Buffer Resources Research Articles

Design and Implementation of Burst Buffer Over-Subscription Scheme for HPC Storage Systems

Burst Buffer is widely used in supercomputer centers to bridge the performance gap between computational power and the high-performance I/O systems. The primary role of Burst Buffer is to temporarily absorb the bursty I/O and reduce the heavy access on Parallel File System (PFS). However, the job resource manager on High-Performance Computer (HPC) systems prefers to use a dedicated Burst Buffer allocation approach, which eventually leads to the severely underutilized Burst Buffer resource. To improve the efficiency of using the expensive Burst Buffer resource, we analyze the I/O patterns on Burst Buffer in depth. We propose Burst Buffer over-subscription allocation method, which improves Burst Buffer utilization by allowing each job to access Burst Buffer only during its I/O phases so that the jobs can overlap each other. Furthermore, we develop a new I/O congestion-aware scheduler and a transparent data management system between Burst Buffer and PFS. Our approach also reduces the memory overhead and improves the data persistence of the data management system by adapting the persistent memory. With the proposed approach, not only the Burst Buffer utilization can be improved, but also HPC applications can achieve high I/O performance by exploiting the powerful Burst Buffer hardware capabilities. Experimental results show that BBOS can improve Burst Buffer utilization by up to 120% while more stable and higher checkpoint performance is guaranteed even under high I/O loads compared to other state-of-the-art schedulers. Besides, our approach can improve the hit ratio of restart requests by up to 96.4% and provides up to 210% higher restart throughput on Burst Buffer.

Combined Distributed Shared-Buffered and Diagonally-Linked Mesh Topology for High-Performance Interconnect.

Networks-on-Chip (NoCs) have become the de-facto on-chip interconnect for multi/manycore systems. A typical NoC router is made up of buffers used to store packets that are unable to advance to their desired destination. However, buffers consume significant power/area and are often underutilized, especially in cases of applications with non-uniform traffic patterns thus leading to performance degradation for such applications. To improve network performance, the Roundabout NoC (R-NoC) concept is considered. R-NoC is inspired by real-life multi-lane traffic roundabouts and consists of lanes that are shared by multiple input/output ports to maximize buffering resource utilization. R-NoC relies on router-internal adaptive routing that decides the lane path based on back pressure. Back pressure makes it possible to assess lane utilization and route packets accordingly. This is made possible thanks to the use of elastic buffers for control flow, a clever type of handshaking in a way similar to asynchronous circuits. Another prominent feature of R-NoC is that internal routing and arbitration are completely distributed which allows for significant freedom in deciding internal router topology and parameters. This work leverages this property and proposes novel yet unexplored configurations for which an in-depth evaluation of corresponding implementations on 45 nm CMOS technology is given. Each configuration is evaluated performance and power-wise on both synthetic and real application traffic. Several R-NoC configurations are identified and demonstrated to provide very significant performance improvements over standard mesh configurations and a typical input-buffered router, without compromising area and power consumption. Exploiting the distributed nature of R-NoC routers, a diagonally-linked configuration is then proposed which incurs moderate area overhead and features yet better performance and energy efficiency.

Traffic-Aware Buffer Management in Shared Memory Switches

Switch buffer serves an important role in the modern internet. To achieve efficiency, today’s switches often use on-chip shared memory. Shared memory switches rely on buffer management policies to allocate buffer among ports. To avoid waste of buffer resources or excessive buffer occupation by a few ports, existing policies tend to maximize overall buffer utilization and pursue queue length fairness. However, blind pursuit of utilization and misleading fairness definition based on queue length lead to buffer occupation with no benefit to throughput but extends queuing delay and undermines burst absorption of other ports. With analysis of current dynamic threshold policies, we demonstrate that meaningless buffer occupation can potentially impair the absorption capability of shared buffer, whereas none of the existing policies have addressed this problem. We contend that a buffer management policy should proactively detect port traffic and adjust buffer allocation accordingly. In this paper, we propose Traffic-aware Dynamic Threshold (TDT) policy. On the basis of the classic dynamic threshold policy, TDT proactively raises or lowers port threshold to absorb burst traffic or evacuate meaningless buffer occupation. We present detailed designs of port control state transition and state decision module that detect real-time traffic and change port thresholds accordingly. Simulation and DPDK-based real testbed demonstrate that TDT simultaneously optimizes for throughput, loss and delay, and reduces up to 50% flow completion time.

Responding to Inconsistent Performance Feedback on Multiple Goals: The Contingency Role of Decision Maker's Status in Introducing Changes

We advance performance feedback theory by showing how decision makers respond to inconsistent performance feedback on multiple goals, and how status shapes these responses for introducing behavioral changes. We argue that inconsistent performance feedback on primary and secondary goals decrease decision makers' propensity to initiate changes, while their high status operates both as a critical asset that motivates solution search for fixing a performance shortfall on a primary goal and as a resource buffer against self-enhancement beliefs when addressing inconsistent feedback on multiple goals. We test our arguments using a dataset of 107,791 Airbnb properties and find support for our hypotheses. We discuss the contributions of our findings to the performance feedback theory.

Humor in the sky: the use of affiliative and aggressive humor in cabin crews facing passenger misconduct

PurposeThe authors present a fist attempt to test the mediating role of humor in the relation between unruly passenger behavior and occupational stress in cabin crews.Design/methodology/approachThis study used an experience sampling design to investigate the relationship between a frequent job hassle in air service provision, namely unruly passenger behavior (UPB), and the stress experienced by flight attendants.FindingsThe results of multilevel analyses show that UPB is positively related to the use of aggressive humor and negatively related to the use of affiliative humor in cabin crews. Moreover, humor mediates the relationship between unruly passenger behavior and stress. In addition the results show that general self-efficacy as a personal resource buffers the association between passenger misconduct and the use of aggressive humor.Originality/valueThis study is among the first empirical attempts to explore the role of humor as a mediator between uncivil customer behavior and stress in air service employee.

Access to resources buffers against effects of current reproduction on future ability to provide care in a burying beetle.

Studies investigating the trade‐off between current and future reproduction often find that increased allocation to current reproduction is associated with a reduction in the number or quality of future offspring. In species that provide parental care, this effect on future offspring may be mediated through a reduced future ability to provide care. Here, we test this idea in the burying beetle Nicrophorus vespilloides, a species in which parents shift the cost of reproduction toward future offspring and provide elaborate parental care. We manipulated brood size to alter the costs females experienced in association with current reproduction and measured the level of parental care during a subsequent breeding attempt. Given that these beetles breed on carcasses of small vertebrates, it is important to consider confounding effects due to benefits associated with resource access during breeding. We, therefore, manipulated access to carrion and measured the level of parental care during a subsequent breeding attempt. We found that females provided the same level of care regardless of previous brood size and resource access, suggesting that neither affected future ability to provide care. This may reflect that parents feed on carrion during breeding, which may buffer against any costs of previous breeding attempts. Our results show that increased allocation to current reproduction is not necessarily associated with a reduction in future ability to provide care. Nevertheless, this may reflect unique aspects of our study system, and we encourage future work on systems where parents do not have access to a rich resource during breeding.

No more career resilient and passionate! A longitudinal experimental insight into work overload

Complex and excess work demands are a byproduct of corporate model in higher educational settings, however, how work overload manifests in long-run career outcomes is yet to be examined. We therefore built on the ‘primacy of resource loss’ corollary of the Conservation of Resources theory to find out the repercussions of work overload on career resilience via reduced harmonious passion. We conducted a longitudinal natural field experiment of 402 faculty members (N=198 in the standardized group (optimal load) and N= 204 in the non-standardized (overload) group working in Higher Educational Institutions in Pakistan. Data were collected in three waves (six months apart). The analysis was performed through group comparisons, trend analysis, and longitudinal mediation analysis, using SPSS and MPlus. Our findings suggest that, as opposed to the much-held conception of harmonious passion and career resilience as buffering resources to sustain work overload, these rescores are also depleted by the same phenomenon. Based on these results, we suggest that faculty members should be given optimal workload to retain their passion and resilience for teaching.

Does Technostress Increase R&D Employees' Knowledge Hiding in the Digital Era?

Technostress as an antecedent factor exploring knowledge hiding continues to be seldomly discussed in the digital era. Based on the job demand-resource theory, this article introduces work exhaustion as a mediator variable and constructs a model that the five sub-dimensions of technostress (i.e., overload, invasion, complexity, insecurity, and uncertainty) affect knowledge hiding for R&D employees. Similarly, this study analyzes the moderation of workplace friendship as the resource buffering effect. Based on data from the 254 questionnaires of the two-stage survey, empirical results show that: (1) Techno-invasion, techno-insecurity, and techno-complexity have significant positive effects on work exhaustion, and techno-invasion has the greatest effect. However, techno-overload and techno-uncertainty have no significant relationship with work exhaustion. (2) Work exhaustion plays a mediating role in the relationships between the three aspects of technostress (techno-invasion, techno-insecurity, techno-complexity) and knowledge hiding; However, its mediating effects are insignificant in the relationships between the two aspects of technostress (techno-overload and techno-uncertainty) and knowledge hiding. (3) Workplace friendship negatively moderates the relationships between the two aspects of technostress (techno-invasion and techno-insecurity) and work exhaustion, leading to less knowledge hiding. Nonetheless, its negative moderation for the relationships between the two aspects of technostress (techno-overload and techno-uncertainty) and work exhaustion are insignificant. Empirical results further show that workplace friendship positively moderates the relationship between techno-complexity and work exhaustion.

Learning a Healthy Rhythm: An Intervention to Increase Children's Resources for Stress Management.

This article describes a pilot test of a community engaged, culturally relevant, arts-based intervention. The purpose was to increase children's personal protective buffering resources. Protective buffering resources help children cope with stressful stimuli, reduce activation of their systemic stress response, mitigate allostatic load, and promote optimal health. The "Learning a Healthy Rhythm" intervention included a stress management component and an ongoing Afro-Latino percussion program for 18 children ages 9-11. The stress management component included educational content about stress, self-assessment of stress symptoms, and stress management techniques. A mixed-method intervention evaluation design was used. Qualitative data, quantitative data, and biometrics including hair cortisol were collected. Six intervention parameters were evaluated: effectiveness, fidelity, feasibility, acceptability, necessity, and safety. Positive results were obtained for all parameters. Reduction in physiological and subjective measures of stress was evident. This stress management intervention was well-received and supported by participants.

Heterogeneous reconfigurable architectures for machine learning dataflows

AbstractThis work explores the placement and routing of machine learning applications' dataflow graphs on different heterogeneous coarse‐grained reconfigurable architectures (CGRA). We analyze three different types of processing element (PE) heterogeneity, the first concerning the interconnection pattern, the second being on the kind of operations a single PE can execute, and the last concerning the PE buffer resources. This analysis aim to propose a fair reduction to the overall cost in comparison to the homogeneous CGRA architecture. We compare our results with the homogeneous case and one of the state‐of‐the‐art tools for placement and routing (P&R). Our algorithm executed, on average, 52% faster than VPR 8.1 (Versatile Place and Route), which is an open‐source academic tool designed for the FPGA placement and routing phases, reaching better mapping in 66% of cases and achieving the same results in 26% of cases. Furthermore, a heterogeneous architecture reduces the cost without losing performance in 76% of the cases considering multiplier heterogeneity. We propose a novel heterogeneous buffer architecture that minimizes the buffer resources by 56.3% for K‐means dataflow patterns. We also show that a heterogeneous border chess architecture outperforms a homogeneous one. In addition, our mapping reaches optimal instances of single tree dataflows compared to classical Lee/Choi and H‐trees.

Design of fully adaptive routing and hybrid VC allocation in wireless NOC

Wireless network-on-chip (WiNoC) establishes a long-range single-hop wireless links on the base of the traditional planar metal interconnects network, which improve network performance in several aspects. The establishment of wireless link improves the diversity of path for packet transmitting, while simultaneously aggravates the deadlock problem, leading it to more complicated condemn. The traditional routing mechanism of wireless network on chip can avoid deadlock by sacrificing the adaptability of transmitting, which leads to unbalanced and insufficient use of limited on chip resources. In this paper, a fully adaptive routing mechanism is proposed to balance buffer utilization. At the same time, we proposed a hybrid virtual channel allocation strategy in order to make full use of buffer resources. For a large number of single-flit wired packets and wireless packets, more aggressive non-atomic allocation is adopted, and the remaining packets are arranged with atomic allocation to avoid potential deadlock. Experimental results showed that compared with the traditional deadlock solutions, the proposed scheme introduces less area and power consumption, and balances the utilization of buffer resources. Under the non-uniform Traffic Patterns, the proposed scheme has better performance in terms of delay and throughput, achieves 62.0% improvement compared to the traditional counterpart.

DTN Architecture With Resource-Aware Rate Adaptation for Multiple Bundle Transmission in InterPlanetary Networks

The world of telecommunications, from some years, is revitalizing an high interest to Deep Space Communication (DSC) and Delay Tolerant Network (DTN) architecture as enabling technologies to offer internet services for aerospace vehicles or orbital stations. The communications between the planets in the solar system is well known to present a series of problems that need to be faced. The fundamental challenge with InterPlanetary Communication (IPC) is the intermittent Line of Sight (LoS) caused by planets movement and the significant propagation time. A possible way to face with it is the use of a DTN approach where routers use a store-and-forward transmission in a hop-by-hop message exchange mechanism. Thanks to this approach, it is possible to face many issues related to the transport layer. The main contribution of this work concerns the proposal of a bundle management layer able to assign at the receiver the appropriate bundle rate on the basis of concurrent bundle transmissions, the output link capacity and buffer resources. The proposed strategy can be integrated with a routing algorithm for InterPlanetary Networks (IPNs) and DTN networks. In our case we consider the Earliest Arrival Optimal Delivery Ratio (EAODR) routing that applies a modified temporal graph to select the IPN nodes. Performance evaluation have been led out to show the benefits of our proposal on the single links and also on the routing scheme in terms of bundle delivery ratio and average bundle delivery time.

Efficacy and acceptability of digital stress management micro-interventions

Tremendous interest exists in the use of brief digital “micro-interventions” for online and mobile intervention. Most digital micro-interventions, however, lack a strong theoretical or empirical basis and have not demonstrated efficacy and acceptability. We developed a suite of brief digital stress management micro-interventions based on theory and empirical evidence and tested the efficacy and acceptability of these micro-interventions in managing stress, mood, and perseverative cognition. Participants were 1,050 US adults (ages 35-65) in good health who received digital micro-intervention content (or comparison content) and then completed post-intervention measures and acceptability ratings. We created 16 brief (<2 min) micro-interventions across four therapeutic domains (relaxation, response modulation, positive experiences, and resource buffers) and a brief active comparison activity. We also created one multi-component micro-intervention (∼20 min) that contained elements of all four domains, and a time-matched active comparison activity. A subset (n=850) of participants received one of the 16 brief micro-interventions or a brief comparator; the remainder (n=200) received either the longer multi-component micro-intervention or the time-matched active comparator task. All micro-intervention stress management content reduced acute stress, negative mood, and perseverative cognitions. For all outcomes, the multi-component intervention showed the strongest effects. Active comparator tasks were more weakly associated with outcomes (except that brief distraction was highly effective at reducing perseverative cognitions). Micro-intervention acceptability was generally high across multiple dimensions. These data demonstrate that a diverse set of 16 brief digital micro-interventions comparison activity were efficacious, and an integrative multi-component micro-intervention was more efficacious. Such micro-interventions hold great potential for scalable digital implementation, including “just-in-time” intervention in response to acute risk states.

A heuristic algorithm in project scheduling and project time optimization: through managing the size of the project buffer, feeding buffers, and resource buffers

Execution of projects is not always done according to the original plan, and in many cases, projects are accompanied by delays, which will lead to financial and time losses for companies and organizations. This research is introduced an innovative heuristic algorithm for more accurate scheduling of projects and minimizing delays in them by managing project buffers and determining their size. Providing a heuristic algorithm to determine the size of project buffer, feeding buffers, and resource buffers simultaneously is the innovation of this research. The proposed heuristic algorithm is coded and run in MATLAB software to analyze the data and also evaluate its performance and the results are compared with similar previous algorithms. The results are showed that using this algorithm will have better results than similar algorithms and will increase project productivity and projects can be completed in a shorter time.

Planning &amp; Controlling Tourism Landscape Design Project using Critical Chain Project Management (CCPM) Schedulling

Project Management must meet the performance criteria of timeliness and budget control agreed between the project owner, contractor and supervisor. However, the level of uncertainty in the project implementation process is very high and causes delays and over budgeting. The Village Nongkojajar is a Proposed Integrated Agropark Project as a result of the revitalization of the non-productive land of the Horticultural Seed Farm owned by the Government of East Java Province. The Landscape Architecture Project has a Creative Value Chain which is divided into Creation - Implementation - Management. Therefore scheduling and controlling the project becomes important in order to to estimate working time and the project remains in accordance with the predetermined project timeline and whenever corrective action is needed. The relevance of Project Management in this research is that with high project problems, require a good planning process. The Project can be completed quickly than the specified time and more cost effective. Critical Chain Project Management (CCPM) is one method for completing and preparing projects that focus on the resource requirements for implementing projects by adding project buffers, feeding buffers and resource buffers. By analyzing using CCPM, researchers can look for optimization and control in implementing project scheduling.

Comparison of the effect of sodium bicarbonate, sodium sesquicarbonate, and zeolite as rumen buffers on apparent digestibility, growth performance, and rumen fermentation parameters of Arabi lambs.

The current research was conducted to compare the effect of various buffers or alkalizers in Arabi lambs and find new and less expensive buffering resources. Forty-five Arabi lambs with an average weight of 29.37 ± 3.63kg were used in a completely randomized design with five treatments and nine replicates. Treatments included 1 - control diet (no buffer); 2 - base diet + 0.75% sodium sesquicarbonate, 0.75% sodium bicarbonate; 3 - base diet + 2% zeolite; 4 - base diet + 1.5% sodium bicarbonate; and 5 - base diet + 1.5% sodium sesquicarbonate. Results showed that rumen pH increased and ammonia nitrogen concentration decreased in diets containing buffer in comparison to control diet (P < 0.05). Rumen concentration of acetate and acetate to propionate ratio showed reduction in experimental diets compared to control (P < 0.05). The concentration of propionate in control diet increased significantly compared to diets receiving buffer (P < 0.05). Using 1.5% sodium bicarbonate in the diet causes a significant increase in rumen protozoa population compared to the control group (P < 0.05). There was no significant difference in dry matter intake and growth performance of lambs. Generally, the effects of using 2% of zeolite were competitive with the effects of other buffers, and caused an increase in the rumen pH and concentration of the acetate. Therefore, the use of buffer in fattening lambs ration fed moderate concentrate diets is beneficial, and it is possible to use low-cost zeolite buffer in the ration of livestock as an alternative to sodium bicarbonate and/or sodium sesquicarbonate.

Job Insecurity and Depressive Symptoms in Mothers and Adolescents: A Dyadic Study

In this study, we examined whether maternal perceived job insecurity (JI) affected depressive symptoms in their adolescent children. Specifically, we tested a mediator-moderator model in which we hypothesized that maternal JI was related to adolescent depressive symptoms indirectly via maternal depressive symptoms. Most importantly, we also explored whether this indirect path was moderated by two buffering resources, i.e., maternal off-job recovery and adolescent self-esteem. There is little research on the effects of parental JI on adolescent children from the viewpoint of buffering moderators. Due to growing insecurity in the labor markets we need more information about resources that mitigate the harmful effects of JI within families. We tested the model via structural equation modeling among Finnish mother-adolescent dyads (N = 601, mothers aged 32–60 years; adolescents aged 14–18 years). Our results showed that mothers’ depressive symptoms did not mediate the relationship between mothers’ JI and adolescents’ depressive symptoms, but mothers’ JI was directly related to increased depressive symptoms among both adolescents and their mothers. Mothers’ off-job recovery and adolescents’ self-esteem buffered against mothers’ JI in relation to depressive symptoms. When off-job recovery (in mothers) and self-esteem (in adolescents) were high, the higher level of JI was not associated with increased depressive symptoms. Off-job recovery among working mothers should be improved as it has potential to protect against JI. Moreover, adolescents’ self-esteem should be boosted as it mitigates the effects of mothers’ job stress (e.g., JI) on adolescents’ well-being.

Hybrid Memory Buffer Microarchitecture for High-Radix Routers

Hierarchical high-radix router microarchitecture consisting of small SRAM-based intermediate buffers has been used in large-scale supercomputers interconnection networks. While hierarchical organization enables efficient scaling to higher switch port count, it requires intermediate buffers which can cause performance bottleneck. Shallow intermediate buffers can cause head-of-line blocking to create backpressure towards input buffers and reduce overall performance. Increasing intermediate buffer size overcomes this problem but becomes infeasible due to the large overhead. In this work, we propose to organise decentralized intermediate buffers as a centralized buffer and leverage alternate memory technology to increase its capacity. In particular, we exploit the high-density nature of Spin-Torque Transfer Magnetic RAM (STT-MRAM) to increase intermediate buffer depth while also providing near-zero leakage power. STT-MRAM has disadvantages such as higher write latency and higher write energy. To overcome these disadvantages, we propose DeepHiR, a novel deep hybrid buffer organization (STT-MRAM and SRAM) combined with a centralized buffer organization to provide high performance with minimal cost. Although the deep intermediate buffer provided by DeepHiR can effectively improve router performance, a large amount of input buffer will still cause a lot of hardware overhead. At the same time, deeper intermediate buffers also makes it take longer for the backpressure to propagate to the source node, thereby reducing the performance of DeepHiR. Therefore, we further propose ElasHiR, which leverages elastic input buffer design in the centralized row buffer to allow a part of the centralized row buffer to act as input buffer. ElasHiR adopts reduced input buffers and automatically determines the length of input buffer in the centralized row buffer. This design minimizes the buffer resource while achieving excellent efficiency. Evaluation results show that DeepHiR can achieve 56.7 percent performance improvement in packet latency under synthetic traffic, and the cost of energy and area is moderate. ElasHiR can reduce the input buffer by 93.8 percent with performance comparable to DeepHiR.

The Lived Experiences of African American Centenarians: Narrative Exploration From a Life Course Theory Perspective.

The purpose of this study was to examine examples of inequality experienced by African American centenarians throughout the life course. Data were derived from 111 oral history transcripts from the Oklahoma Oral History Project. A subsample of all centenarians who identified as African American (n = 12) was utilized. An interpretive phenomenological approach and thematic analysis were used to review and code each narrative. Findings revealed 2 themes: (a) historical time and place with subthemes on family lineage, segregation, discrimination, and racism, and societal changes; (b) resource buffers against inequality with subthemes on multigenerational transmission, support and cultural connectedness, religion and spirituality, and self-reliance. Several centenarians acknowledged having grandparents who were slaves and others reported hidden family lineage due to complicated mixed-race dynamics. Work histories were affected by the historical era, with limited "domestic" positions for women. Experiences with education varied, with one centenarian commenting on his legacy of teaching at historically Black schools whereas others expressed difficulty in keeping up with school and regret for not finishing. This is one of the first studies to explore life course experiences specific to African American centenarians. Centenarians in this study possessed survival skills and various support systems including community and religious/spiritual support, perhaps buffering against the impact of cumulative racial inequality.

Determination of early warning time window for bottleneck resource buffer

As an advanced warning system, resource buffer can notify the use of resources and provide added protection to critical chain activities. However, most of the existing studies in the literature focus on locating actual time buffers at various locations. In this study, we locate bottleneck resource buffers along the critical chain and determines optimal time windows for the first time. Drawing on the theory of constraints, we take into account factors such as bottleneck resource sensitivity, the idle cost, the start time flexibility and the work flow on the critical chain, and develop a quantitative model for the optimal time window determination of resource buffer. The proposed method is tested and compared with the currently widely adopted buffer management method. Results of the computational experiment of the proposed method demonstrates its relative dominance over the predominant buffer management approach in terms of overall completion time, project cost, the probability of delay and the average buffer consumption ratio.