Scalability Of System Research Articles

AutoML with Bayesian Optimizations for Big Data Management

The field of automated machine learning (AutoML) has gained significant attention in recent years due to its ability to automate the process of building and optimizing machine learning models. However, the increasing amount of big data being generated has presented new challenges for AutoML systems in terms of big data management. In this paper, we introduce Fabolas and learning curve extrapolation as two methods for accelerating hyperparameter optimization. Four methods for quickening training were presented including Bag of Little Bootstraps, k-means clustering for Support Vector Machines, subsample size selection for gradient descent, and subsampling for logistic regression. Additionally, we also discuss the use of Markov Chain Monte Carlo (MCMC) methods and other stochastic optimization techniques to improve the efficiency of AutoML systems in managing big data. These methods enhance various facets of the training process, making it feasible to combine them in diverse ways to gain further speedups. We review several combinations that have potential and provide a comprehensive understanding of the current state of AutoML and its potential for managing big data in various industries. Furthermore, we also mention the importance of parallel computing and distributed systems to improve the scalability of the AutoML systems while working with big data.

Web-based mixed reality video fusion with remote rendering

BackgroundMixed reality (MR) video fusion systems merge video imagery with 3D scenes to make the scene more realistic and help users understand the video content and temporal–spatial correlation between them, reducing the user′s cognitive load. MR video fusion are used in various applications; however, video fusion systems require powerful client machines because video streaming delivery, stitching, and rendering are computationally intensive. Moreover, huge bandwidth usage is another critical factor that affects the scalability of video-fusion systems. MethodsOur framework proposes a fusion method for dynamically projecting video images into 3D models as textures. ResultsSeveral experiments on different metrics demonstrate the effectiveness of the proposed framework. ConclusionsThe framework proposed in this study can overcome client limitations by utilizing remote rendering. Furthermore, the framework we built is based on browsers. Therefore, the user can test the MR video fusion system with a laptop or tablet without installing any additional plug-ins or application programs.

Enabling technology for global 3D+3V hybrid-Vlasov simulations of near-Earth space

We present methods and algorithms that allow the Vlasiator code to run global, three-dimensional hybrid-Vlasov simulations of Earth's entire magnetosphere. The key ingredients that make Vlasov simulations at magnetospheric scales possible are the sparse velocity space implementation and spatial adaptive mesh refinement. We outline the algorithmic improvement of the semi-Lagrangian solver for six-dimensional phase space quantities, discuss the coupling of Vlasov and Maxwell equations' solvers in a refined mesh, and provide performance figures from simulation test runs that demonstrate the scalability of this simulation system to full magnetospheric runs.

PartitionChain: A Scalable and Reliable Data Storage Strategy for Permissioned Blockchain

Blockchain, a specific distributed database which maintains a list of data records against tampering and corruption, has aroused wide interests and become a hot topic in the real world. Nevertheless, the increasingly heavy storage consumption brought by the full-replication data storage mechanism, becomes a bottleneck to the system scalability. To address this problem, a reliable storage scheme named BFT-Store (Qi <i>et al.</i> 2020), integrating erasure coding with Byzantine Fault Tolerance (BFT), was proposed recently. While, three critical problems are still left open: (i) The complex re-initialization process of the blockchain when the number of nodes varies; (ii) The high computational overload of downloading data; (iii) The massive communication on the network. This paper proposes a better trade-off for blockchain storage scheme termed PartitionChain which addresses the above three problems, maintaining the merits of BFT-Store. First, our scheme allows the original nodes to merely update a single aggregate signature (e.g., 320 bits) when the number of nodes varies. Using aggregate signatures as the proof of the encoded data not only saves the storage costs but also gets rid of the trusted third party. Second, the computational complexity of retrieving data by decoding, compared to BFT-Store, is greatly reduced by about <inline-formula><tex-math notation="LaTeX">$2^{18}$</tex-math></inline-formula> times on each node. Third, the amount of transmitted data for recovering each block is reduced from <inline-formula><tex-math notation="LaTeX">$O(n)$</tex-math></inline-formula> (assuming <inline-formula><tex-math notation="LaTeX">$n$</tex-math></inline-formula> is the number of nodes) to <inline-formula><tex-math notation="LaTeX">$O(1)$</tex-math></inline-formula> , by partitioning each block into smaller pieces and applying Reed-Solomon coding to each block. Furthermore, this paper also introduces a reputation ranking system where the malicious behaviors of the nodes can be detected and marked, enabling PartitionChain to check the credits of each node termly and expel the nodes with misbehavior to the specific extent. Comparing with BFT-Store, our scheme allows blockchain system to suit dynamic network with higher efficiency and scalability.

Domain-Driven Design in Modern Software Architecture Best Practices and Patterns

Domain driven design (DDD) is crucial, in shaping software structures to meet business goals effectively while also ensuring flexibility to accommodate changing needs over time. This article delves into the fusion of DDD with microservices architecture and highlights how DDD concepts like bounded contexts and common language enhance the organization and scalability of systems. The collaboration between DDD and microservices allows for service deployment and expansion easily by tackling issues such, as distributed transactions and data coherence. Align service boundaries, with domain models to improve system flexibility and align with business objectives effectively in organizations. This strategy is especially valuable in paced environments like IoT and cloud computing where scalability and adaptabilities are crucial. The article also explores methods for implementing DDD such as creating a language among stakeholders and using iterative development approaches to keep domain models in sync with business requirements. In summary DDD offers a framework, for handling software intricacies encouraging creativity and making systems to shifts efficiently

The efficient employment of the space division multiplexing technique with OADMs for more flexibility and scalability in optical systems

Abstract The optical add drop multiplexers (OADMs) to be modeled and analyzed depends up on the simulation variables are demonstrated. Transmitted power per channel variations are studied against both fiber length and number of transmitted channels at receiver bandwidth of 40 GHz and temperature of 25 °C. Received power from each channel variations are demonstrated against both fiber length and number of transmitted channels at receiver bandwidth of 40 GHz and temperature of 25 °C. The bit error rate variations are clarified against both fiber length and number of transmitted channels at receiver bandwidth of 40 GHz and temperature of 25 °C. The optical link map through fiber cable core is clarified. Total pulse broadening variations are illustrated against both fiber length and number of transmitted channels at receiver bandwidth of 40 GHz and temperature of 25 °C. The lighted signal/noise ratio variations are demonstrated against both receiver bandwidth and number of transmitted channels at transmission distance of 200 km and temperature of 25 °C. Channel capacity variations are also demonstrated against both receiver bandwidth and number of transmitted channels at transmission distance of 200 km and temperature of 25 °C.

Hierarchical Clustering-based Personalized Federated Learning for Robust and Fair Human Activity Recognition

Currently, federated learning (FL) can enable users to collaboratively train a global model while protecting the privacy of user data, which has been applied to human activity recognition (HAR) tasks. However, in real HAR scenarios, deploying an FL system needs to consider multiple aspects, including system accuracy, fairness, robustness, and scalability. Most existing FL frameworks aim to solve specific problems while ignoring other properties. In this paper, we propose FedCHAR, a personalized FL framework with a hierarchical clustering method for robust and fair HAR, which not only improves the accuracy and the fairness of model performance by exploiting the intrinsically similar relationship between users but also enhances the robustness of the system by identifying malicious nodes through clustering in attack scenarios. In addition, to enhance the scalability of FedCHAR, we also propose FedCHAR-DC, a scalable and adaptive FL framework which is featured by dynamic clustering and adapting to the addition of new users or the evolution of datasets for realistic FL-based HAR scenarios. We conduct extensive experiments to evaluate the performance of FedCHAR on seven datasets of different sizes. The results demonstrate that FedCHAR could obtain better performance on different datasets than the other five state-of-the-art methods in terms of accuracy, robustness, and fairness. We further validate that FedCHAR-DC exhibits satisfactory scalability on three large-scale datasets regardless of the number of participants.

An evaluation of relational and NoSQL distributed databases on a low-power cluster.

The constant growth of social media, unconventional web technologies, mobile applications, and Internet of Things (IoT) devices create challenges for cloud data systems in order to support huge datasets and very high request rates. NoSQL databases, such as Cassandra and HBase, and relational SQL databases with replication, such as Citus/PostgreSQL, have been used to increase horizontal scalability and high availability of data store systems. In this paper, we evaluated three distributed databases on a low-power low-cost cluster of commodity Single-Board Computers (SBC): relational Citus/PostgreSQL and NoSQL databases Cassandra and HBase. The cluster has 15 Raspberry Pi 3 nodes with Docker Swarm orchestration tool for service deployment and ingress load balancing over SBCs. We believe that a low-cost SBC cluster can support cloud serving goals such as scale-out, elasticity, and high availability. Experimental results clearly demonstrated that there is a trade-off between performance and replication, which provides availability and partition tolerance. Besides, both properties are essential in the context of distributed systems with low-power boards. Cassandra attained better results with its consistency levels specified by the client. Both Citus and HBase enable consistency but it penalizes performance as the number of replicas increases.

A Development of an IoT-Based Connected University System: Progress Report.

In this paper, a report on the development of an Internet of Things (IoT)-based connected university system is presented. There have been multiple smart solutions developed at the university over recent years. However, the user base of these systems is limited. The IoT-based connected university system allows for integration of multiple subsystems without the need to implement all of them in the same environment, thus enabling end-users to access multiple solutions through a single common interface. The implementation is based on microservice architecture, with the focus mainly on system robustness, scalability, and universality. In the system design, four subsystems are currently implemented, i.e., the subsystem for indoor navigation, the subsystem for parking assistants, the subsystem for smart classrooms or offices, and the subsystem for news aggregation from university life. The principles of all implemented subsystems, as well as the implementation of the system as a web interface and a mobile application, are presented in the paper. Moreover, the implementation of the indoor navigation subsystem that uses signals from Bluetooth beacons is described in detail. The paper also presents results proving the concept of the Bluetooth-based indoor navigation, taking into account different placements of nodes. The tests were performed in a real-world environment to evaluate the feasibility of the navigation module that utilizes deterministic fingerprinting algorithms to estimate the positions of users' devices.

"Just add small molecules" cell-free protein synthesis: Combining DNA template and cell extract preparation into a single fermentation.

Cell-free protein synthesis (CFPS) is a versatile biotechnology platform enabling a broad range of applications including clinical diagnostics, large-scale production of officinal therapeutics, small-scale on-demand production of personal magistral therapeutics, and exploratory research. The shelf stability and scalability of CFPS systems also have the potential to overcome cost and infrastructure challenges for distributing and using essential medical tests at home in both high- and low-income countries. However, CFPS systems are often more time-consuming and expensive to prepare than traditional in vivo systems, limiting their broader use. Much work has been done to lower CFPS costs by optimizing cell extract preparation, small molecule reagent recipes, and DNA template preparation. In order to further reduce reagent cost and preparation time, this work presents a CFPS system that does not require separately purified DNA template. Instead, a DNA plasmid encoding the recombinant protein is transformed into the cells used to make the extract, and the extract preparation process is modified to allow enough DNA to withstand homogenization-induced shearing. The finished extract contains sufficient levels of intact DNA plasmid for the CFPS system to operate. For a 10mL scale CFPS system expressing recombinant sfGFP protein for a biosensor, this new system reduces reagent cost by more than half. This system is applied to a proof-of-concept glutamine sensor compatible with smartphone quantification to demonstrate its viability for further cost reduction and use in low-resource settings.

IoT Data Sharing Platform in Web 3.0 Using Blockchain Technology

As Internet of Things (IoT)-based systems become more prevalent in the era of data-driven intelligence, they are prone to some unprecedented challenges in terms of data security and systems scalability in an era of context-sensitive data. The current advances in IoT-driven data sensing and sharing rely on third-party sources of information (TTPs) that gather data from one party, then transmit it to the other. As a result of TTPs’ involvement, such IoT systems suffer from many issues including but not limited to security, transparency, trust, and immutability as a result of the involvement of the company. Moreover, a multitude of technical impediments, such as the computation and storage poverty of IoTs, privacy concerns, and energy efficiency, enhances the challenges for IoTs. To address these issues of IoT security, we propose a blockchain-enabled open IoT data-sharing framework based on the potential of the interplanetary file system (IPFS). We have used a case study-based approach to evaluate the proposed solution. It is submitted that the proposed scenario is implemented by building smart contracts in Solidity and deploying them on the local Ethereum test network, using the Solidity programming language. With the implementation of smart contracts on the blockchain for access roles in IoT data sensing, the proposed solution advocates for a blockchain-based approach to data security for IoT systems that makes use of smart contracts for access roles.

SoK: DAG-based Blockchain Systems

Limitations on high latency and low scalability of classical blockchain systems retard their adoptions and applications. Reconstructed blockchain systems have been proposed to avoid the consumption of competitive transactions caused by linear sequenced blocks. These systems, instead, structure transactions/blocks in the form of Directed Acyclic Graph (DAG) and consequently rebuild upper layer components. The promise of DAG-based blockchain systems is to enable fast confirmation (complete transactions within million seconds) and high scalability (attach transactions in parallel) without significantly compromising security. However, this field still lacks systematic work that summarises DAG techniques. To bridge the gap, this Systematization of Knowledge (SoK) provides a comprehensive analysis of ever-existing and ongoing DAG-based blockchain systems. We abstract a general model to capture the main features and identify six types of design patterns. Then, we evaluate these systems from the perspectives of structure, consensus, property, security, and performance. We further discuss the trade-off between different factors, open challenges, and the potentiality of DAG-based solutions, indicating their promising directions for future research.

Harnessing our greatest asset: Solving margin issues by investing in frontline leaders

The rate of change in combination with the scope and magnitude and the need for increased efficacy in healthcare operations are on the rise, especially with the advent of a plethora of patient needs in the post-COVID-19 world. Senior leaders in the field are faced with various operational and financial challenges on account of the new and rapidly rising patient requirements in the healthcare sector, and, instead of optimising existing resources to meet current challenges, they are often persuaded to just add on new programmes, leaving the existing programmes to languish. Thus, to circumvent some of the major issues in healthcare operations, a more effective strategy has to be pursued than the usual one of simply buying and implementing a new productivity system with updated targets or performance benchmarks. The solution is to assess the efficacy and profitability of existing improvement programmes and resources in the advent of a new set of requirements: deploying burst versus incremental improvement; engaging middle managers to enable their solutions; and achieving organisation speed, spread and scalability across the organisation [Caldwell, C., Cook, K., (2020), ‘Achieving speed, spread, scalability, and sustainability in health systems’, American College of Healthcare Executives-Cluster Session, Clearwater, FL, 17–20th January]. This paper describes how Dignity Health East Valley was able to optimise the use of existing performance improvement resources and deploy an accelerated improvement structure to achieve rapid gains, followed by incremental improvement to both sustain and further the gains made. The approach included mentoring middle managers to become adaptive change agents and achieve 5 per cent improvement in margins in six months. The burst method described in this paper is based on the findings from the research undertaken by top-performing hospitals that identified five essential characteristics that healthcare’s frontline leaders are expected to possess and which are a clear mark of their leadership skills and competence [Caldwell, C., Cook, K., (2020), ‘Achieving speed, spread, scalability, and sustainability in health systems’, American College of Healthcare Executives-Cluster Session, Clearwater, FL, 17–20th January]. The paper presents the background to this research as well as the structure that builds these top-performer capabilities in the front-line team in order for them to lead change and drive its implementation. The positive results that emerged from the leadership training provided to frontline leaders are substantial, including improved margin, higher patient experience and higher employee engagement.

Evidence in the fog – Triage in fog computing systems

Fog computing promises improved service scalability and lower latency for IoT systems. The concept closes the gap between full computing capabilities at the network's edge and cloud systems' centrally located processing infrastructure. The drawback of the former is the high power requirements at the edge nodes, and the latter is the high latency for the data being transmitted from the edge to the cloud and back. One of the challenges for a digital forensic investigator facing a fog is the number of possible data locations, as the node functioning as the server processing data can be selected among several nodes in the network. An investigator typically has limited resources for an investigation; the more possible evidence locations, the more resources are required to collect and examine the data locations. A triage is thus needed to prioritize collecting and examining the evidence. This work analyzes measures that can identify which fog nodes are more likely to contain data, and it uses simulations to test the measures' precision and sensitivity. It aims for digital forensic investigators to maximize the utility of the available investigation resources, such that all relevant evidence is found on time.

High-Throughput FPGA-Based Inference of Gradient Tree Boosting Models for Position Estimation in PET Detectors

In modern large-scale positron emission tomography (PET) systems, transferring the digitized raw detector data at high count rates to a centralized processing unit is a challenge. Processing data on field programmable gate arrays (FPGAs) close to the detectors can reduce data early on and improve scalability of the PET system. We present and evaluate an FPGA implementation of gradient tree boosting (GTB) for 1-D position estimation of gamma interactions in the scintillator. GTB is a supervised machine learning algorithm based on building ensembles of binary decision trees. Models were trained offline and inferred in an FPGA (XC7K410T-2FFG676 Kintex-7). Input features and GTB parameters influencing both positioning performance and model size were varied while evaluating the inferred models concerning data throughput and FPGA resource consumption as well as positioning performance. We achieved throughputs per detector between <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2.94\times 10^{6}$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$4.55\times 10^{6}$ </tex-math></inline-formula> gamma interactions per second. For an optimized GTB model, resource consumption could be reduced by factors of 17 and 10 to less than 1% ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2.51\times 10^{3}$ </tex-math></inline-formula> look-up tables) of available logic and 1.26% (20 BRAMs) of memory resources, while maintaining a positioning performance of 98.63% when compared to the model with the best positioning performance. The presented framework can be easily adapted to other photosensors and scintillator influencing.

Enantioselective Friedel–Crafts Alkylation of Indoles with β,γ-Unsaturated α-Ketoesters Catalyzed by New Copper(I) Catalysts

New Cu(I) catalysts are effective in enantioselective Friedel-Crafts alkylation of a variety of indoles with different β,γ-unsaturated α-ketoesters. A control study shows that such a catalyst system is less sensitive to air, and the reactions can be carried out without special cares such as glovebox operation or moisture/oxygen-free conditions. Preliminary computation results suggest that there exists π-π stacking between the substrate and the catalyst, and such an interaction seems to play a role in stabilizing the reaction intermediate and enhancing the stereoselectivity of the reactions. The desired products can be obtained in up to 98% yield at 99% enantiomeric excess. The same high enantioselectivity can be observed when the reaction is carried in a gram scale, indicating a good scalability of the catalyst system in enantioselective Friedel-Crafts alkylation of different indoles with β,γ-unsaturated α-ketoesters.

Consensus-Based Intelligent Distributed Secondary Control for Multiagent Islanded Microgrid

Isolated microgrids (MGs) face challenges in performance stability and active/reactive power sharing as a result of frequency/voltage deviations and mismatched line impedance issues. In this paper, a consensus-based multiagent system (MAS) is proposed as a solution to restore voltage/frequency deviations and enable true power sharing. The invention of an Intelligent Distributed Secondary Control Scheme (IDSCS) can efficiently achieve hoped-for outcomes. The proposed IDSCS features estimation and compensation sublayers. For the estimation sublayer, discrete dynamic consensus algorithm-based state estimators are presented to collect average information of frequency, voltage, and reactive power. Each DG is viewed as an agent sharing information with its immediate neighbors through a sparse cyber communication network. In the compensation sublayer, online tuned proportional integral (PI) controllers using artificial neural networks (ANNs) are proposed as an intelligent voltage and frequency compensators. This combination uses the simplicity of the PI controller mathematical formula and ANN’s ability to deal with parameter variations and nonlinearity. Due to the global nature of the frequency parameter, the active power-sharing compensator is unnecessary. For compensating reactive power deviations, ANNs-based reactive power controllers are proposed. Furthermore, at the primary control level, the proposed strategy employs discrete-time proportional resonant (PR) controllers in a stationary reference frame, eliminating the need for any α β / d q or d q / α β transformations. Distributed implementation of the proposed method guarantees system scalability without MG topology or demand pattern expertise. The control scheme was validated using hypothetical MAS in MATLAB Simulink platform. The simulation findings indicate the proposed MG system can effectively distribute power among the DGs while maintaining voltage and frequency stable.

Evaluation of the Use of Cloud Computing Technology in Managing Organizational Information Systems

Cloud Computing technology has become a major focus in the development of organizational information systems, providing flexible and efficient solutions for data and application management. The aim of this research is to evaluate the use of Cloud Computing technology in managing organizational information systems with a focus on effectiveness, efficiency and security. The research method used involves literature analysis and case studies from several organizations that have adopted Cloud Computing technology. The research results show that the use of Cloud Computing significantly increases data accessibility, system scalability, and application performance. However, challenges related to data security and privacy remain a major concern, requiring the implementation of appropriate strategies for risk mitigation. In conclusion, the integration of Cloud Computing in the management of an organization's information systems can provide significant benefits, but it needs to be balanced with careful attention to security and privacy issues.

Automatic PV Array Reconfiguration under Partial Shading Conditions

Photovoltaic (PV) arrays have been recognized as one of the main clean renewable sources of electrical energy. In practice, PV arrays are subject to unavoidable non-uniform partial shading (PS) due to clouds, buildings, dust, etc. PS causes the hot spot problem, mismatch losses, and output power degradation. A widely used approach to mitigate the effect of PS is dynamic reconfiguration, where, the interconnection of PV panels is modified according to the shading condition to achieve maximum output. Dynamic reconfiguration techniques use sensors, a programmable controller, and a switch matrix for optimal operation. However, they have drawbacks such as system complexity and scalability issues. For instance, reducing the number of switches limits the flexibility of the reconfiguration algorithm. In this paper, an automatic dynamic reconfiguration scheme is proposed in which PV panel irradiation activity controls the interconnection of switches, without the need for a programmable controller. A modular building block is proposed from which easily scalable PV arrays can be constructed in a hierarchical manner. Experimental and analytic results have verified the effectiveness of the proposed reconfiguration scheme.

A high-fidelity quantum matter-link between ion-trap microchip modules

System scalability is fundamental for large-scale quantum computers (QCs) and is being pursued over a variety of hardware platforms. For QCs based on trapped ions, architectures such as the quantum charge-coupled device (QCCD) are used to scale the number of qubits on a single device. However, the number of ions that can be hosted on a single quantum computing module is limited by the size of the chip being used. Therefore, a modular approach is of critical importance and requires quantum connections between individual modules. Here, we present the demonstration of a quantum matter-link in which ion qubits are transferred between adjacent QC modules. Ion transport between adjacent modules is realised at a rate of 2424 s−1 and with an infidelity associated with ion loss during transport below 7 × 10−8. Furthermore, we show that the link does not measurably impact the phase coherence of the qubit. The quantum matter-link constitutes a practical mechanism for the interconnection of QCCD devices. Our work will facilitate the implementation of modular QCs capable of fault-tolerant utility-scale quantum computation.