Concurrent Users Research Articles

The Associations of Diuretics and Laxatives Use with Cardiovascular Mortality. An Individual Patient-Data Meta-analysis of Two Large Cohort Studies

PurposeTo investigate the associations of diuretics overall, non-potassium-sparing diuretics in specific, and laxative use with cardiovascular mortality (CVM) in subjects with antihypertensive treatment.MethodsAnalyses included 4253 participants, aged 50 to 75 years, from the German ESTHER cohort and 105,359 participants, aged 50 to 69 years, from the UK Biobank. Cox proportional hazard regression models were applied in both studies, and then results were pooled using random-effects model meta-analyses.ResultsDuring 14 and 7 years of follow-up, 476 and 1616 CVM cases were observed in the ESTHER study and the UK Biobank, respectively. Compared to non-users, a 1.6-fold (hazard ratio [95% confidence interval] 1.57 [1.29; 1.90]), a 1.4-fold (1.39 [1.26; 1.53]), and no statistically significantly increased (1.13 [0.94; 1.36]) CVM were observed in users of diuretics overall, non-potassium-sparing diuretics in specific, and laxatives, respectively. Concurrent use of non-potassium-sparing diuretics and laxatives was associated with a 2-fold increased CVM (2.05 [1.55; 2.71]) when compared to users of neither diuretics nor laxatives. However, a test for interaction slightly missed statistical significance (p = 0.075).ConclusionsThese consistent results from two large cohort studies imply that more research is needed on the safety of diuretics in routine care. Although not statistically significant in this study, a drug-drug interaction of non-potassium-sparing diuretics and laxatives appears plausible. Physicians and pharmacists are advised to clarify additional laxative use in users of non-potassium-sparing diuretics and inform about the risk of concurrent use. Moreover, closer potassium monitoring intervals (e.g., every 3 months) might be indicated in concurrent users to prevent fatal cardiovascular events.

Do menstrual cups increase risk of IUD expulsion? A survey of self-reported IUD and menstrual hygiene product use in the United States

Purpose: To gather preliminary data about menstrual hygiene product use and associated risk of intrauterine device (IUD) expulsion.Materials and methods: We conducted an Internet-based survey using the Amazon Mechanical Turk (MTurk) platform and Researchmatch.org. Eligible participants were women at least 18 years old and self-identified as current or previous IUD users.Results: Of 902 survey responses, 71% reported current or previous IUD use and 19.7% reported menstrual cup use. IUD users were significantly more likely than non-IUD users to use a menstrual cup or tampons. Among all IUD users, 56 individuals reported experiencing at least one expulsion (8.8%). We found a positive association between concurrent menstrual cup use and IUD expulsion (OR: 2.75, 95% CI: 1.40–5.42, p = .002), but no association with concurrent tampon or pad use. Of concurrent IUD and cup users who experienced an expulsion, one reported it occurring ‘while using a menstrual cup’, with others reporting the expulsion during other events.Conclusions: Menstrual cup use may be higher than previously reported, particularly in IUD users. Concurrent menstrual cup use may increase the risk of IUD expulsion. Prospective research is necessary to fully explore the relationship between menstrual hygiene product use and IUD expulsions.

New Coprescription of Opioids and Benzodiazepines and Mortality Among Veterans Affairs Patients With Posttraumatic Stress Disorder.

Opioids and benzodiazepines are commonly coprescribed medications. The mortality risk associated with their concurrent use is unknown. To estimate the all-cause mortality risk for patients newly prescribed opioids and benzodiazepines concurrently relative to patients prescribed benzodiazepines only, opioids only, or neither medication. This propensity score-matched, retrospective, cohort study included 17,476 patients receiving Veterans Affairs (VA) health care between October 1, 2009, and September 30, 2011, and diagnosed with posttraumatic stress disorder identified using ICD-9-CM code 309.81. One-year total and cause-specific mortality was assessed by hazard ratios and subhazard ratios, adjusted for propensity score, age, baseline psychiatric and medical comorbidity, and daily medication dose. Concurrent users (n = 4,369) were propensity score matched 1:1 with benzodiazepine-only users, opioid-only users, and nonusers. One year after medication start, the concurrent cohort had higher rates of all-cause mortality (116 deaths) relative to benzodiazepine-only (75 deaths; adjusted hazard ratio = 1.52; 95% CI, 1.14-2.03), opioid-only (67 deaths; 1.76; 95% CI, 1.32-2.35), and nonuser (60 deaths; 1.85; 95% CI, 1.30-2.64) cohorts. Risk of overdose death was greater among patients in the concurrent cohort relative to patients in the benzodiazepine-only (adjusted subhazard ratio = 2.59; 95% CI, 1.00-6.66), opioid-only (2.58; 95% CI, 1.09-6.11), and nonuser (9.16; 95% CI, 2.27-37.02) cohorts. For circulatory disease-related deaths, the adjusted subhazard ratio for concurrent medication users was 1.81 (95% CI, 1.01-3.24) relative to nonusers. New coprescription of opioids and benzodiazepines was associated with increased all-cause mortality and overdose death compared with new prescription of benzodiazepines only, opioids only, or neither medication and increased circulatory disease-related death relative to neither medication.

Guidepost: Scalable MU-MIMO User Selection via Indirect Channel Orthogonality Evaluation

Multi-user MIMO (MU-MIMO) can serve multiple users concurrently, and is the key technology to enable ultra-high-speed wireless access. However, in practice MU-MIMO networks are far from their full potential due to the poor scalability problem, including high computational complexity at PHY layer and large-overhead channel contention at MAC layer. Moreover, cross-cell interference among multiple MU-MIMO cells also counteracts network performance. In this paper, we perform a systematic study on MU-MIMO and propose a fully scalable MU-MIMO user selection protocol called Guidepost. In contrast with previous works, Guidepost builds on a novel principle of indirection channel orthogonality evaluation, so as to decouple and simplify the complicated computational/contention interaction among users. Based on the principle, Guidepost first achieves scalable MU-MIMO user selection with only linear computational complexity. Second, Guidepost realizes distributed user selection through a two-dimensional prioritized contention mechanism, which can single out the best concurrent users efficiently by utilizing both the time and frequency domain resources. Third, Guidepost incorporates a lightweight AP-assisted contention mechanism to handle cross-cell interference in distributed MU-MIMO (netMIMO) where users are widely distributed and cannot sense each other. Software-radio based implementation and experimentation show that Guidepost significantly outperforms state-of-the-art methods under various traffic patterns and node mobility.

Serverless Computing: An Investigation of Deployment Environments for Web APIs

Cloud vendors offer a variety of serverless technologies promising high availability and dynamic scaling while reducing operational and maintenance costs. One such technology, serverless computing, or function-as-a-service (FaaS), is advertised as a good candidate for web applications, data-processing, or backend services, where you only pay for usage. Unlike virtual machines (VMs), they come with automatic resource provisioning and allocation, providing elastic and automatic scaling. We present the results from our investigation of a specific serverless candidate, Web Application Programming Interface or Web API, deployed on virtual machines and as function(s)-as-a-service. We contrast these deployments by varying the number of concurrent users for measuring response times and costs. We found no significant response time differences between deployments when VMs are configured for the expected load, and test scenarios are within the FaaS hardware limitations. Higher numbers of concurrent users or unexpected user growths are effortlessly handled by FaaS, whereas additional labor must be invested in VMs for equivalent results. We identified that despite the advantages serverless computing brings, there is no clear choice between serverless or virtual machines for a Web API application because one needs to carefully measure costs and factor-in all components that are included with FaaS.

Psychological correlates and binge drinking behaviours among Canadian youth: a cross-sectional analysis of the mental health pilot data from the COMPASS study.

ObjectiveThe objective of this study was to examine associations between depression, anxiety and binge drinking among a large sample of Canadian youth, while testing the moderating effect of flourishing. This research uses data from the Cannabis, Obesity, Mental health, Physical activity, Alcohol, Smoking, Sedentary Behaviour (COMPASS) study (2012–2021) with a large sample size collecting data on youth health behaviours within Canadian secondary schools.DesignCross-sectionalSetting14 secondary schools across Ontario and British Columbia, Canada.ParticipantsA sample of grade 9–12 students (n=6570) who participated in the Mental Health pilot of the COMPASS studyPrimary and secondary outcome measuresSelf-reported questionnaires assessed student binge drinking behaviours (5≥drinks), symptoms of depression (Center for Epidemiologic Studies Depression Scale (Revised)−10 scores≥10) and anxiety (Generalised Anxiety Disorder 7-item Scale scores≥10), and flourishing (Diener’s Flourishing Scale: 8–40).ResultsIn our sample of 6570 students, 37.0% of students reported binge drinking in the last year, and 41.4% and 31.7% of students report clinically-relevant symptoms of depression and anxiety, respectively. Anxiety (adjusted OR (AOR): 0.57, (99% CI 0.15 to 2.22)) and depression (AOR: 1.98, (99% CI 0.76 to 5.13)) symptoms were not found to be associated with binge drinking and we did not detect any moderating role of flourishing. Rather, factors that were associated with increased odds of binge drinking included sports team participation (AOR: 1.67, (99% CI 1.37 to 2.03)) and use of other substances (tobacco (AOR: 3.00, (99% CI 2.12 to 4.25)) and cannabis (AOR: 7.76, (99% CI 6.36 to 9.46))). Similar associations were found for frequency of binge drinking.ConclusionsConsistent with existing literature, binge drinking behaviours were problematic, as well as clinically-relevant symptoms of depression and anxiety. However, mental health problems and well-being may not be responsible for explaining patterns of binge drinking in youth. Targeted intervention efforts towards student athletes and concurrent substance users are necessary for addressing binge drinking in youth populations.

Modelling the region‐based VoLTE cell capacity estimation using resource optimisation

In this study, the authors introduce analytical models for region-based Voice over Long Term Evolution (VoLTE) cell capacity estimation. Using the authors' proposed analytical models, they estimate cell capacity in different regions by using several user-specific voice traffic characteristics such as Voice Activity Factor (VAF) and Silent Insertion Descriptor (SID), and different advanced techniques like Packet bundling and Transmission Time Interval (TTI) bundling. It also incorporates Hybrid automatic repeat request (HARQ) retransmission. In this work, the cell is divided into multiple regions, each corresponding to the area in which user equipment generates a certain level of Channel Quality Indication (CQI) value based on Signal to Interference plus Noise Ratio (SINR). This proposed model for region-based VoLTE capacity estimation optimises the network resource utilisation and enhances the number of concurrent active VoLTE users within the cell in each region. In this work, they have defined packet bundling and TTI bundling factors for appropriate regions to optimally allocate the radio resources for capacity estimation. Numerical results obtained using their proposed model considering different radio specific parameters are found improving the VoLTE cell capacity significantly. Further, the impact of HARQ on region-based VoLTE cell capacity is also obtained.

Prevalence, attitudes, behaviours and policy evaluation of midwakh smoking among young people in the United Arab Emirates: Cross-sectional analysis of the Global Youth Tobacco Survey.

IntroductionNon-cigarette tobacco products are an increasing public health concern globally. Little is known about midwakh, a pipe indigenous to the United Arab Emirates (UAE). This study aimed to assess the prevalence, attitudes, behaviours and policy evaluation of midwakh smoking among 13 to 15 year olds in the UAE.MethodsWe conducted secondary analyses of the 2013 UAE Global Youth Tobacco Survey. The main three outcomes were ever use, current use (past-30 days), and the number of midwakhs smoked per day. We assessed cessation, attitude, and policy measures. Regression models identified the association between each outcome measure and sex, school grade, nationality, weekly spending money, cigarette use, and parent and peer tobacco use.ResultsThe prevalence of ever and current midwakh use were 18.5% and 9.0%, respectively. Daily midwakh users smoked a median of 8.0 per day while non-daily users smoked 3.8 per month. Higher midwakh prevalence was reported among wealthier males, older age groups, concurrent cigarette users and among participants having peers or parents who use tobacco. There was also variation by nationality. Reduced harm perception was greater among midwakh users than non-users. About 39.6% reported being declined a midwakh purchase due to age, and 35.5% reported noticing health warnings on packages.ConclusionsMidwakh use is prevalent among 13 to 15 year olds in the UAE, and burden lies mainly with daily users. Further needed research should not delay implementation and evaluation of policies known to curb tobacco use among youth, including taxation, media campaigns, and provision of cessation services.

A new classification system for describing concurrent use of nicotine vaping products alongside cigarettes (so-called 'dual use'): findings from the ITC-4 Country Smoking and Vaping wave 1 Survey.

To determine whether a simple combination of level of smoking and level of vaping results in a useful typology for characterizing smoking and vaping behaviours. Cross-sectional data from adults (≥18years) in the 2016 wave 1 ITC Four Country Smoking and Vaping Survey in the United States (n=2291), England (n=3591), Australia (n=1376) and Canada (n=2784) were used. Participants who either smoked, vaped or concurrently used both at least monthly were included and divided into eight groups based on use frequency of each product (daily, non-daily, no current use). This resulted in four concurrent use groups (predominant smokers, dual daily users, predominant vapers and concurrent non-daily users). These groups were compared with each other and with the four exclusive use groups, on socio-demographics, nicotine dependence, beliefs and attitudes about both products, and quit-related measures using data weighted to reference population surveys in each country. Of the sample, 10.8% were concurrent users, with daily smokers vaping non-daily (predominant smokers), constituting 51.6% of this group. All eight categories differed from other categories on at least some measures. Concurrent daily nicotine users reported higher levels of indicators of nicotine dependence, and generally more positive attitudes toward both smoking and vaping than concurrent non-daily users. Among daily nicotine users, compared with exclusive daily smokers, reports of interest in quitting were higher in all concurrent use groups. Dual daily users had the most positive attitudes about smoking overall, and saw it as the least denormalized, and at the same time were equally interested in quitting as other concurrent users and were most likely to report intending to continue vaping. In Australia, Canada, England and the United States in 2016, daily nicotine users differed considerably from non-daily nicotine users. Among daily nicotine users, dual daily users (those who smoke and vape concurrently) should be treated as a distinct grouping when studying relationships between smoking and vaping. The eight-level typology characterizing concurrent and exclusive use of smoking and vaping should be considered when studying both products.

ROBOTIC PROCESS AUTOMATION IN SOFTWARE PERFORMANCE TESTING WORKLOAD MODELING

Software Performance Testing(SPT) is the kind of Non-Functional Testing(NFT) to prove the to be delivered system by the project team/product team is working as expected with respect to system health, consistency, endurance robustness etc. This testing is a very critical part of system going live. Now a days customers/stakeholders are very keen to carry on this test to get absolute confidence to the software product they are going to use for their business. It has been seen several instances in the history of software industry ,live application has been demoted or scrapped due to the system not able to behave as it is expected with its features and functionalities when the system is used concurrently in high load .This type of verification can be only simulated when we do performance testing. Now workload distribution for performance testing till date has been generally performed by manually capturing load details like list of very frequently performed critical transactions, no. of call/hours, no. of concurrent users running from the log or details from business. And it’s a laborious time consuming job. With the advent of Robotic Process Automation (RPA) we can overcome the drawbacks of manual hurdles workload model with a futuristic reusable and reliable one, about which I discussed in following sections.

Performance bottleneck analysis and resource optimized distribution method for IoT cloud rendering computing system in cyber-enabled applications

This paper analyzes current cloud computing, cloud rendering industry, and related businesses. In this field, cloud system performance lacks unified evaluation criterion. A novel analysis method and a related measure of cloud rendering system performance are presented in this paper. The main paper investigates the number of system concurrent users and average response delay about user access, average frame speed, system operation speed, and average response about user browsing system. This paper makes a theoretical analysis of a core business process about cloud rendering system, multi-task rendering processes especially. This paper analyzes the efficiency, average frame rate, rendering performance bottleneck of the cloud rendering system, and put forward a unique parameter adjustment strategy to improve system performance, by optimizing related server and rendering machine configuration. This paper puts forward a method to reduce the bottleneck of the system and prevent system performance deterioration in the new scheme. This paper puts forward a set of system optimization strategies to improve system performance. This is a new cloud rendering system performance optimization configuration scheme and optimization strategy.

Study on Integrated Communication Enhanced Link Based on Signal Detection, Frequency Conversion and Forwarding

Remote mountainous areas have complex terrain, so the information transmission capacity of communication links have been greatly restricted. In addition, the maritime areas have the problem of limited communication base stations and limited communication links as well. Currently, the most commonly used air-based platforms for spectrum forwarding, which only support a small coverage of areas, fail to achieve wide-area or cross-regional signal connectivity, and only support a limited number of concurrent users. Therefore, this paper proposes an integrated communication enhanced link based on frequency detection, conversion and forwarding, which solves the problem that information sharing is not stable. And the simulation results verify that the proposed integrated communication enhanced link addresses the problem of multiple users (up to 150) to communicate concurrently in a cross-region area (radius 50 km) which may have mountains in between. And also, the proposed air-based platform only requires a low power of energy, so that the platform can work longer time in the air.

The Relationship Between Social Anxiety and Alcohol and Marijuana Use Outcomes Among Concurrent Users: A Motivational Model of Substance Use.

College students with more social anxiety symptoms are particularly vulnerable to problematic alcohol and marijuana use given their susceptibility for elevated anxiety symptoms in social settings combined with the normative nature of substance use. Existing research has established substance use as coping motivated for these students when examining alcohol and marijuana use problems separately. The next step is to determine whether students with more social anxiety who use both substances do so for similar or different reasons. The current study tested a comprehensive (i.e., all variables in the same model) motivational model of alcohol/marijuana use in a sample of college students from 10 universities across the United States who endorsed both past-month alcohol and marijuana use. College students were recruited through psychology department participant pools and completed an online survey assessing mental health symptoms, substance use motives, and substances use behaviors. Current sample comprised concurrent alcohol/marijuana users (n=2,034), 29.6% of whom endorsed clinically indicated levels of social anxiety and nearly one-fourth exceeded the cutoff for hazardous drinking (23.2%) and hazardous marijuana use (21.9%). Across both substances, coping motives significantly mediated the positive relationship between social anxiety symptoms and substance use problems. Unique to alcohol, conformity motives mediated the association between social anxiety symptoms and alcohol-related problems. Taken together, students with more social anxiety who are focused on anxiety management may use either alcohol or marijuana; however, these students may experience more alcohol-related problems when drinking to fit in with peers.

Smart-Edge-CoCaCo: AI-Enabled Smart Edge with Joint Computation, Caching, and Communication in Heterogeneous IoT

The development of mobile communication technology, hardware, distributed computing, and artificial intelligence (AI) technology has promoted the application of edge computing in the field of heterogeneous Internet of Things (IoT). In order to overcome the defects of the traditional cloud computing model in the era of big data. In this article, we first propose a new AIenabled smart edge with heterogeneous IoT architecture which combines edge computing, caching, and communication. Then, we propose the Smart-Edge-CoCaCo algorithm. To minimize total delay and confirm the computation offloading decision, Smart-Edge-CoCaCo uses joint optimization of the wireless communication model, the collaborative filter caching model in edge cloud, and the computation offloading model. Finally, we built an emotion interaction testbed to perform computational delay experiments in real environments. The experiment results showed that the computation delay of the Smart-Edge-CoCaCo algorithm is lower than that of traditional cloud computing model with the increase of computing task data and the number of concurrent users.

Multi-User Redirected Walking and Resetting Using Artificial Potential Fields.

Head-mounted displays (HMDs) and large area position tracking systems can enable users to navigate virtual worlds through natural walking. Redirected walking (RDW) imperceptibly steers immersed users away from physical world obstacles allowing them to explore unbounded virtual worlds while walking in limited physical space. In cases of imminent collisions, resetting techniques can reorient them into open space. This work introduces categorically new RDW and resetting algorithms based on the use of artificial potential fields that "push" users away from obstacles and other users. Data from human subject experiments indicate that these methods reduce potential single-user resets by 66% and increase the average distance between resets by 86% compared to previous techniques. A live multi-user study demonstrates the viability of the algorithm with up to 3 concurrent users, and simulation results indicate that the algorithm scales efficiently up to at least 8 users and is effective with larger groups.

Pilot Study - Composite Risk Score for Pre-Transplant Risk Assessment in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Objective An internal pre-transplant risk assessment tool or Risk Monitor was developed by the Baylor Blood & Marrow Transplant (BMT) Data Management and Analytics team to assist with optimization of patient selection. The tool assigns a total risk score for allogeneic transplants based on attributes such as disease status, disease risk index (DRI), patient age, Karnofsky performance status (KPS), hematopoietic cellular therapy co-morbidity index (HCT-CI), and CMV status, along with psychosocial factors such as caregiver support, coping skills or motivation, health literacy, medication compliance, socioeconomic viability, distance of residence from the transplant center, as well as others. A retrospective pilot study was conducted to predict which attributes have the highest correlation with the primary outcome variable, post-transplant patient mortality. Method The Risk Monitor was developed using Microsoft Access (Microsoft®, Redmond, WA). A score of 0 (no risk) or 1 (risk) was assigned for each variable and an aggregate score was calculated for each patient (Figure 1). The data was analyzed using Waikato Environment for Knowledge Analysis (Weka). Weka is an open-source data mining tool consisting of a collection of machine learning algorithms for data mining tasks. It contains tools for data preparation, classification, regression, clustering, association rules mining, and visualization. Results We evaluated the total risk score for 97 patients that had undergone an allogeneic stem cell transplant (SCT) at Baylor University Medical Center in 2016. The total risk score for the patients varied between 0 and 9. Weka provided us with a list of independent attributes with rankings based on the intensity of correlation with the dependent variable, early mortality. Evaluation of the data set suggests that one attribute (Total Score) has the highest correlation with mortality, while other individual attributes (KPS, CMV status, DRI, HCT-CI and psychosocial factors such as caregiver support, coping skills, and health literacy) demonstrated a moderate to high correlation (Figure 2). Conclusion The Risk Monitor can serve as a useful tool for screening adult patients prior to allogeneic transplant. In addition, the tool may help to identify patients who need additional screening and social support to minimize the possibility of early mortality after transplantation. The Access database has been transitioned to a web-based (SQL Server/PHP/JavaScript) application to overcome certain Access-related issues such as client software installation, file corruption and interface update issues. Furthermore, this will enable easy cross platform usage, enhance security, and allow more concurrent users.

The Relationship Between Risk Factors and Alcohol and Marijuana Use Outcomes Among Concurrent Users: A Comprehensive Examination of Protective Behavioral Strategies

Among college samples, both alcohol and marijuana protective behavioral strategies (PBS) have been shown to mediate the effects of known risk factors (i.e., sex, age at substance use onset, college substance use beliefs, substance use motives, and impulsivity-like traits) on alcohol and marijuana outcomes. However, it is unknown whether PBS use would operate similarly for both substances among concurrent users. The present study examined which risk factors relate to alcohol/marijuana outcomes via alcohol/marijuana PBS use among a large group of concurrent alcohol/marijuana users. Participants were college students who consumed both alcohol and marijuana at least 1 day in the previous month (n = 2,034; 69.08% female). Across both substances, PBS use significantly mediated the effects of sex (women reported higher PBS use), age at first use (having an older age at first use was associated with more PBS use), and college substance use beliefs (higher beliefs was associated with lower PBS use). Unique to alcohol outcomes, alcohol PBS use significantly mediated the effects of negative urgency, social motives, and enhancement motives (all associated with lower PBS use). Unique to marijuana outcomes, marijuana PBS use mediated the effects of coping, expansion (both associated with lower PBS use), and conformity motives (associated with more PBS use). Our results suggest that PBS for both alcohol and marijuana can help explain why some risk factors are associated with alcohol/marijuana outcomes. Taken together, PBS use seems to be an important intervention target for alcohol/marijuana concurrent users.

Interference Mitigation for 5G Millimeter-Wave Communications

We present a novel ultra-wideband low-cost transmit/receive architecture with reduced power and hardware requirements, which includes interference mitigation capabilities and accommodates multiple users. Specifically, a novel encoding approach is introduced to multiplex signals across large bandwidths and achieve secure communication and interference mitigation through spread signals. In this paper, a detailed description of the proposed high data rate system architecture is presented. Simulations are also provided in the presence of high-power interferers with eight concurrent users. To assess the performance and impact of the encoding/decoding process, a noise analysis was conducted for the transmitter and receiver chains. Specifically, bit error rate curves are generated at different stages of the system. The measurements show interference margin ratios improvement by at least 16 dB.

Syntix: A Profiling Based Resource Estimator for CUDA Kernels

Trending applications such as AI and data analytics have mandated the use of GPUs in modern datacenters for performance reasons. Current practice dictates to dedicate GPUs to applications, which limits the amount of concurrent users to the available GPUs. That use of GPUs contradicts with the policy of datacenters to oversubscribe resources and accommodate as many user applications as possible. To address this issue, providers will inevitably resort to GPU sharing. In this work we introduce Syntix, a mechanism that we deploy on GPU sharing system and 1) profiles CUDA kernels in order to learn their resource requirements in terms of threads and blocks and 2) assigns those resources to kernels in order to be efficiently collocated into streams. Syntix is able to exploit the resources that are possibly wasted from the execution of an individual kernel and save the 80% of them on average.

A New Dynamic Load Balancing Algorithm for Multi-ROIA

Real-time Online Interactive Application (ROIA) is an emerging class of large-scale distributed application which can support millions of concurrent users around the world. Due to the dynamic changes in the number of concurrent users as well as the uncertainty of user operations, the dynamic load balancing is a key issue for ROIA. However, most of previous works are dedicated to the load balancing in a single ROIA without considering the variety of different type ROIAs. We take the advantage of differences between ROIAs and propose a new load balancing algorithm for multi-ROIA to improve the scalability of ROIA and increase the resource utilization of system. This paper firstly describes the motivation of the new load balancing algorithm, then presents the dynamic load balancing algorithm for multi-ROIA. Finally, the simulation results are also presented to show the efficiency and feasibility of the new algorithm.