User Response Time Research Articles

Cooperative Computation Offloading in FiWi Enhanced 4G HetNets Using Self-Organizing MEC

Multi-access edge computing (MEC) is an emerging paradigm to meet the rapidly growing computation demands of mobile applications. This paper investigates the performance gains of cooperative computation offloading for MEC enabled FiWi enhanced HetNets with capacity-limited backhaul links. After presenting the envisioned two-tier MEC architecture for a FiWi based networking infrastructure, we propose a simple but efficient offloading strategy, which relies on the flexible trilateral cooperation between end-device, edge servers, and the remote cloud. We then present an analytical framework to estimate the average response time and energy consumption of mobile users for various offloading scenarios with different wireless access modes (i.e., WiFi and 4G LTE-A). The presented analysis flexibly allows for incorporating both offloaded and conventional human-to-human (H2H) traffic of mobile users as well as fixed (wired) subscribers. Finally, we present our self-organization based mechanism, which enables mobile users to make suitable energy-delay trade-offs by jointly minimizing the average task execution time and energy consumption, using only their local information. The obtained results demonstrate the feasibility of the proposed cooperative self-organizing offloading strategy and its superior performance over schemes with MEC- or cloud-only offloading strategies.

PRIMA: Subscriber-Driven Interference Mitigation for Cloud Services

Network services, e.g., video streaming services, are increasingly being deployed on public cloud platforms. Such services often employ horizontal scaling where a group of resource instances, e.g., virtual machines (VMs), handle incoming workload. The response time of such services is often affected by interference, i.e., contention among resource instances belonging to multiple cloud subscribers for shared cloud resources. Most commercial cloud platforms do not support built-in mechanisms to detect interference and mitigate its impact. Consequently, subscribers of such platforms, i.e., network service providers, need to deploy their own mechanisms to ensure a specified end user response time target is continuously met even in the face of fluctuations in workload and interference. This paper describes PRIMA, our implementation of such a mechanism. PRIMA uses automated and controlled performance tests to build models that capture the joint impact of workload and interference on the response time of each resource instance employed by a service. It adapts the system to changing workload and interference conditions by using these models at runtime to control the number of instances in the system and the distribution of load among these instances. Unlike existing subscriber-oriented interference mitigation techniques in literature, PRIMA guarantees that a subscriber-specified response time threshold is satisfied at every resource instance assigned to a service. Furthermore, in contrast to these approaches PRIMA can help a subscriber avoid using more instances than necessary by automatically selecting at runtime the least number of instances required for handling the observed workload and interference. We experimentally validate the effectiveness of PRIMA in both private and public cloud environments. Results show that PRIMA outperforms competing approaches proposed by us and others, including those that are commonly used in practice. They also reveal that PRIMA can automatically calibrate its models at runtime to account for any model prediction errors.

Helmet Prototype Response Time Assessment using NCAA Division 1 Collegiate Football Athletes

Background: With advances in concussion research, an increasing amount of resources are being allocated to advancing football helmet technology. Objective: This study assesses the claim that a new modified helmet prototype provides greater field of view for the user as compared to a commonly worn helmet by players. Method: The helmets—Riddell SpeedFlex and the modified helmet—were compared based on user response time while performing a response test task using the FITLIGHT Trainer system, actual helmet field of view blockage, users’ subjective perception of field of view, and balance tests. Eighteen National Collegiate Athletic Association (NCAA) Division 1 American football student-athletes completed the response test task and questionnaire. Results: The results demonstrate evidence that the SpeedFlex helmet provided by the equipment staff significantly increases wearers’ response times, F(2,20) = 5.646, p < 0.05. Also, while the quantification of the field of view perception was similar across helmet types, the student-athlete participants perceived the modified helmet to have significantly more field of view while performing the response test task, 1.56 v. 2.56; p < 0.05 for frontal vision and 2.83 v. 5.39; p < 0.05 for peripheral vision. Conclusion: In addition to the findings, this study also lays out a response time test protocol using the FITLIGHT Trainer system that can be used for assessment of response time testing of football and other helmets in future studies.

Types and Frequency of Infusion Pump Alarms and Infusion-Interruption to Infusion-Recovery Times for Critical Short Half-Life Infusions: Retrospective Data Analysis.

BackgroundAlarm fatigue commonly leads to a reduced response to alarms. Appropriate and timely response to intravenous pump alarms is crucial to infusion continuity. The difficulty of filtering out critical short half-life infusion alarms from nonurgent alarms is a key challenge for risk management for clinicians. Critical care areas provide ample opportunities for intravenous medication error with the frequent administration of high-alert, critical short half-life infusions that require rigorous maintenance for continuity of delivery. Most serious medication errors in critical care occur during the execution of treatment, with performance-level failures outweighing rule-based or knowledge-based mistakes.ObjectiveOne objective of this study was to establish baseline data for the types and frequency of alarms that critical care clinicians are exposed to from a variety of infusion devices, including both large volume pumps and syringe drivers. Another objective was to identify the volume of these alarms that specifically relate to critical short half-life infusions and to evaluate user response times to alarms from infusion devices delivering these particular infusions.MethodsThe event logs of 1183 infusion pumps used in critical care environments and in general care areas within the European region were mined for a range of alarm states. The study then focused on a selection of infusion alarms from devices delivering critical short half-life infusions that would warrant rapid attention from clinicians in order to avoid potentially harmful prolonged infusion interruption. The reaction time of clinicians to infusion-interruption states and alarms for the selected critical short half-life infusions was then calculated.ResultsInitial analysis showed a mean average of 4.50 alarms per infusion in the general critical care pump population as opposed to the whole hospital rate of 1.39. In the pediatric intensive care unit (PICU) group, the alarms per infusion value was significantly above the mean average for all critical care areas, with 8.61 alarms per infusion. Infusion-interruption of critical short half-life infusions was found to be a significant problem in all areas of the general critical care pump population, with a significant number of downstream (ie, vein and access) occlusion events noted. While the mean and median response times to critical short half-life infusion interruptions were generally within the half-lives of the selected medications, there was a high prevalence of outliers in terms of reaction times for all the critical short half-life infusions studied.ConclusionsThis study gives an indication of what might be expected in critical care environments in terms of the volume of general infusion alarms and critical short half-life infusion alarms, as well as for clinician reaction times to critical short half-life infusion-interruption events. This study also identifies potentially problematic areas of the hospital for alarm fatigue and for particular issues of infusion and infusion-line management. Application of the proposed protocols can help create benchmarks for pump alarm management and clinician reaction times. These protocols can be applied to studies on the impact of alarm fatigue and for the evaluation of protocols, infusion-monitoring strategies, and infusion pump-based medication safety software aimed at reducing alarm fatigue and ensuring the maintenance of critical short half-life infusions. Given the frequency of infusion alarms seen in this study, the risk of alarm fatigue due to the white noise of pump alarms present in critical care, to which clinicians are constantly exposed, is very high. Furthermore, the added difficulties of maintaining critical short half-life infusions, and other infusions in specialist areas, are made clear by the high ratio of downstream occlusion to infusion starts in the neonatal intensive care unit (NICU). The ability to quantitatively track the volume of alarms and clinician reaction times contributes to a greater understanding of the issues of alarm fatigue in intensive care units. This can be applied to clinical audit, can allow for targeted training to reduce nuisance alarms, and can aid in planning for improvement in the key area of maintenance of steady-state plasma levels of critical short half-life infusions. One clear conclusion is that the medication administration rights should be extended to include right maintenance and ensured delivery continuity of critical short half-life infusions.

An Analysis on ECSE Load Balancing Algorithm with CDC as a Service Broker Policy in Cloud with Heterogeneous Communication and Device Characteristics

Load balancing algorithms and service broker policies plays a crucial role in determining the performance of cloud systems. User response time and data center request servicing time are largely affected by the load balancing algorithms and service broker policies. Several load balancing algorithms and service broker polices exist in the literature to perform the data center allocation and virtual machine allocation for the given set of user requests. In this paper, we investigate the performance of equally spread current execution (ESCE) based load balancing algorithm with closest data center(CDC) service broker policy in a cloud environment that consists of homogeneous and heterogeneous device characteristics in data centers and heterogeneous communication bandwidth that exist between different regions where cloud data centers are deployed. We performed a simulation using CloudAnalyst an open source software with different settings of device characteristics and bandwidth. The user response time and data center request servicing time are found considerably less in heterogeneous environment.

Analysis of Heterogeneous Device Characteristics in Round Robin Based Load Balancing Algorithm with Closest Data Center as Service Broker Policy in Cloud

Load balancing is an important aspect in cloud to share load among different virtual machines running on various physical nodes. The user response time which is an important performance metric is being highly influenced by the efficient load balancing algorithm for cloud data centers. Virtual machines which are part of the cloud data centers consist of various types of physical devices. The user response time is affected significantly by the capacity of physical devices that exist as part of the data centers. Several load balancing algorithms exist in the literature to allocate task effectively on various virtual machines running in data centers. We investigate the performance of round robin based load balancing algorithm with closest data center as service broker policy in cloud data centers. We have performed a simulation with data centers that consist of devices with different physical characteristics such as memory, storage, bandwidth, processor speed and scheduling policy using Round Robin load balancing algorithm with closest data centers as service broker policy. We present the merits of heterogeneous device characteristics in reducing the user response time and the data center request service time. We used Cloud Analyst, an open source simulation tool for cloud computing environment.

Development of an Automated 30 Second Chair Stand Test Using Smartwatch Application.

This paper presents development of the smartwatch application for automation of the standard 30 Second Chair Stand Test (30SCST). 30SCST is primarily used to test leg strength and endurance, but also speed and mobility and assess risk of falls. We use inertial signals on smartwatch to detect and count stands during the test. The application notifies the user to start and stop the test using vibration on the smartwatch. Synchronization of notifications and signal acquisition allows assessment of user's response time during the test. Our application monitors baseline heart rate before the test, heart rate increase during the test, and heart rate recovery after the test that might allow assessment of cardiovascular fitness of the user. The application is developed using Wear OS and tested on two smartwatch platforms: Fossil G4 and Polar M600. Pilot test included 12 subjects, six male and six female (mean age 39.1, S.D. 19 years). Overall accuracy of detection of the number of standups is 98.8%. Smartwatch application can be used for automated testing in clinical setups as well as for self-monitoring at home.

Distributed cooperative computation offloading in multi-access edge computing fiber–wireless networks

Novel 5G applications will require the processing of large amounts of computation-intensive, delay-sensitive tasks. Mobile devices, however, are subject to limited hardware resources (e.g., battery life, CPU, and storage) and therefore may fall short of satisfying the stringent latency requirements of complex tasks. To fill the gap between the increasing demands of computation-intensive, delay-sensitive tasks driven by emerging 5G applications and the availability of limited resources on smart mobile devices, we introduce our proposed architecture of multi-access computing (MEC) enabled Fiber–Wireless (FiWi) LTE enhanced HetNets, enabling the coexistence and cooperation between conventional (remote) cloud and MEC servers. After presenting our analytical framework to evaluate the energy-delay performance, we propose our distributed cooperative offloading strategy with the objective of minimizing the average response time for mobile users.

Cost Optimization for the Coupled Video Delivery Networks

Video delivery networks (VDNs) are one of the key enablers for alleviating contradiction between demands increasing for online video and response time of users accessing video content. However, existed works mainly adopt either proactive or reactive delivery scheme, which leads to the problems of backbone traffic overloaded and delays increased. In these works, the impact of performance indicators, such as bandwidth, delay, and personalized demand on delivery cost are not comprehensively taken into account. To overcome these problems, we propose a cost optimization for coupled video delivery model (CO-CVDM), which combines coupled proactive delivery and coupled reactive delivery method to minimize delivery cost in terms of bandwidth consumption, delay performance, personalized demand and construct a multicast delivery tree for the video delivery. The coupled delivery cost minimization problem under constraint conditions is formulated. Since this problem is NP-hard and is prohibitively difficult to solve, we develop a heuristic multicast delivery tree (HMDT) algorithm to provide more optimized approximate cost. Based on this, we extend CO-CVDM to distribute video in a dynamic scenario. Specifically, with dynamic change of user's requirement for video, the process of video delivery is truncated into a sequence of static graphs and the change of interest attenuation is modeled with human forgetting curve. Finally, the numerical simulations are conducted to validate the advantage of CO-CVDM in terms of optimization delivery cost, and the delivery cost is inversely proportional to the attenuation of user interest.

A Fast Approach to Scale Up Disk Arrays With Parity Declustered Data Layout by Minimizing Data Migration

Parity declustering is widely deployed in erasure-coded storage systems so as to provide fast recovery and high data availability. However, to perform scaling on such redundant array of inexpensive disks (RAIDs), it is necessary to preserve parity declustered data layout so as to preserve the properties after scaling. Unfortunately, existing scaling algorithms fail to achieve this goal so they cannot be applied for scaling RAIDs with parity declustering. To address this challenge, we develop an efficient online scaling scheme called parity declustering scaling (PDS), which employs an auxiliary balanced incomplete block design to define the data migration so as to preserve parity declustered data layout. Furthermore, PDS can also be applied to scale RAIDs for improving reliability and/or storage efficiency as options by allocating more parity blocks and/or data blocks in stripes. We provide theoretical proofs to formally show that PDS preserves parity declustered data layout, and achieves uniform distributions of data and parity blocks after scaling while requiring only the minimal data migration. We implement PDS in Linux kernel 3.14.72 and evaluate its performance with real-world traces. The results show that PDS can reduce 82.37 percent of scaling time and 18.25 percent of user response time during scaling on average, compared with “moving-everything” round-robin approach adapted to achieve parity declustered data layout after scaling.

184. Implementation and Evaluation of an Electronic Antimicrobial Prior Authorization Approval Process at a Large Academic Medical Center

BackgroundPrior authorization (PA) of antimicrobial agents is recommended by CDC as a core antimicrobial stewardship (AMS) intervention and has been a cornerstone of the Hospital of the University of Pennsylvania (HUP) AMS program since its inception. Previously, prescribers called the PA pager, waited for a callback, discussed the case with the AMS team, and received approval or an alternative recommendation. The objective of this study was to implement a new electronic prior authorization approval platform within existing clinical decision support software (CDSS) and evaluate its utilization.MethodsElectronic PA was introduced to HUP prescribers in January 2017. At the point of order entry for a restricted agent, an approver’s name is a required field. In the CDSS, the request is initiated by selecting the patient and indication for use. The next screen displays additional risk factors to select and auto-populates the patient’s current and historic cultures. The requestor selects the relevant cultures and desired antimicrobial(s), enters contact information, and submits the request. Review and approval by AMS team are completed in the CDSS, often without requiring a phone call. Pharmacy views a log of requests to determine whether the name given on the order as approver is true. Process metrics were evaluated using a retrospective cohort study from January 2017 through December 2017.ResultsThere were 437 unique users of the PA system. Uptake over time is displayed in Figure 1. 1,934 unique patients had 3,329 requests submitted. The most frequent indications were for prophylaxis, other, and pneumonia. Levofloxacin was most commonly requested (36.76%, 1,297 of 3,528), followed by meropenem, caspofungin, and fluconazole. 88.7% (2,952 of 3,329) of overall requests approved. The frequency of PA requests by hour is presented in Figure 2. During AMS hours, the median time to response was 18 (8–42) minutes. 18.4% (649 of 3,528) of requests were submitted during off hours.Figure 1.Cumulative total of new requestorsFigure 2.Frequency of PA requests by hourConclusionBased on user uptake and response times, electronic PA was successfully implemented at HUP. Software to facilitate PA shows promise to assist in facilitating PA and tracking relevant process metrics.Disclosures J. Mehta, ILUM Health Solutions: Consultant, Salary. J. Olson, ILUM Health Solutions: Consultant, Salary. W. J. Ketcherside, ILUM Health Solutions: Consultant, Salary. L. Hunt, ILUM Health Solutions: Consultant, Salary. B. Palermo, ILUM Health Solutions: Employee, Salary.

A Comparison of Predictive Spatial Augmented Reality Cues for Procedural Tasks.

Previous research has demonstrated that Augmented Reality can reduce a user's task response time and mental effort when completing a procedural task. This paper investigates techniques to improve user performance and reduce mental effort by providing projector-based Spatial Augmented Reality predictive cues for future responses. The objective of the two experiments conducted in this study was to isolate the performance and mental effort differences from several different annotation cueing techniques for simple (Experiment 1) and complex (Experiment 2) button-pressing tasks. Comporting with existing cognitive neuroscience literature on prediction, attentional orienting, and interference, we hypothesized that for both simple procedural tasks and complex search-based tasks, having a visual cue guiding to the next task's location would positively impact performance relative to a baseline, no-cue condition. Additionally, we predicted that direction-based cues would provide a more significant positive impact than target-based cues. The results indicated that providing a line to the next task was the most effective technique for improving the users' task time and mental effort in both the simple and complex tasks.

Supporting Continuous Skyline Queries in Dynamically Weighted Road Networks

The paper focuses on the design of an optimum method for handling the continuous skyline query problem in road networks. Existing studies on processing the continuous skyline query focus exclusively on static road networks, which are limited because the state of roads in road networks is constantly changing. Therefore, to apply current methods for dynamically weighted road networks, a distributed skyline query method based on a grid partition method has been proposed in this paper. The method adopts the concepts of a distributed computing framework and road network preprocessing computations in which multiple parallel computing nodes are allocated and organized in grids. Using this approach, the road network map is simplified to a hub graph with much smaller scale such that the query load of the central node can be significantly reduced. The theoretical analysis and experimental results both indicate that, by using the proposed method, the system can achieve quick response time for users as well as a good balance between response times and accuracy. Therefore, it can be concluded that using the proposed method is beneficial for handling continuous skyline queries in a dynamically weighted road network.

SDN-Based Load Balancing Service for Cloud Servers

With the continuous growth, heterogeneity, and ever increasing demand of services, load balancing of cloud servers is an emerging challenge to meet highly demanding requirements (e.g., data rates, latency, quality of service) of 5G network applications. Although various load balancing techniques have been proposed, some of these techniques either require installation of dedicated additional load balancers for each service, or manual reconfiguration of the device to handle new services is desired. These techniques are expensive, time-consuming, and impractical. Moreover, most of the existing load balancing schemes do not consider service types. This article presents an SDN-based load balancing (SBLB) service for cloud servers to maximize resource utilization and minimize response time of users. The constituents of the proposed scheme are an application module that runs on top of an SDN controller and server pools that connect to the controller through OpenFlow switches. The application module consists of a service classification module, a dynamic load balancing module, and a monitoring module. The controller handles all messages, manages host pools, and maintains the load of host in real time. Experimental results validate the performance of the proposed scheme. Through experimental results, SBLB demonstrates significant decrease in average response time and reply time.

Real-world large-scale study on adaptive notification scheduling on smartphones

Human attention has bottlenecked today’s ubiquitous computing environment where users are consuming increasing amounts of information from numerous applications and services. Since the system-to-user provision of information is becoming more proactive, mainly via push notifications that often cause interruption at the users’ side, attention management is becoming very important. Despite the many existing studies concerned with detecting opportune moments to present such push information to the users (in a way that preserves the user’s attention and lowers their cognitive load and frustration), there is little evaluation of such systems in the real-world production environments. Overlooked areas of study also include the examination of real users and notification contents. In this paper, we present various results from the first extensive evaluation on user’s interruptibility and engagement in the real-world environment with a market-leading smartphone application that boasts a large number of users, including real notification content on the application. Following our previous mobile-sensing and machine learning-based interruptibility estimation approach, which was an effective study in its own right (Okoshi et al., 2015 [18,19]), we embedded a logic with the same approach in the “Yahoo! JAPAN” Android app (one of the most popular applications on the national market). The results from our large-scale in-the-wild user study (that included more than 680,000 users and spanned three weeks) indicate that, in most cases, delaying the notification delivery until an interruptible moment is detected is beneficial to users. The practice results in significant reduction of user response time (49.7%) when compared to delivering the notifications immediately. We observed a higher number of notifications opened in our system and constant improvement in user engagement levels throughout the entire study period. We also observed differences in click rates among different days and time among users with different attributes (e.g., age, gender, and occupation). Additional evaluation of our revised system, which can train and distribute different models for weekdays and weekends, improved the click rates during weekends. This negated the performance degradation previously observed during weekends.

Towards optimal scheduling policy for heterogeneous memory architecture in many-core system

With the advent of Intels second-generation many-core processor (Knights Landing: KNL), high-bandwidth memory (HBM) with potentially five times more bandwidth than existing dynamic random-access memory has become available as a valuable computing resource for high-performance computing (HPC) applications. Therefore, resource management schemes should now be able to consider existing central processing unit cores, conventional main memory, and this newly available HBM to improve the overall system throughput and user response time. In this paper, we present our profiling mechanism and related scheduling policy that analyzes the resource usage patterns of various HPC workloads. By carefully allocating memory-intensive workloads to HBM in KNL, we show that the overall performance of multiple message passing interface workloads can be improved in terms of the execution time and system utilization. We evaluate and verify the effectiveness of our scheme for optimizing the use of HBM by using NAS Parallel Benchmarks.

Latency Aware and Service Delay with Task Scheduling in Mobile Edge Computing

In a traditional Mobile Cloud Computing (MCC), a stream of data produced by mobile users (MUs) is uploaded to the remote cloud for additional processing throughout the Internet. Though, due to long WAN distance it causes high End to End latency. With the intention of minimize the average response time and key constrained Service Delay (network and cloudlet Delay) for mobile users (MUs), offload their workloads to the geographically distributed cloudlets network, we propose the Multi-layer Latency Aware Workload Assignment Strategy (MLAWAS) to allocate MUs workloads into optimal cloudlets, Simulation results demonstrate that MLAWAS earns the minimum average response time as compared with two other existing strategies.

Towards effective scheduling policies for many-task applications: Practice and experience based on HTCaaS

Summary In this paper, we conduct a comparative study of relatively simple yet effective scheduling policies for many-task applications where multiple users with varying numbers of tasks are actively sharing a common system infrastructure. We have implemented three different scheduling mechanisms that can address fairness and user response time respectively in a common middleware stack called HTCaaS, which is a pilot-job–based multi-level scheduling system running on top of production-level clusters. As a representative case of our many-task applications, we have leveraged the virtual screening application, which is a computational technique used in drug discovery process to select the most promising candidate drugs for in vitro testing from millions of chemical compounds, which typically requires a substantial amount of computing resources and efficient processing of docking simulations. Our comparative experimental results of different scheduling policies show how we can effectively support multiple users in a shared resource environment by balancing between user satisfaction and overall system performance, provide guidelines to improve system utilization, and address additional technical challenges to support various many-task applications.

Latency Aware Workload Offloading in the Cloudlet Network

We propose a novel network architecture by leveraging the cloudlet concept, the software defined networking technology, and the cellular network infrastructure to bring the computing resource to the mobile edge. In order to minimize the average response time for mobile users (MUs) in offloading their application workloads to the geographically distributed cloudlets, we propose the latency-aware workload offloading (LEAD) strategy to allocate MUs’ application workloads into suitable cloudlets. Simulation results demonstrate that LEAD incurs the lowest average response time as compared with two other existing strategies.

Capacity planning of the registration server in cloud storage

Cloud storage is one of the key services in cloud computing. In order to appeal more users to the cloud storage, free cloud experience is provided to potential users. A registration server is set specially for intentional users who are satisfied with the free experience and want to register as very important persons (VIPs). In order to mathematically evaluate the system performance and optimally plan the registration server, we establish a queueing network-based system model. We analyse the steady-state distribution of the system model and derive the performance measures of the system in terms of the average response time of potential users and the utilisation of registration server. For balancing different performance measures, we establish a cost function and develop an optimisation algorithm to plan the capacity of the registration server with a reasonable service rate.