, a framework that enables discovery and extraction of text-based software configurations in the cloud. ConfEx uses a novel vocabulary-based technique to identify configuration files in cloud system instances with unlabeled content. To extract the information in these files, ConfEx leverages existing configuration parsers and post-processes the extracted data for analysis. We show that ConfEx achieves over 99% precision and 100% recall in identifying configuration files on 7805 popular Docker Hub images. Using two applied examples, we demonstrate that ConfEx also enables detecting misconfigurations in the cloud via existing tools that are designed for configurations represented as key-value pairs, revealing 184 errors in public Docker Hub images.

cs.SE↗

arXiv2020

Counterfactual Explanations for Machine Learning on Multivariate Time Series Data

Applying machine learning (ML) on multivariate time series data has growing popularity in many application domains, including in computer system management. For example, recent high performance computing (HPC) research proposes a variety of ML frameworks that use system telemetry data in the form of multivariate time series so as to detect performance variations, perform intelligent scheduling or node allocation, and improve system security. Common barriers for adoption for these ML frameworks include the lack of user trust and the difficulty of debugging. These barriers need to be overcome to enable the widespread adoption of ML frameworks in production systems. To address this challenge, this paper proposes a novel explainability technique for providing counterfactual explanations for supervised ML frameworks that use multivariate time series data. The proposed method outperforms state-of-the-art explainability methods on several different ML frameworks and data sets in metrics such as faithfulness and robustness. The paper also demonstrates how the proposed method can be used to debug ML frameworks and gain a better understanding of HPC system telemetry data.

cs.LG↗

arXiv2016

Optimizing Energy Storage Participation in Emerging Power Markets

The growing amount of intermittent renewables in power generation creates challenges for real-time matching of supply and demand in the power grid. Emerging ancillary power markets provide new incentives to consumers (e.g., electrical vehicles, data centers, and others) to perform demand response to help stabilize the electricity grid. A promising class of potential demand response providers includes energy storage systems (ESSs). This paper evaluates the benefits of using various types of novel ESS technologies for a variety of emerging smart grid demand response programs, such as regulation services reserves (RSRs), contingency reserves, and peak shaving. We model, formulate and solve optimization problems to maximize the net profit of ESSs in providing each demand response. Our solution selects the optimal power and energy capacities of the ESS, determines the optimal reserve value to provide as well as the ESS real-time operational policy for program participation. Our results highlight that applying ultra-capacitors and flywheels in RSR has the potential to be up to 30 times more profitable than using common battery technologies such as LI and LA batteries for peak shaving.

eess.SY↗