Service Instances Research Articles

Trust your local scaler: A continuous, decentralized approach to autoscaling

Autoscaling is a critical component of modern cloud computing environments, improving flexibility, efficiency, and cost-effectiveness. Current approaches use centralized autoscalers that make decisions based on averaged monitoring data of managed service instances in fixed intervals. In this scheme, autoscalers are single points of failure, tightly coupled to monitoring systems, and limited in reaction times, making non-optimal scaling decisions costly. This paper presents an approach for continuous decentralized autoscaling, where decisions are made on a service instance level. By distributing scaling decisions of instances over time, autoscaling evolves into a quasi-continuous process, enabling great adaptability to different workload patterns. We analyze our approach on different abstraction levels, including a model-based, simulation-based, and real-world evaluation. Proof-of-concept experiments show that our approach is able to scale different applications deployed in containers and virtual machines in realistic environments, yielding better scaling performance compared to established baseline autoscalers, especially in scenarios with highly dynamic workloads.

Investigating Deployment Issues of DNS Root Server Instances From a China-Wide View

Investigating Deployment Issues of DNS Root Server Instances From a China-Wide View

Assuring Service Quality through SERVQUAL at the Land Bank of the Philippines North NCR Branches

This study aims to assess and assure service quality through SERVQUAL in the Land Bank of the Philippines North NCR Branches. Quantitative methods were employed, with 100 clients/customers, 216 front-liners, and 66 administrators from 82 LANDBANK North NCR Branches participating as respondents. A structured questionnaire served as the primary data collection tool. Statistical analyses included weighted mean, frequency and percentage distribution, standard deviation, Slovin’s formula, and Cronbach alpha reliability test. This study assesses the service quality perceptions among administrators, front-liners, and clients at LANDBANK. Results revealed a majority of female presence across all respondent groups. Administrators, aged 51 years and above, are mainly single and highly educated with postgraduate degrees. In contrast, majority of the front-liners and clients are aged 26-30 years and are married and hold mostly college degrees. Assessment across five dimensions of LANDBANK's service quality – tangibility, reliability, responsiveness, assurance, and empathy – revealed different perceptions. Tangibility was rated as commonly observed, with concerns about outdated equipment and system functionality. Reliability was rated higher by administrators than the front-liners and clients, issues encountered include long waiting times, system maintenance, ATM failures, and client-reported poor customer service. Administrators rated responsiveness higher than frontliners and clients, who highlighted delays and system failures. Assurance received higher ratings from administrators and front-liners but there are concerns encountered regarding hidden charges and security from clients. Empathy was consistently rated higher by administrators and front-liners compared to clients, citing instances of unsatisfactory customer service. Administrators rated LANDBANK’s service quality higher than the front-liners and clients, suggesting opportunities for enhancing alignment with customer expectations. Key issues identified include system reliability, long waiting times, and ATM failures. Recommendations focus on facility upgrades, IT enhancements, staffing increases, professionalism development, and improved customer service training to elevate service quality, customer satisfaction, and competitiveness at LANDBANK North NCR Branches.

Dynamic pickup and delivery problem for autonomous delivery robots in an airport terminal

Autonomous delivery robot (ADR) operation for short-range delivery purposes is becoming an increasingly popular mode of service. Assigning orders to robots is critical in ensuring high quality in these automated delivery operations. This paper considers a dynamic pickup and delivery problem using autonomous robots (DPDP-AR), where a fleet of ADRs picks up desired items from stores and delivers them to customers. The arrival time of orders is uncertain, with a hard delivery deadline for each order. This study considered the battery consumption of ADRs, which has been less extensively covered in previous research, and devised a battery recharging strategy for the operation. To handle this stochasticity and dynamism in the problem, we developed a reassignment algorithm that reschedules previously assigned orders at the arrival of a new order. Additionally, as periods of high demand can be estimated, peak time management of the battery is proposed to enhance ADR utilization at peak periods. Computational experiments were performed on real-world ADR food delivery service instances in an airport terminal. Test instances on various demand scenarios demonstrated improvement in service quality when devised policies were used compared to current practice. We substantiated that the proposed algorithms found efficient solutions within short computation times, validating their applicability to real-world operations.

DNS root server resolution anomaly detection

The DNS root server is at the top of the hierarchical structure of the DNS system and is the initial node that bootstraps all DNS queries. If the root server resolves abnormally, all domain name resolutions will fail, and many users cannot access the Internet. For this reason, this paper detects the root server itself resolution anomalies and non-self resolution anomalies by constructing high-confidence root zone file and anomaly judgment rules. First, we use the weighted voting statistics method to build high-confidence root zone file by calculating the confidence of multi-source root zone files. Based on high-confidence root zone files, we construct three types of anomaly detection judgment rules: (1) root-side resolution anomaly judgment rules based on feature value matching, (2) response hijacking judgment rules by correlating response anomaly features and resolution routing information, (3) root zone file synchronization anomaly judgment rules by calculating the relative synchronization delay of multi-source root zone files. Finally, using three anomaly judgment rules, we perform anomaly detection on the root resolution data obtained by active measurement. Our detection results show that root zone file synchronization delay distributions of different root server instances vary greatly. Some instances even show minute-level convergence, resulting in incorrect resolution for some TLDs. We also detect one response hijacking incident for 2 TLDs resolution, caused by the domain takeover mechanism adopted by the ISP to reduce inter-domain traffic settlement and decrease resolution latency. Except for the unresponsive exception caused by network packet loss, no root-side resolution anomaly is found, indicating that there is no artificial manipulation of TLD resolution on the root server and reflecting the responsibility of each root server operator aiming to maintain global Internet interconnection. The detection results show that the detection rules proposed in this paper can effectively achieve the anomaly detection of root server resolution and help to maintain the health and stability of the DNS system.

Key technologies of end-side computing power network based on multi-granularity and multi-level end-side computing power scheduling

With the development of modern society, business organizations have higher and higher requirements for the efficiency of cloud computing services. In order to improve the comprehensive computing capability of cloud computing network, it is very important to optimize its end-side computing power. This research takes the Hadoop platform as the computing end-side cloud computing network structure as the research object, and designs a Hadoop end-side multi-granularity and multi-level multi-level network that integrates the Graphics processing unit (GPU) and the information transfer interface (Multi Point Interface, MPI). Hierarchical computing power optimization scheduling model and improved microservice deployment s11trategy that integrates multi-level resources. The performance verification experiment results show that the mean value of all node balance ratios of the original strategy and the improved strategy on computing resource-oriented, memory resource-oriented, and disk resource-oriented microservices are 0.13 and 0.12, 0.21 and 0.17, and 0.22 and 0.19, respectively. The value of the service instance cost in the scheme using the critical path optimization scheduling strategy is always at a low level, while the instance cost value of the native strategy is significantly higher than the former. It can be seen that the end-side computing power optimization scheduling model designed in this study can indeed play a role in improving the computing performance of the end-side computing power network.

Automated management of AWS instances for training.

Amazon Web Services (AWS) instances provide a convenient way to run training on complex 'omics data analysis workflows without requiring participants to install software packages or store large data volumes locally. However, efficiently managing dozens of instances is challenging for training providers. We present a set of Bash scripts that make it quick and easy to manage Linux AWS instances pre-configured with all the software analysis tools and data needed for a course, and accessible using encrypted login keys and optional domain names. Creating over 30 instances takes 10-15minutes. A comprehensive online tutorial describes how to set up and use an AWS account and the scripts, and how to customise AWS instance templates with other software tools and data. We anticipate that others offering similar training may benefit from using the scripts regardless of the analyses being taught.

Beyond Cloud Boundaries: An Analytical Case Study on the Migration of a Modular Monolith to Azure, AWS, and Google Cloud Platform

This study primarily explores the feasibility of transferring a modular monolith from an on-premises to Google Cloud Platform (GCP) utilizing the Lift and Shift technique, with a focus on cost-effectiveness. Our objective is to identify cost-effective ways to host the application in the cloud. Our application, Hermit Portal, is built with Angular for the front end and .NET Core for the back end, with Clean Architecture serving as the primary design pattern. It examines the impact of architectural decisions on a modular monolith use case.In Section 2, we look at several cloud models, such as IaaS, PaaS, and Serverless. Our findings indicate that the IaaS model can be a very appealing option for GCP in terms of cost, with the potential to decrease costs by about 65%. We can conclude that, in terms of pricing, the GCP IaaS is the best choice based on our earlier research. We also discovered that the initial prices of PaaS can be higher due to the higher cost of SQL Server instances compared to on-premises deployments (this is also true for Amazon Web Services). To address this, we advise moving to a less expensive database (PostgreSQL or MySQL). This is made feasible by the flexibility of our architecture, which allows changes to other layers, such as databases, with relatively little influence on costs.Three cloud service providers (Amazon Web Services, Google Cloud Platform, and Microsoft Azure) will be compared in Section 3. Based on the insights gathered from this comparison, we will suggest enhancements to facilitate a seamless cloud migration for this category of applications. We will also include two essential components (Compute and Storage) that could provide useful data for a multi-cloud migration strategy or a hybrid-cloud migration strategy.

Challenges and solutions developed by the infant-toddler court teams to support child health services during the COVID-19 pandemic

This qualitative study examines Infant-Toddler Court Teams’ (ITCT) responses to the impact of the COVID-19 pandemic on the accessibility and timeliness of services for children involved with an Infant Toddler Court Team (ITCT) within the United States. ITCTs utilize collaborative practice to improve, align, and integrate systems and build community capacity to improve outcomes for very young children and their families. This study reports findings from 350 transcripts of community partner interviews and an analysis of written case notes from ITCT Community Coordinators related to more than 700 instances of child service needs. We describe challenges faced and solutions and innovations developed by ITCTs to support and maintain child health services during the first year of the pandemic, as well as findings related to challenges associated with other types of services that reduced accessibility and timeliness. Understanding the challenges experienced and solutions developed can provide the child welfare field with insight and guidance on how to approach disruptions in care in the future and the persistent lack of service providers for court involved families.

Optimized resource usage with hybrid auto-scaling system for knative serverless edge computing

In the most popular serverless platform - Knative, dynamic resource allocation is implemented using horizontal auto-scaling algorithms to create or delete service instances based on different monitored metrics. However, the assigned resources for each instance are fixed. Vertical scaling up or down assigned resources per instance is required to avoid over-provisioning resources which are limited at the edge. Hybrid (horizontal and vertical) auto-scaling solutions proposed by existing works have several limitations. These solutions are optimized for separated services and get degraded performance when applied in a normal environment with multiple concurrent services. Further, most methods make significant changes to the original Knative platform, and have not been considered to be adopted since then. In this article, instead of Knative modification, we developed separated Kubernetes operators and custom resources (CRs) that can assist the Knative auto-scaler with optimal hybrid auto-scaling configurations based on traffic prediction. First, we characterize each service with a profile of different assigned resource levels pairing with their optimal target Knative’s horizontal scaling request concurrency. Then, based on these profiles, we calculate the best-assigned resources level, target concurrency level, and the number of required instances corresponding to each future time step’s predicted traffic. Finally, these configurations are applied to Knative’s default auto-scaler and services’ CR. Experiments done on our testbed compared our solution with a Knative hybrid auto-scaler solution that does not consider the service’s target request concurrency, and the default Knative horizontal auto-scaler. The results show our solution improvements up to 14% and 20% in terms of resource usage, respectively.

The NCI Imaging Data Commons as a platform for reproducible research in computational pathology

Background and objectivesReproducibility is a major challenge in developing machine learning (ML)-based solutions in computational pathology (CompPath). The NCI Imaging Data Commons (IDC) provides >120 cancer image collections according to the FAIR principles and is designed to be used with cloud ML services. Here, we explore its potential to facilitate reproducibility in CompPath research. MethodsUsing the IDC, we implemented two experiments in which a representative ML-based method for classifying lung tumor tissue was trained and/or evaluated on different datasets. To assess reproducibility, the experiments were run multiple times with separate but identically configured instances of common ML services. ResultsThe results of different runs of the same experiment were reproducible to a large extent. However, we observed occasional, small variations in AUC values, indicating a practical limit to reproducibility. ConclusionsWe conclude that the IDC facilitates approaching the reproducibility limit of CompPath research (i) by enabling researchers to reuse exactly the same datasets and (ii) by integrating with cloud ML services so that experiments can be run in identically configured computing environments.

Boosting the visibility of services in microservice architecture

Monolithic software architectures are no longer sufficient for the highly complex software-intensive systems, which modern society depends on. Service Oriented Architecture (SOA) surpassed monolithic architecture due to its reusability, platform independency, ease of maintenance, and scalability. Recent SOA implementations made use of cloud-native architectural approaches such as microservice architecture, which has resulted in a new challenge: the discovery difficulties of services. One way to dynamically discover and route traffic to service instances is to use a service discovery tool to locate the Internet Protocol (IP) address and port number of a microservice. In the event that replicated microservice instances are found to provide the same function, it is crucial to select the right microservice that provides the best overall experience for the end-user. Parameters including success rate, efficiency, delay time, and response time play a vital role in establishing a microservice’s Quality of Service (QoS). These assessments can be performed by means of a live health-check service, or, alternatively, by making a prediction of the current state of affairs with the application of machine learning-based approaches. In this research, we evaluate the performance of several classification algorithms for estimating the quality of microservices using the QWS dataset containing traffic data of 2505 microservices. Our research also analyzed the boosting algorithms, namely Gradient Boost, XGBoost, LightGBM, and CatBoost to improve the overall performance. We utilized parameter optimization techniques, namely Grid Search, Random Search, Bayes Search, Halvin Grid Search, and Halvin Random Search to fine-tune the hyperparameters of our classifier models. Experimental results demonstrated that the CatBoost algorithm achieved the highest level of accuracy (90.42%) in predicting microservice quality.

Stateless Q-learning algorithm for service caching in resource constrained edge environment

In resource constrained edge environment, multiple service providers can compete to rent the limited resources to cache their service instances on edge servers close to end users, thereby significantly reducing the service delay and improving quality of service (QoS). However, service providers renting the resources of different edge servers to deploy their service instances can incur different resource usage costs and service delay. To make full use of the limited resources of the edge servers to further reduce resource usage costs, multiple service providers on an edge server can form a coalition and share the limited resource of an edge server. In this paper, we investigate the service caching problem of multiple service providers in resource constrained edge environment, and propose an independent learners-based services caching scheme (ILSCS) which adopts a stateless Q-learning to learn an optimal service caching scheme. To verify the effectiveness of ILSCS scheme, we implement COALITION, RANDOM, MDU, and MCS four baseline algorithms, and compare the total collaboration cost and service latency of ILSCS scheme with these of these four baseline algorithms under different experimental parameter settings. The extensive experimental results show that the ILSCS scheme can achieve lower total collaboration cost and service latency.

MagicScaler: Uncertainty-Aware, Predictive Autoscaling

Predictive autoscaling is a key enabler for optimizing cloud resource allocation in Alibaba Cloud's computing platforms, which dynamically adjust the Elastic Compute Service (ECS) instances based on predicted user demands to ensure Quality of Service (QoS). However, user demands in the cloud are often highly complex, with high uncertainty and scale-sensitive temporal dependencies, thus posing great challenges for accurate prediction of future demands. These in turn make autoscaling challenging---autoscaling needs to properly account for demand uncertainty while maintaining a reasonable trade-off between two contradictory factors, i.e., low instance running costs vs. low QoS violation risks. To address the above challenges, we propose a novel predictive autoscaling framework MagicScaler , consisting of a Multi-scale attentive Gaussian process based predictor and an uncertainty-aware scaler. First, the predictor carefully bridges the best of two successful prediction methodologies---multi-scale attention mechanisms, which are good at capturing complex, multi-scale features, and stochastic process regression, which can quantify prediction uncertainty, thus achieving accurate demand prediction with quantified uncertainty. Second, the scaler takes the quantified future demand uncertainty into a judiciously designed loss function with stochastic constraints, enabling flexible trade-off between running costs and QoS violation risks. Extensive experiments on three clusters of Alibaba Cloud in different Chinese cities demonstrate the effectiveness and efficiency of MagicScaler , which outperforms other commonly adopted scalers, thus justifying our design choices.

Orchestrating distributed 5G edges for automotive cross-border trials: Validation of an experimental prototype

The automotive industry requires ultra-reliable low-latency connectivity for its vehicles, and as such, it is one of the promising customers of 5G ecosystems and their orchestrated network infrastructure. In particular, Multi-Access Edge Computing (MEC) provides moving vehicles with localized low-latency access to service instances. However, given the mobility of vehicles, and various resource demand patterns at the distributed MEC nodes, challenges such as fast reconfiguration of the distributed deployment according to mobility pattern and associated service and resource demand need to be mitigated. In this paper, we present the orchestrated edges platform, which is a solution for orchestrating distributed edges in complex cross-border network environments, tailored to Connected, Cooperative, and Automated Mobility (CCAM) use cases within a 5G ecosystem. The proposed solution enables collaboration between orchestrators that belong to different tiers, and various federated edge domains, with the goal to enable service continuity for vehicles traversing cross-border corridors. The paper presents the prototype that we built for the H2020 5G-CARMEN trials, including the validation of the orchestration design choices, followed by the promising results that span both orchestration (orchestration latency) and application performance-related metrics (client-to-edge and edge-to-edge service data plane latencies).

Budgeted Edge Service Selection in Mobile Edge Computing Environment

Mobile edge computing allows end-users to retrieve required services from edge servers with low network latency. However, when massive service instances with the same or interchangeable functionalities are available at the network edge, how to select appropriate service instances for use to minimize the overall latency perceived is a critical problem. This is not trivial due to that: 1) usually multiple service instances are involved to collectively fulfil an end-user's ever-increasing complicated need; 2) the overall cost incurred should not exceed end-user's service recruiting budget. In this article, we formally define the budgeted edge service selection (BESS) problem. Then, we model the BESS problem as a constrained optimization problem and theoretically prove its <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathcal {NP}$</tex-math></inline-formula> -hardness. Next, to solve the BESS problem effectively, we propose two approaches named BESS-O and BESS-A, respectively. Specifically, BESS-O is an optimal approach based on nonlinear integer programming. It can find out optimal solutions to BESS problems. Moreover, BESS-A is a greedy approach. It can find out approximate BESS solutions with higher efficiency, which is more useful for dealing with large-scale BESS problems. Finally, we conduct comprehensive experiments to evaluate our approaches against five representative approaches. Experimental results demonstrate that given a specific service usage budget, our approaches proposed in this article can achieve the highest completion ratio and the lowest service delay.

Changes in the epidemiological characteristics of prehospital emergency services before and during the COVID-19 pandemic, Chengdu, 2016–2021

The coronavirus disease 2019 (COVID-19) brought a healthy crisis all around the world. It’s not only affected healthcare utilization, but also some disease’s incidence. We collected pre-hospital emergency data in Chengdu from January 2016 to December 2021, and studied the demand for emergency medical services (EMSs), emergency response times (ERTs), and the disease spectrum in the city proper of Chengdu as a whole. A total of 1,122,294 prehospital emergency medical service (EMS) instances met the inclusion criteria. Under the influence of COVID-19, notably in 2020, the epidemiological characteristics of prehospital emergency services in Chengdu were considerably altered. However, as the pandemic was brought under control, they returned to normal or even earlier in 2021.The COVID-19 pandemic had an effect on people's medical care-seeking behavior and the performance of prehospital emergency services. Although the indicators associated with prehospital emergency services eventually recovered as the epidemic was brought under control, they remained marginally different from before the outbreak.

The association between the Patient Protection and Affordable Care Act and healthcare affordability among US adults with intellectual disability.

Historically, US adults with intellectual disability (ID) experience worse healthcare access than the general population. However, the implementation of the Patient Protection and Affordable Care Act (ACA) may have reduced disparities in healthcare access. Using a pre-ACA 2011-2013 sample and a post-ACA implementation 2014-2016 sample from the National Health Interview Survey data, we examined the association between the ACA's introduction and healthcare access among adults with ID (N=623). Negative binomial regression models were used to test the association between the ACA and the total number of foregone healthcare services. Binary logistic regression was used to explore whether the ACA's implementation was associated with the increased likelihood of possessing health insurance as well as the decreased likelihood of any and particular measures of foregone healthcare services due to cost. The study provides evidence that the ACA's implementation was associated with the decreased likelihood of the total number and any foregone care services owing to cost. Findings also revealed that the ACA's implementation was associated with expansion of health insurance coverage and decreasing instances of foregone care services for medical care, dental care, specialist visit and mental care among adults with ID. However, persons with ID were still at a higher risk of foregone prescription medicines, follow-up medical care and eyeglasses due to cost in the post-ACA years. The study provides evidence that healthcare access among Americans with ID improved after the ACA's implementation. However, challenges in access to follow-up care, eyeglasses and prescription medicines persist and require policy solutions, which extend beyond the ACA's provisions.

Exploitation of steroids in coronavirus disease 2019: An overview

AbstractCorticosteroids, more specifically glucocorticoids, are one of the most prescribed drugs. Corticosteroids are adrenal hormones that serve significant physiologic activities such as modulating glucose metabolism, protein catabolism, calcium metabolism, bone turnover control, immunosuppression, and downregulation of the inflammatory cascade. Corticosteroids are regarded as life-saving due to their various effects and have been used therapeutically to treat a broad range of autoimmune, rheumatologic, inflammatory, neoplastic, and viral illnesses. However, the therapeutic benefits of glucocorticoids are restricted by their adverse effects. The most serious side effects of corticosteroids are associated with the use of higher doses for longer periods and over the counter medicine availability in specific pharmacies, which leads to dependency, as well as its usage in mild and moderate server instances, which is contrary to guidelines. In recent times, the use of corticosteroids has been multiplied with the emergence of the coronavirus disease 2019 (COVID-19) pandemic. WHO and the standard guidelines have recommended the usage of corticosteroids in critically ill COVID-19 patients, but their usage in mild and moderate cases caused more harm than benefit. This illicit usage has resulted in the development of opportunistic fungal illnesses such as mucormycosis, posing an extra risk to patients in terms of quality of life and finances. Other adverse effects of systemic corticosteroids include morphological changes, increased blood sugar levels, delayed wound healing, infections, decreased bone density, truncal obesity, cataracts, glaucoma, blood pressure abnormalities, and muscle fiber atrophy. In this review, we want to discuss the significance and detrimental effects of corticosteroids emphasizing on recent times, that is, COVID-19. Anabolic-androgenic steroids are the best-studied class of APEDs (appearance and performance-enhancing drugs). Anabolic steroids increase lean muscle mass when used in conjunction with weight training. The aim, for nonathlete weightlifters is typically the improvement of appearance. The vast majority of people who misuse steroids are male nonathlete weightlifters in their 20s or 30s.

Collaborative Orchestration of Multi-Domain Edges From a Connected, Cooperative and Automated Mobility (CCAM) Perspective

The 5G ecosystem is comprised of the cellular 5G System, as well as a managed and orchestrated infrastructure providing virtualized network and service functions. The automotive industry with its stringent requirements for connected vehicles is a promising and yet challenging consumer of such 5G ecosystem. Deployment of service instances at distributed cloud resources of cellular network infrastructure edges enables localized low-latency access to these services from moving vehicles but comes along with challenges, such as the need for fast reconfiguration of the distributed deployment according to mobility pattern and associated service and resource demand. In this paper, we investigate a solution for the collaborative orchestration of services for Connected, Cooperative and Automated Mobility (CCAM) within such 5G ecosystem. A key objective is the service continuity for a highly dynamic automotive scenario, achieved by the associated management and orchestration of these services in distributed edge clouds. The proposed solution leverages a multi-tier orchestration system as well as localized management and protocol operations for collaborative edge resources. By means of both analytical and experimental evaluations, the paper draws conclusions on the gain in accelerating orchestration decisions and enforcements, while balancing associated protocol and computational load over the highly distributed and multi-layered orchestration system.