Scalable Algorithm Research Articles

Concurrent service auto-scaling for Knative resource quota-based serverless system

Serverless computing platforms currently provide application developers with two ways to control their services’ resource usage cost: resource usage-based and resource quota-based. In usage-based systems, resources are assigned to services based on the amount they consume. Meanwhile, a fixed resource amount is reserved and shared between services for quota-based systems. Several studies have been proposed to enhancing the default serverless auto- scaling algorithms to optimize resource usage and service quality. However, almost all previous works targeted usage-based systems and optimized auto-scaling performance for each separate service. In contrast, this work targets auto-scaling of concurrent services in serverless quota-based systems. In serverless quota-based systems, over-quota resource usage and latency Service Level Objective (SLO) violation may occur when auto-scaling new instances during burst traffic moments. Hence, we aim to find an optimal hybrid auto-scaling decision to minimize over-quota resource usage and latency SLO violation. To solve this problem, We applied deep reinforcement learning and traffic prediction techniques. Our solution is developed and implemented based on the characteristics of the most popular open-source serverless platform Knative. For evaluation, we compared our solution with the default Knative auto-scaler and our previously proposed usage-based hybrid Knative auto-scaler. We configured the default Knative auto-scaler with two different settings that can optimize the total resource usage of all services or minimize service latency. Compared to the optimized resource usage version, our solution traded a slightly higher resource usage for an 11.3% latency SLO violation duration reduction. Meanwhile, compared with the optimized latency version, our solution reduced the over-quota resource usage rate by 18.25% while achieving a similar latency SLO preservation performance. Compared to the usage-based hybrid auto-scaler, the latency SLO violation rate and over-quota resource usage were approximately reduced by half.

Performance evaluation of anti-predatory nature-inspired algorithm for small scale to very high scale problems

Performance evaluation of anti-predatory nature-inspired algorithm for small scale to very high scale problems

Exploring the depths of Multi-Armed Bandit algorithms: From theoretical foundations to modern applications

This paper provides an in-depth and comprehensive analysis of Multi-Armed Bandit (MAB) algorithms, which are crucial in decision-making under uncertainty. It begins with a detailed explanation of the fundamental scenarios where MAB algorithms are applicable, focusing on their features and key strategies. The paper then introduces and explains the core algorithms: Explore-Then-Commit (ETC), Upper Confidence Bound (UCB), and Thompson Sampling (TS). Utilizing various plots, the paper not only analyzes these classical algorithms but also compares them with several of their advanced versions. Additionally, the paper highlights two practical applications of MAB algorithms - in recommendation systems and in wireless digital twin networks - to illustrate their real-world relevance and potential. However, the paper also acknowledges the challenges posed by the complexity of different bandit settings, which affect the efficiency and scalability of MAB algorithms, indicating the need for ongoing research in this field. This review aims to delve deep into the realm of MAB algorithms, offering a thorough understanding of their theoretical underpinnings as well as practical implications.

Finding small feedback arc sets on large graphs

The minimum feedback arc set problem (FASP), which seeks to remove a minimum set of arcs from a directed graph to make the remaining graph acyclic, is fundamental in graph algorithms with many applications in social network analysis, circuit testing, deadlock resolution, algorithmic games, and other fields. However, in theory, FASP is NP-Hard and cannot be approximated within any constant under the Unique-Games Conjecture. In practice, the performance of existing algorithms on large graphs is still unsatisfactory. To solve FASP on large graphs, we propose two novel methods in this paper. First, we systematically study the reduction rules for FASP that can quickly reduce the scale of input graphs without losing optimality. We integrate the reduction rules into an algorithm that can be used as a preprocessor for all existing FASP algorithms to simplify the input graphs. Second, we provide a scalable heuristic algorithm based on the divide-and-conquer method for better solution quality. Extensive experiments on a wide range of real-world graphs show that the reduction algorithm effectively reduces the size of input instances. Furthermore, for half of the circuit benchmark graphs, our final algorithm improves the quality of the existing heuristic solution by 24% to 40% within an acceptable running time.

Asynchronous Parallel Large-Scale Gaussian Process Regression.

Gaussian process regression (GPR) is an important nonparametric learning method in machine learning research with many real-world applications. It is well known that training large-scale GPR is a challenging task due to the required heavy computational cost and large volume memory. To address this challenging problem, in this article, we propose an asynchronous doubly stochastic gradient algorithm to handle the large-scale training of GPR. We formulate the GPR to a convex optimization problem, i.e., kernel ridge regression. After that, in order to efficiently solve this convex kernel problem, we first use the random feature mapping method to approximate the kernel model and then utilize two unbiased stochastic approximations, i.e., stochastic variance reduced gradient and stochastic coordinate descent, to update the solution asynchronously and in parallel. In this way, our algorithm scales well in both sample size and dimensionality, and speeds up the training computation. More importantly, we prove that our algorithm has a global linear convergence rate. Our experimental results on eight large-scale benchmark datasets with both regression and classification tasks show that the proposed algorithm outperforms the existing state-of-the-art GPR methods.

Bayesian Analysis of Exponential Random Graph Models Using Stochastic Gradient Markov Chain Monte Carlo

The exponential random graph model (ERGM) is a popular model for social networks, which is known to have an intractable likelihood function. Sampling from the posterior for such a model is a long-standing problem in statistical research. We analyze the performance of the stochastic gradient Langevin dynamics (SGLD) algorithm (also known as noisy Longevin Monte Carlo) in tackling this problem, where the stochastic gradient is calculated via running a short Markov chain (the so-called inner Markov chain in this paper) at each iteration. We show that if the model size grows with the network size slowly enough, then SGLD converges to the true posterior in 2-Wasserstein distance as the network size and iteration number become large regardless of the length of the inner Markov chain performed at each iteration. Our study provides a scalable algorithm for analyzing large-scale social networks with possibly high-dimensional ERGMs.

P‐248: Late‐News Poster: Image Upscaling Techniques for IT Products; Evaluation and Analysis

This paper introduces an upscaler technology which is proven to be effective in improving power reduction for IT products at the system level. Proposed upscaler algorithm utilizes Lanczos and bilinear interpolation both of which are widely used in currently existing upscaler designs on the industry, and carefully analyzes the image distortion caused by 2x scaling of the input image and manipulates the possible artifacts by considering the key performance criteria in the up‐scaling. Our experimental results show that image quality distortion due to the proposed scaling algorithm is significantly improved while the small details in the images are preserved at most thanks to the combination of “blended upscale” and “edge enhancement” sub‐functions in the proposed upscaling design. According to our subjective evaluation, we got the average score of 4.39 over 15 persons with 21 images from well‐known image benchmark suites.

Blockchian Based Hybrid Consensus Algorithm for Data Security in Edge Computing

Abstract: Edge computing has emerged as a paradigm to process data closer to its source, enabling low-latency and real-time applications. However, the distributed and heterogeneous nature of edge environments poses significant challenges for ensuring data security and integrity. Blockchain technology offers a promising solution by providing a decentralized and immutable ledger for recording transactions. This abstract proposes a consensus algorithm tailored for securing data in edge computing environments using blockchain technology. The proposed consensus algorithm leverages a hybrid approach combining Proof of Authority (PoA) and Practical Byzantine Fault Tolerance (PBFT). PoA ensures that only trusted nodes within the edge network are eligible to participate in the consensus process, mitigating the risk of malicious attacks. PBFT enhances the efficiency and scalability of the consensus algorithm by allowing fast decision-making among a subset of nodes, thereby reducing latency in data validation and verification. Furthermore, smart contracts are employed to enforce data access control and execute predefined rules for data processing and sharing. These smart contracts facilitate secure and transparent interactions among edge devices, ensuring that only authorized parties can access and manipulate sensitive data. The integration of blockchain technology with edge computing offers several benefits, including enhanced data security, traceability, and accountability. By employing a tailored consensus algorithm, edge computing environments can effectively mitigate security threats while maintaining low-latency data processing capabilities. Future research directions may explore optimization techniques to further improve the performance and scalability of the proposed solution in large-scale edge networks.

A Scalable Joint Estimation Algorithm for SOC and SOH of All Individual Cells within the Battery Pack and Its HIL Implementation

A Scalable Joint Estimation Algorithm for SOC and SOH of All Individual Cells within the Battery Pack and Its HIL Implementation

Efficient calculation of unbiased atomic forces in ab initio variational Monte Carlo

quantum Monte Carlo (QMC) is a state-of-the-art numerical approach for evaluating accurate expectation values of many-body wave functions. However, one of the major drawbacks that still hinders widespread QMC applications is the lack of an affordable scheme to compute unbiased atomic forces. In this study, we propose an efficient method to obtain unbiased atomic forces and pressures in the variational Monte Carlo (VMC) framework with the Jastrow-correlated Slater determinant ansatz or the Jastrow antisymmetrized geminal power ansatz, exploiting the gauge-invariant and locality properties of their geminal representation. We demonstrate the effectiveness of our method for H2 and Cl2 molecules and for the cubic boron nitride crystal. Our framework has a better algorithmic scaling with the system size than the traditional finite-difference method and, in practical applications, is as efficient as single-point VMC calculations. Thus, it paves the way to study dynamical properties of materials, such as phonons, and is beneficial for pursuing more reliable machine-learning interatomic potentials based on unbiased VMC forces. Published by the American Physical Society 2024

Early adverse physiological event detection using commercial wearables: challenges and opportunities

Data from commercial off-the-shelf (COTS) wearables leveraged with machine learning algorithms provide an unprecedented potential for the early detection of adverse physiological events. However, several challenges inhibit this potential, including (1) heterogeneity among and within participants that make scaling detection algorithms to a general population less precise, (2) confounders that lead to incorrect assumptions regarding a participant’s healthy state, (3) noise in the data at the sensor level that limits the sensitivity of detection algorithms, and (4) imprecision in self-reported labels that misrepresent the true data values associated with a given physiological event. The goal of this study was two-fold: (1) to characterize the performance of such algorithms in the presence of these challenges and provide insights to researchers on limitations and opportunities, and (2) to subsequently devise algorithms to address each challenge and offer insights on future opportunities for advancement. Our proposed algorithms include techniques that build on determining suitable baselines for each participant to capture important physiological changes and label correction techniques as it pertains to participant-reported identifiers. Our work is validated on potentially one of the largest datasets available, obtained with 8000+ participants and 1.3+ million hours of wearable data captured from Oura smart rings. Leveraging this extensive dataset, we achieve pre-symptomatic detection of COVID-19 with a performance receiver operator characteristic (ROC) area under the curve (AUC) of 0.725 without correction techniques, 0.739 with baseline correction, 0.740 with baseline correction and label correction on the training set, and 0.777 with baseline correction and label correction on both the training and the test set. Using the same respective paradigms, we achieve ROC AUCs of 0.919, 0.938, 0.943 and 0.994 for the detection of self-reported fever, and 0.574, 0.611, 0.601, and 0.635 for detection of self-reported shortness of breath. These techniques offer improvements across almost all metrics and events, including PR AUC, sensitivity at 75% specificity, and precision at 75% recall. The ring allows continuous monitoring for detection of event onset, and we further demonstrate an improvement in the early detection of COVID-19 from an average of 3.5 days to an average of 4.1 days before a reported positive test result.

A modular framework for multi-scale tissue imaging and neuronal segmentation.

The development of robust tools for segmenting cellular and sub-cellular neuronal structures lags behind the massive production of high-resolution 3D images of neurons in brain tissue. The challenges are principally related to high neuronal density and low signal-to-noise characteristics in thick samples, as well as the heterogeneity of data acquired with different imaging methods. To address this issue, we design a framework which includes sample preparation for high resolution imaging and image analysis. Specifically, we set up a method for labeling thick samples and develop SENPAI, a scalable algorithm for segmenting neurons at cellular and sub-cellular scales in conventional and super-resolution STimulated Emission Depletion (STED) microscopy images of brain tissues. Further, we propose a validation paradigm for testing segmentation performance when a manual ground-truth may not exhaustively describe neuronal arborization. We show that SENPAI provides accurate multi-scale segmentation, from entire neurons down to spines, outperforming state-of-the-art tools. The framework will empower image processing of complex neuronal circuitries.

Prospective Validation of Glial Fibrillary Acidic Protein, d ‐Dimer, and Clinical Scales for Acute Large‐Vessel Occlusion Ischemic Stroke Detection

Background Large‐vessel occlusion (LVO) ischemic stroke is responsible for significant morbidity and mortality. We have previously described a novel tool for acute LVO detection that combines blood‐based biomarkers (glial fibrillary acidic protein and d ‐dimer) with stroke severity scales to achieve high accuracy. Accordingly, the present study sought to prospectively validate cutoff values that we had previously established for biomarkers and scales. Methods The TIME (Testing for Identification Markers of Stroke) trial was designed as a prospective observational diagnostic accuracy study. All ambulance‐identified stroke code activations <18 hours from symptom onset were recruited at Brandon Regional Hospital (Brandon, FL) between May 2021 and August 2022. Previously determined cutoff concentrations of plasma glial fibrillary acidic protein (213 pg/mL) and d ‐dimer (600 ng/mL) were used in combination with prehospital stroke scales to detect LVO. We compared rates of LVO detection against a reference standard using computed tomography/magnetic resonance angiography. Results A total of 382 patients with suspected stroke were recruited. The final cohort was composed of 323 patients with suspected stroke with the following distribution: LVO ischemic stroke (n = 29, 9%), non‐LVO ischemic stroke (n = 48, 15%), hemorrhagic stroke (n = 13, 4%), transient ischemic attack (n = 12, 3.9%), and stroke mimics (n = 220, 68.1%). Combining blood‐based biomarkers (glial fibrillary acidic protein and d ‐dimer) with the scale field assessment stroke triage for emergency destination yielded the best performance for LVO detection, with specificity of 94% and sensitivity of 71%. Performance was found to be higher in a subanalysis focusing on patients presenting <6 hours from symptom onset, with 93% specificity and 81% sensitivity. Critically, application of the biomarker and stroke scale algorithms ruled out all patients with hemorrhage. Conclusion The present work prospectively validated the potential utility of previously defined glial fibrillary acidic protein and d ‐dimer cutoff levels (ie, 213 pg/mL and 600 ng/mL, respectively), demonstrating their value for discrimination of LVO stroke from differential diagnoses during code stroke workups. (ClinicalTrials.gov number, NCT04292600.)

A systematic review of optimization methods for recovery planning in cyber–physical infrastructure networks: Current state and future trends

Cyber–Physical Systems are increasingly complex and frequently integrated into modern societies via critical infrastructure systems, products, and services. Consequently, there is a need for reliable functionality of these complex systems under various scenarios, from physical failures due to aging to cyber attacks. However, the development of effective strategies to restore disrupted cyber–physical infrastructure systems continues to be a major challenge. Even though there have been an increasing number of papers evaluating recovery planning in cyber–physical infrastructures networks, a comprehensive literature review focusing on mathematical modeling and optimization methods is still lacking. Thus, this study critically analyzes the literature on optimization techniques for recovery planning of cyber–physical infrastructure networks after a disruption, to synthesize key findings on the current methods in this domain. A total of 152 relevant research papers are reviewed following an extensive assessment of all major scientific databases. The main mathematical modeling practices and optimization methods are identified for both deterministic and stochastic formulations, categorizing them based on the solution approach (exact, heuristic, metaheuristic), objective function, and network size. Having identified the gaps, a set of future trends for both the methodology and application of optimization algorithms is presented. Overall, there is a need to shift toward scalable optimization solution algorithms, empowered by data-driven methods and machine learning algorithms, to provide reliable and computationally efficient decision-support systems for decision-makers and practitioners.

Imaging swiFTly: streaming widefield Fourier Transforms for large-scale interferometry

We describe a scalable distributed imaging algorithm framework for next-generation radio telescopes, managing the Fourier transform from apertures to sky (or vice versa) with a focus on minimising memory load, data transfers, and computation. Our algorithm uses smooth window functions to isolate the influence between specific regions of spatial-frequency and image space. This allows the distribution of image data between nodes and the construction of segments of frequency space exactly when and where needed. The developed prototype distributes terabytes of image data across many nodes, while generating visibilities at throughput and accuracy competitive with existing software. Scaling is demonstrated to be better than cubic in problem complexity (for baseline length and field of view), reducing the risk involved in growing radio astronomy processing to large telescopes like the Square Kilometre Array.

Diversity of the genus Chondrilla L. (Asteraceae) in Eastern Europe

Abstract. Kashin AS, Parkhomenko AS, Kondratieva AO, Kritskaya TA, Bogoslov AV, Shilova IV. 2024. Diversity of the genus Chondrilla L. (Asteraceae) in Eastern Europe. Biodiversitas 25: 1901-1910. The genus Chondrilla (Asteraceae) has a complex taxonomic structure due to widespread apomictic reproduction. This study aimed to identify the variability and clarify the status of Chondrilla species growing in Eastern Europe (European Russia and Western Kazakhstan). The study of morphological variability in Chondrilla species was assessed using 28 morphometric parameters. Moreover, 840 specimens from 28 populations of 8 Chondrilla species from Eastern Europe were analyzed. Principal Component Analysis (PCA), Nonmetric Multi-Dimensional Scaling (NMDS), recursive decomposition, and random forest algorithm were used for analysis. The ordination methods divided the studied samples into two groups: the first group included samples of C. ambigua and C. pauciflora, and the second group included all remaining samples. Following the resulting binary trees, obtained by recursive decomposition method and the random forest algorithm, a determination key was compiled for the Chondrilla species of Eastern Europe. The high level of morphological variation in Chondrilla and its distribution characteristics revealed in this study by numerical taxonomy and recursive decomposition methods, along with our previous studies results, show that only C. ambigua is a morphologically and genetically well-separated species, including C. pauciflora as a triploid cytotype; C. canescens, C. graminea, C. latifolia and C. juncea are synonyms with the priority name C. juncea and should be considered as subspecies; and C. brevirostris and C. laticoronata are hybrids with C. ambigua as maternal parent.

Scalable and unbiased sequence-informed embedding of single-cell ATAC-seq data with CellSpace

Standard scATAC sequencing (scATAC-seq) analysis pipelines represent cells as sparse numeric vectors relative to an atlas of peaks or genomic tiles and consequently ignore genomic sequence information at accessible loci. Here we present CellSpace, an efficient and scalable sequence-informed embedding algorithm for scATAC-seq that learns a mapping of DNA k-mers and cells to the same space, to address this limitation. We show that CellSpace captures meaningful latent structure in scATAC-seq datasets, including cell subpopulations and developmental hierarchies, and can score transcription factor activities in single cells based on proximity to binding motifs embedded in the same space. Importantly, CellSpace implicitly mitigates batch effects arising from multiple samples, donors or assays, even when individual datasets are processed relative to different peak atlases. Thus, CellSpace provides a powerful tool for integrating and interpreting large-scale scATAC-seq compendia.

PROTAX-GPU: a scalable probabilistic taxonomic classification system for DNA barcodes.

DNA-based identification is vital for classifying biological specimens, yet methods to quantify the uncertainty of sequence-based taxonomic assignments are scarce. Challenges arise from noisy reference databases, including mislabelled entries and missing taxa. PROTAX addresses these issues with a probabilistic approach to taxonomic classification, advancing on methods that rely solely on sequence similarity. It provides calibrated probabilistic assignments to a partially populated taxonomic hierarchy, accounting for taxa that lack references and incorrect taxonomic annotation. While effective on smaller scales, global application of PROTAX necessitates substantially larger reference libraries, a goal previously hindered by computational barriers. We introduce PROTAX-GPU, a scalable algorithm capable of leveraging the global Barcode of Life Data System (>14 million specimens) as a reference database. Using graphics processing units (GPU) to accelerate similarity and nearest-neighbour operations and the JAX library for Python integration, we achieve over a 1000 × speedup compared with the central processing unit (CPU)-based implementation without compromising PROTAX's key benefits. PROTAX-GPU marks a significant stride towards real-time DNA barcoding, enabling quicker and more efficient species identification in environmental assessments. This capability opens up new avenues for real-time monitoring and analysis of biodiversity, advancing our ability to understand and respond to ecological dynamics. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'.

Deep Learning Approaches for Medical Image Analysis and Diagnosis.

In addition to enhancing diagnostic accuracy, deep learning techniques offer the potential to streamline workflows, reduce interpretation time, and ultimately improve patient outcomes. The scalability and adaptability of deep learning algorithms enable their deployment across diverse clinical settings, ranging from radiology departments to point-of-care facilities. Furthermore, ongoing research efforts focus on addressing the challenges of data heterogeneity, model interpretability, and regulatory compliance, paving the way forseamless integration of deep learning solutions into routine clinical practice. As the field continues to evolve, collaborations between clinicians, data scientists, and industry stakeholders will be paramount in harnessing the full potential of deep learning for advancing medical image analysis and diagnosis. Furthermore, the integration of deep learning algorithms with other technologies, including natural language processing and computer vision, may foster multimodal medical data analysis and clinical decision support systems to improve patient care. The future of deep learning in medical image analysis and diagnosis is promising. With each success and advancement, this technology is getting closer to being leveraged for medical purposes. Beyond medical image analysis, patient care pathways like multimodal imaging, imaging genomics, and intelligent operating rooms or intensive care units can benefitfrom deep learning models.

Matrix nearness problems with off-block-diagonal rank constraints

This paper explores the problem of finding a rank-structured approximation to a given Hermitian positive definite matrix. We first study a fundamental matrix nearness problem which seeks an approximation whose off-diagonal block must have low rank. Our investigation focuses on three objective functions: the spectral norm, Frobenius divergence, and log-determinant divergence. We derive closed-form solutions for the fundamental nearness problem, extending our analysis to include problems with a range restriction on the off-diagonal block. Building upon these results, we propose a practical, scalable algorithm to compute a positive definite hierarchical off-diagonal low-rank approximation to a Hermitian positive definite matrix. Through numerical experiments, we explore the use of our approximation as a general-purpose preconditioner for the conjugate gradient method. Our results demonstrate that the log-determinant divergence-based approximation exhibits good performance across a diverse set of test problems.