Search

It is a challenging task to achieve the minimum average CCT (coflow completion time) and provide isolation guarantees in multi-tenant datacenters without prior knowledge of coflow sizes. State-of-the-art solutions either focus on minimizing the average CCT or providing optimal isolation guarantees. However, achieving the minimum average CCT and isolation guarantees in multi-tenant datacenters is difficult due to the conflicting nature of these objectives. Therefore, we propose FIGCS-TF (Fast and Isolation Guarantees Coflow Scheduling via Traffic Forecasting), a coflow scheduling algorithm that does not require prior knowledge. FIGCS-TF utilizes a lightweight forecasting module to predict the relative scheduling priority of coflows. Moreover, it employs the MDRF (monopolistic dominant resource fairness) strategy for bandwidth allocation, which is based on super-coflows and helps achieve long-term isolation. Through trace-driven simulations, FIGCS-TF demonstrate communication stages that are 1.12 × , 1.99 × , and 5.50 × faster than DRF (Dominant Resource Fairness), NCDRF (Non-Clairvoyant Dominant Resource Fairness) and Per-Flow Fairness, respectively. In comparison with the theoretically minimum CCT, FIGCS-TF experiences only a 46% increase in average CCT at the top 95th percentile of the dataset. Overall, FIGCS-TF exhibits superior performance in reducing average CCT compared to other algorithms. [ABSTRACT FROM AUTHOR]

Copyright of Advances in New & Renewable Energy is the property of Editorial Office of Advances in New & Renewable Energy and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

It is shown that the velocity fluctuation spectra measured using a hot wire in the potential flow region of the near field of a turbulent jet with a coflow can be converted into pressure fluctuation spectra. The proposed conversion method is based on the fact that the structure of instability waves, which make a decisive contribution to jet near-field fluctuations, resembles homogeneous one-dimensional waves, which makes it possible to locally link pressure fluctuations and the fluctuations of the streamwise velocity component measured by a hot wire. [ABSTRACT FROM AUTHOR]

Recently, it has been proposed to develop space power systems based on filtration combustion of metal powders with oxygen supplied by a chemical oxygen generator. The experiments with lithium and magnesium powders at natural infiltration of oxygen have shown propagation of both counterflow and coflow combustion waves. However, natural filtration combustion of metal powders at relatively low pressures is not sufficiently understood. In the present paper, we investigate the natural coflow combustion waves propagating through a porous medium. The porous matrix is made of metal particles that react with oxygen flowing from the open end of the sample to the reaction zone where it is consumed forming a condensed product which also has a porous structure. The gas flow is due to the pressure difference between the pressure at the open end and that in the reaction zone (the so-called natural filtration). The open end is where the sample is ignited, so that the gas flows through the reaction products, i.e. in the same direction as the combustion wave propagates (coflow filtration). Our mathematical model involves the conservation of energy equation and gas mass, solid reactant mass, and gas momentum balances, as well as an equation of state, and appropriate boundary and initial conditions. It is studied analytically under the combustion front approximation. When the reaction zone is close to the open end, there is sufficient amount of oxygen in the reaction zone and the reaction is controlled by kinetic factors (the kinetic regime of propagation). As the reaction moves away from the open end, it is gas supply that becomes a limiting factor (filtration regime). Both kinetic and filtration regimes of propagation as well the transition between them are analytically studied in this paper. [ABSTRACT FROM AUTHOR]

Many emerging Internet of Things (IoT) applications deployed on cloud platforms have strict latency requirements or deadline constraints, and thus meeting the deadlines is crucial to ensure the quality of service for users and the revenue for service providers in these delay-stringent IoT applications. Efficient flow scheduling in data center networks (DCNs) plays a major role in reducing the execution time of jobs and has garnered significant attention in recent years. However, only few studies have attempted to combine job-level flow scheduling and routing to guarantee meeting the deadlines of multi-stage jobs. In this paper, an efficient heuristic joint flow scheduling and routing (JFSR) scheme is proposed. First, targeting maximizing the number of jobs for which the deadlines have been met, we formulate the joint flow scheduling and routing optimization problem for multiple multi-stage jobs. Second, due to its mathematical intractability, this problem is decomposed into two sub-problems: inter-coflow scheduling and intra-coflow scheduling. In the first sub-problem, coflows from different jobs are scheduled according to their relative remaining times; in the second sub-problem, an iterative coflow scheduling and routing (ICSR) algorithm is designed to alternately optimize the routing path and bandwidth allocation for each scheduled coflow. Finally, simulation results demonstrate that the proposed JFSR scheme can significantly increase the number of jobs for which the deadlines have been met in DCNs.

An experimental study was conducted on the effect of the shape of the fuel nozzle outlet on the suppression of soot emission from a coaxial flow diffusion flame. Three nozzle shapes were used, namely circular, square, and an equilateral triangle. Propane was used as the fuel and air as the oxidizer. The fuel flow rates Qf were 2, 3, and 4 cm3/s. At the maximum Qf of 4 cm3/s, the nozzle shape changed the soot discharge characteristics and the flame structure. The soot discharge was suppressed the most with the triangular nozzle and the least with the circular nozzle. The shape of the soot concentration distribution near the flame tip was similar to the nozzle shape, indicating that the influence of the nozzle shape reached the flame tip. Soot formation started earliest in the triangular nozzle and latest in the circular nozzle. In non-circular nozzles, soot started to form from the downstream regions of the corners of the nozzle rims. This is because the profile of the fuel flow was close to the flame front in these downstream regions. Soot oxidation was fastest in the triangular nozzle and slowest in the circular nozzle. Therefore, the flame was the lowest in the triangular nozzle and the highest in the circular nozzle. The oxidizer flow velocity toward the flame base was high in the downstream region of the flat side of the non-circular nozzles. The velocity was especially high in the triangular nozzle, and the CH radical emission, which is closely related to the heat release rate, was also strong at the flame base. As mentioned above, the fuel concentration distribution, the convective transport of the oxidizer, and the flame structures changed in the non-circular nozzles, and in particular, soot emission was suppressed in the triangular nozzle.

• IoT boom introduces requirements of processing streams at the edge. • A data locality-aware approach optimizes the resource allocation in edge computing. • A load-aware mechanism further improves the query performance. • Coflow dependencies are considered to shape the network usage. • Implementing on Apache Storm and Performing a real cluster experiment on CloudLab. The proliferation of Internet-of-Things (IoT) devices is rapidly increasing the demands for efficient processing of low latency stream data generated close to the edge of the network. Edge computing-based stream processing techniques that carefully consider the heterogeneity of the computational and network resources available in the infrastructure provide significant benefits in optimizing the throughput and end-to-end latency of the data streams. In this paper, we propose a novel stream query processing framework called Amnis that optimizes the performance of the stream processing applications through a careful allocation of computational and network resources available at the edge. The Amnis approach differentiates itself through its consideration of data locality and resource constraints during physical plan generation and operator placement for the stream queries. Additionally, Amnis considers the coflow dependencies to optimize the network resource allocation through an application-level rate control mechanism. We implement a prototype of Amnis in Apache Storm. Our performance evaluation carried out in a real testbed shows that the proposed techniques achieve as much as 200X improvement on the end-to-end latency and 10X improvement on the overall throughput compared to the default resource aware scheduler in Storm. [ABSTRACT FROM AUTHOR]

Applications in data-parallel computing typically consist of multiple stages. In each stage, a set of intermediate parallel data flows (Coflow) is produced and transferred between servers to enable starting of next stage. While there has been much research on scheduling isolated coflows, the dependency between coflows in multi-stage jobs has been largely ignored. In this paper, we consider scheduling coflows of multi-stage jobs represented by general DAGs (Directed Acyclic Graphs) in a shared data center network, so as to minimize the total weighted completion time of jobs. This problem is significantly more challenging than the traditional coflow scheduling, as scheduling even a single multi-stage job to minimize its completion time is shown to be NP-hard. In this paper, we propose a polynomial-time algorithm with approximation ratio of $O(\mu \log (m)/\log (\log (m)))$ , where $\mu $ is the maximum number of coflows in a job and $m$ is the number of servers. For the special case that the jobs’ underlying dependency graphs are rooted trees, we modify the algorithm and improve its approximation ratio. To verify the performance of our algorithms, we present simulation results using real traffic traces that show up to 53% improvement over the prior approach. We conclude the paper by providing a result concerning an optimality gap for scheduling coflows with general DAGs. [ABSTRACT FROM AUTHOR]

Non-premixed impinging jet flames with different coflow conditions are performed using PIV technology combined with numerical simulation to investigate flame instability in the vicinity of wall. Results indicate that the increase of coflow velocity results in a more chaotic flow field and higher fuel efficiency, and the increase of coflow temperature leads to ignition advance and the increase of NO concentration. These can be attributed to the coupling effect of Kelvin-Helmholtz instability, convective instability and Rayleigh-Taylor instability. High coflow velocity is more likely to induce Kelvin-Helmholtz instability and convective instability, and the increase of coflow temperature enhances Rayleigh-Taylor instability and convective instability. Due to the impact effect in the vicinity of wall, the flame instability is more likely to be induced at high coflow velocity. Meanwhile, the increase of coflow temperature can inhibit flame wrinkles. The flame dynamics is affected by turbulent mixing, head-on collision, shear and convective behaviors in non-premixed flames. [ABSTRACT FROM AUTHOR]

Achieving the minimum average coflow completion time(CCT) and the isolation guarantees for multi-tenant, is considered a challenge in a cloud environment. This is because the minimum average CCT and isolation guarantees are two conflicting targets, and they cannot be achieved simultaneously. Prior solutions have implemented a single target either minimizing the average CCT or isolation guarantees. The prior solutions are also limited to clairvoyant scheduling. They also assume the availability of the complete knowledge of coflow sizes before the communication starts. In this paper, we propose an efficient scheduling algorithm smallest-height-first DRF(SHFDRF) for near-optimal scheduling and isolation guarantees without prior knowledge of coflow size. SHFDRF achieves the long-term isolation guarantees and the minimum average CCT by the smallest height first and the monopolistic dominant resource fairness bandwidth allocation strategy. The smallest height first and the monopolistic dominant resource fairness bandwidth allocation strategy can also improve link utilization and system throughput. The trace-driven simulation shows that SHFDRF enables communication stages to 1.28 × , 2.27 × , and 6.28 × faster on the 95th percentile compared to DRF, NCDRF, and Per-Flow Fairness. Even compared with minimum CCT, the completion time of coflow only slowed down by 13.9% on the 95th percentile. Overall, the performance of SHFDRF is acceptable, and it can be applied to the actual datacenter without the limitation of complete prior knowledge. [ABSTRACT FROM AUTHOR]

Communication in data centers often involves many parallel flows that all share the same performance goal (e.g. to minimize the average completion time). A useful abstraction, coflow , is proposed to express the communication requirements of prevalent data parallel paradigms such as MapReduce and Spark. The multiple coflow routing and scheduling problem makes it challenging to derive a good theoretical performance ratio, as coexisting coflows may compete for the same network resources such as link bandwidths. In this paper, we focus on the coflow problem in one popular data center infrastructure: the Leaf-Spine topology. We first formulate the problem and study the path selection issue on this two-tier structure. In order to minimize the average coflow completion time (CCT), we propose the Multi-hop Coflow Routing and Scheduling strategy (MCRS) and prove that our method has a reasonably good competitive ratio. Extensive experiments and large-scale simulations show that MCRS outperforms the state-of-the-art heuristic schemes under the Leaf-Spine topology. [ABSTRACT FROM AUTHOR]

The traffic flows in parallel computing systems show clustered, correlative nature and the flows are always latency-sensitive. These flows have been abstracted as “Coflow” to pursue overall optimization. Concurrent Coflows on the network show very novel traffic patterns. On the other hand, multiple optical interconnection network architectures have been proposed to enable the traffic adaption topology reconstructions. Nevertheless, topology reconstruction strategies are application-agnostic, and their optimization objective of network performance cannot meet the Coflow demand. In order to exert the flexibility of optical topology to promote the performance of parallel computing application by Coflow acceleration, the traffic patterns are preferred to be well recognized and then an adaptive topology is generated accordingly. To avoid further complex, such recognition is expected to finish without prior knowledge from the application layer. Then, the topology should be reconstructed to minimize the Coflow completion time. To implement these procedures, we proposed a traffic pattern-aware topology reconstruction strategy. Our strategy first combines CNN and spectral clustering to realize the traffic patterns awareness. And then, the genetic searching algorithm is used to mind the proper topology. Based on real traffic trace from Facebook computing application, large-scale simulations have verified the efficiency of such a strategy by lowering the completion time of computing jobs. In addition, the experimental demonstration has confirmed the conclusions.

Satellite-based communication technology has gained much attention in the past few years, where satellites play mainly the supplementary roles as relay devices to terrestrial communication networks. Unlike previous work, we treat the low-earth-orbit (LEO) satellites as secure data storage mediums. We focus on data acquisition from a LEO satellite based data storage system (also referred to as the LEO based datacenters), which has been considered as a promising and secure paradigm on data storage. Under the LEO based datacenter architecture, one fundamental challenge is to deal with energy-efficient downloading from space to ground while maintaining the system stability. In this paper, we aim to maximize the amount of data admitted while minimizing the energy consumption, when downloading files from LEO based datacenters to meet user demands. To this end, we first formulate a novel optimization problem and develop an online scheduling framework. We then devise a novel coflow-like “Join the first $K$ K -shortest Queues (JKQ)” based job-dispatch strategy, which can significantly lower backlogs of queues residing in LEO satellites, thereby improving the system stability. We also analyze the optimality of the proposed approach and system stability. We finally evaluate the performance of the proposed algorithm through conducting emulator based simulations, based on real-world LEO constellation and user demand traces. The simulation results show that the proposed algorithm can dramatically lower the queue backlogs and achieve high energy efficiency. [ABSTRACT FROM AUTHOR]

The temperature distribution of an operating planar solid oxide fuel cell (SOFC) is experimentally investigated under direct internal reforming conditions. An in situ measurement is conducted using a cell holder and an infrared (IR) camera. The effects of the gas flow configuration, exothermic power generation reaction, and endothermic steam–methane reforming reaction are examined at a furnace temperature of 770 °C. The fuel flow and airflow are set to a coflow or counterflow configuration. The heat generation and absorption by the reactions are varied by tuning the average current density and the concentration of methane in the supplied fuel. The maximum value of the local temperature gradient along the cell tends to increase with increasing internal reforming ratio, regardless of the gas flow configuration. From the view point of a small temperature gradient, the counterflow configuration clearly shows better characteristics than that of the coflow, regardless of the internal reforming ratio. • The temperature distribution of a planar SOFC was experimentally investigated. • An in situ measurement was conducted using an infrared camera. • The effects of gas flow, power generation and internal reforming were examined. • The counterflow showed a small temperature gradient that of the coflow. [ABSTRACT FROM AUTHOR]

The blowout behavior of non-premixed turbulent coflow jet flames under microgravity environment was studied experimentally by utilizing a 3.6 s drop tower. Variations of flames leading to liftoff as well as blowout were examined by varying the coflow velocity and compared with those obtained under the normal gravity condition. A modeling work was conducted to incorporate the effects of the gravity (buoyancy) and coflow velocity on blowout behavior. Major findings include: (1) the flame length in microgravity was longer than that in normal gravity and decreased with increasing coflow velocity. The flame in microgravity showed more intense yellow luminosity with larger sooting zone; (2) the flame liftoff height increased with increasing coflow velocity in both gravity levels. The flame base was closer to the burner in microgravity as compared with that in normal gravity; (3) the blowout velocity in microgravity was appreciably larger than that obtained in normal gravity; and (4) a physical model based on Damköhler number was developed by using similarity solutions to characterize the differences in the blowout limits considering both the coflow and gravity (buoyancy) effects. The proposed model can successfully predict the experimental data. This work provided new data and basic scaling analysis for blowout limit of non-premixed turbulent jet flames considering both the coflow and gravity (buoyancy) effects. [ABSTRACT FROM AUTHOR]

Chlorine is extensively used in disinfection processes in water and wastewater treatment processes. The chlorine is typically dosed by jet mixing, and yet no systematic study of chlorine jets has been conducted. In the Stonecutters Island Sewage Treatment Works (SCISTW) of the Hong Kong Harbour Area Treatment Scheme (HATS), concentrated chlorine solution (in sodium hypochlorite form) is injected into primary treated effluent in the form of turbulent dense jets. In view of the importance of sewage disinfection to the environmental impact, it is essential to understand the mixing and reaction of chlorine with ammonia nitrogen for disinfection dosage optimization. In this study, an integral buoyant jet model is developed for a chlorine jet reacting with ammonia nitrogen in CEPT sewage. The jet mixing and chlorine demand is studied by accounting for the stoichiometric reaction of chlorine and ammonia including breakpoint chlorination. Based on experiments of chlorine jet in ammonia solution, the stoichiometric mass ratio of chlorine to ammonia under dynamic mixing process is found to be 7.48. The chlorination kinetics model is then incorporated in a general model of a chlorine jet discharging into a coflowing primary treated sewage effluent, with the reaction ratio calibrated based on experimental measurements for the stagnant case. Significant chlorine demand can occur within a short distance (in the order of 0.1 m, or 1 s or less of travel time) from the dosing point due to the turbulent mixing and reaction with ammonia nitrogen and organic impurities. The chlorine demand of CEPT effluent increases with jet dosing total residual chlorine (TRC) concentration. At TRC < 2000 mg/L, the chlorine demand of sewage is mainly due to ammonia nitrogen (>60%), while for TRC ≫ 2000 mg/L, up to two thirds of chlorine demand can be induced by the oxidation of organic debris. Model predictions of chlorine demand and concentration agree well with laboratory experiments and field data. [ABSTRACT FROM AUTHOR]

The objective of this paper is to investigate the flame structure and liftoff behaviors of a dimethyl ether central jet in CH4/air vitiated coflow in a coflow burner. The liftoff behaviors of dimethyl ether jet flames in the air flow were studied firstly. The flame stability of the burner was analyzed by measuring the flow field temperature with thermocouples. By changing the coflow rate and CH4 equivalence ratio, the liftoff behaviors of dimethyl ether jet flames under different vitiated coflow environments were discussed. The jet flame structure was also analyzed qualitatively by high-speed photography. [ABSTRACT FROM AUTHOR]

A coflow is a collection of related parallel flows that occur typically between two stages of a multi-stage computing task in a network, such as shuffle flows in MapReduce. The coflow abstraction allows applications to convey their semantics to the network so that application-level requirements can be better satisfied. In this paper, we study the routing and scheduling of multiple coflows to minimize the total weighted coflow completion time (CCT). We first propose a rounding-based randomized approximation algorithm, called OneCoflow, for single coflow routing and scheduling. The multiple coflow problem is more challenging as coexisting coflows will compete for the same network resources, such as link bandwidth. To minimize the total weighted CCT, we derive an online multiple coflow routing and scheduling algorithm, called OMCoflow. We then derive a competitive ratio bound of our problem and prove that the competitive ratio of OMCoflow is nearly tight. To the best of our knowledge, this is the first online algorithm with theoretical performance guarantees which considers routing and scheduling simultaneously for multi-coflows. Compared with existing methods, OMCoflow runs more efficiently and avoids frequently rerouting the flows. Extensive simulations on a Facebook data trace show that OMCoflow outperforms the state-of-the-art heuristic schemes significantly (e.g., reducing the total weighted CCT by up to 41.8% and the execution time by up to 99.2% against RAPIER). [ABSTRACT FROM AUTHOR]

Traditional network resource management mechanisms are mainly flow or packet based. Recently, coflow has been proposed as a new abstraction to capture the communication patterns in a rich set of data parallel applications in data centers. Coflows effectively model the application-level semantics of network resource usage, so high-level optimization goals, such as reducing the transfer latency of applications, can be better achieved by taking coflows as the basic elements in network resource allocation or scheduling. Although efficient coflow scheduling methods have been studied, in this paper, we advocate to schedule weighted coflows as a further step in this direction, where weights are used to express the importances or priorities of different coflows or their corresponding applications. We propose the Weighted Coflow Completion Time (WCCT) minimization problem and a $(2-\frac{2}{n+1})$(2-2n+1)-approximate optimal offline algorithm, where n is the concurrent number of coflows. We then design an information-agnostic online algorithm named IAOA to dynamically schedule coflows according to their weights and the instantaneous network condition. We also design and implement a coflow scheduling system named FlyTransfer, which can use the online algorithm as its scheduling method. We test the performance of FlyTransfer by trace-driven simulations as well as real deployment in openstack. Our evaluation results show that, compared to the latest information-agnostic coflow scheduling algorithms, FlyTransfer can reduce more than 40 percent of the WCCT, and more than 30 percent of the completion time for coflows with above-the-average level of importance. It even outperforms the most efficient clairvoyant coflow scheduling method by reducing around 30 percent WCCT, and 25- 30 percent of the completion time for coflows with above-the-average importance, respectively. [ABSTRACT FROM AUTHOR]

Abstract The mechanistic pathways of PAH formation and growth remain uncertain. In addition, our current understanding of the transformation of PAHs into young soot is limited. The PAHs participating in this transformation remain ambiguous. Simultaneous measurements of PAHs and particles are necessary to better understand how particulates are formed in flames. A comprehensive analysis has been performed on a n-dodecane doped methane coflow diffusion flame. PAHs have been analyzed with GC/MS while the particulates have been collected from the flame centreline for studying under a Transmission Electron Microscope (TEM). The soot measurements from a previous study have been used to complete the comprehensive analysis. The novelty of the current study lies in the fact that it provides quantitative information on the growth of PAHs ranging from naphthalene (A2) to pyrene and fluoranthene (together they are referred to as A4 in this study) and how these species participate in the formation, growth and aging of the young soot. The experimental results show that before the commencement of the young soot, the gaseous phase is dominated by species equal to and smaller than A4. PAHs larger than A4 were not detected. With the growth of the young soot, the mole fraction of A4 decreases. The results suggest that small and medium-sized PAHs (A4 and smaller) are responsible for the formation and growth of the young soot. As the young soot transform into mature solid soot, A4 increases abruptly. The study shows that PAH growth and soot maturity are not mutually exclusive. The target flame has also been simulated numerically to identify the problems associated with the state-of-the-art model. The numerical model with irreversible nucleation can not capture the formation of young soot even though it considers nucleation from A4. A comprehensive database of PAHs and soot has been created for future numerical model validations. [ABSTRACT FROM AUTHOR]

In modern data centers, many flow-based and task-based schemes have been proposed to speed up the data transmission in order to provide fast, reliable services for millions of users. However, the existing flow-based schemes treat all flows in isolation, contributing less to or even hurting user experience due to the stalled flows. Other prevalent task-based approaches, such as centralized and decentralized scheduling, are sophisticated or unable to share task information. In this work, we first reveal that the relinquishing bandwidth of leading flows to the stalled ones effectively reduces the task completion time. We further present the design and implementation of a general supporting scheme that shares the flow-tardiness information through a receiver-driven coordination. Our scheme can be flexible and widely integrated with the state-of-the-art TCP protocols designed for data centers in either single stage or multiple stage scenario, while making no modification on switches. Through the testbed experiments and simulations of typical data center applications, we show that in single stage scenario, our scheme reduces the task completion time by 70% and 50% compared with the flow-based protocols (e.g., DCTCP, L2DCT) and task-based scheduling (e.g., Baraat), respectively. Moreover, our scheme also outperforms other approaches by 18%~25% in prevalent topologies of the data center. For multiple stage scenario, our scheme also has up to 50% improvement compared to other schemes. [ABSTRACT FROM AUTHOR]

Flickering of laminar diffusion flames is known to be caused by the buoyancy-induced toroidal vortices attached to the outer flame surface, making a classical model for understanding flame-vortex interactions. This work experimentally investigates the impacts of ambient coflow on the flickering behaviors of a buoyant jet diffusion flame. Utilizing a simultaneous high-speed flame imaging/OH⁎ chemiluminescence imaging/particle image velocimetry measurement system, the evolutions and interactions of coherent flame and flow structures are resolved. The results demonstrate that the periodic deformation of the flame surface (i.e., flame flickering) arises from the periodic formation, growth, and shedding of the buoyancy-induced outer vortex rings (OVRs), manifesting as a hydrodynamic instability developing along the outer shear layer (OSL). Upon applying a weak-to-moderate coflow, the flame bulge's size gradually reduces, and the flickering frequency slightly increases with increasing coflow velocity. The notable frequency jump phenomenon is confirmed as the coflow velocity rises to a threshold value such that the flickering frequency undergoes a sudden increase of 3–4 Hz; this is also accompanied by a significant reduction in the flame bulge's size and a delay in flame pinch-off. Moreover, a second frequency jump of 2–3 Hz occurs at a higher coflow velocity. Based on quantitative analysis of the OVR core's evolution, we find that the frequency jumps result from a sudden downstream shift of the initial OVR core, essentially indicating the jump of the instability onset point of the OSL. From a hydrodynamic instability perspective, this phenomenon can be explained as a stabilization effect of the coflow on the OSL, resulting in the re-establishment of the instability condition in a downstream location. [ABSTRACT FROM AUTHOR]

Coflow is a network abstraction used to represent communication patterns in data centers. The coflow scheduling problem encountered in large data centers is a challenging NP -hard problem. Many previous studies on coflow scheduling mainly focus on the single-core model. However, with the growth of data centers, this single-core model is no longer sufficient. This paper addresses the coflow scheduling problem within heterogeneous parallel networks, which feature an architecture consisting of multiple network cores running in parallel. In this paper, two polynomial-time approximation algorithms are developed for the flow-level scheduling problem and the coflow-level scheduling problem in heterogeneous parallel networks, respectively. For the flow-level scheduling problem, the proposed algorithm achieves an approximation ratio of O (log ⁡ m / log ⁡ log ⁡ m) when all coflows are released at arbitrary times, where m represents the number of network cores. On the other hand, in the coflow-level scheduling problem, the proposed algorithm achieves an approximation ratio of O (m (log ⁡ m / log ⁡ log ⁡ m) 2) when all coflows are released at arbitrary times. Moreover, we propose a heuristic algorithm for the flow-level scheduling problem. Simulation results using synthetic traffic traces validate the performance of our algorithms and show improvements over the previous algorithm. • Algorithm achieves O (log ⁡ m / log ⁡ log ⁡ m) approximation ratio for flow-level scheduling, where m is network cores. • For coflow-level scheduling, algorithm achieves O (m (log ⁡ m / log ⁡ log ⁡ m) 2) approximation ratio. • Moreover, this paper proposes a heuristic algorithm for the flow-level scheduling problem. [ABSTRACT FROM AUTHOR]

In data centers, the coflow abstraction is proposed to better express the requirements and communication semantics of a group of parallel flows generated by the jobs of cluster computing frameworks. Knowing the coflow-level information, such as coflow size, previous coflow scheduling proposals improve the performance over flow-level scheduling schemes. Recently, since some information of coflow is difficult to obtain in cloud environments, designing coflow scheduling mechanisms with partial or even without any information attracts much attention. However, existing information-agnostic mechanisms are generally built on the least attained service heuristic algorithm that schedules coflows only according to the sent bytes of different coflows, and they all ignore other useful coflow-level information like width, length, and communication patterns. In this paper, we investigate that the coflow completion time could be further decreased by jointly leveraging multiple coflow-level attributes. Based on this investigation, we present a Multiple-attributes-based Coflow Scheduling (MCS) mechanism to reduce the coflow completion time. In MCS, at the start of a coflow, a shortest and narrowest coflow first algorithm is designed to assign the initial priority based on the coflow width. During the transmission of coflows, based on the sent bytes of coflows, we proposed a double-threshold scheme to adjust the priorities of different classes of coflows according to different thresholds. Accordingly, the optimal thresholds are analyzed by using the M/M/1 queuing model. Testbed evaluations and simulations with production workloads show that MCS outperforms the previous information-agnostic scheduler Aalo, and reduces the completion time of small coflows. [ABSTRACT FROM AUTHOR]

In data-parallel computing frameworks, intermediate parallel data is often produced at various stages which needs to be transferred among servers in the datacenter network (e.g., the shuffle phase in MapReduce). A stage often cannot start or be completed unless all the required data pieces from the preceding stage are received. Coflow is a recently proposed networking abstraction to capture such communication patterns. We consider the problem of efficiently scheduling coflows with release dates in a shared datacenter network so as to minimize the total weighted completion time of coflows. Several heuristics have been proposed recently to address this problem, as well as a few polynomial-time approximation algorithms with provable performance guarantees. Our main result in this paper is a polynomial-time deterministic algorithm that improves the prior known results. Specifically, we propose a deterministic algorithm with approximation ratio of 5, which improves the prior best known ratio of 12. For the special case when all coflows are released at time zero, our deterministic algorithm obtains approximation ratio of 4 which improves the prior best known ratio of 8. The key ingredient of our approach is an improved linear program formulation for sorting the coflows followed by a simple list scheduling policy. Extensive simulation results, using both synthetic and real traffic traces, are presented that verify the performance of our algorithm and show improvement over the prior approaches. [ABSTRACT FROM AUTHOR]

Multiple single-channel meso-scale combustors assembling into a multi-channel meso-scale planar combustor is a good solution for elevating the output power of the micro-thermophotovoltaic system. Consequently, how to assembly a multi-channel meso-scale planar combustor for achieving higher output power comes to be an interesting issue. In this work, four representative four-channel meso-scale planar combustors, namely, combustor A, combustor B, combustor C and combustor D are designed. Extensive numerical investigations are conducted to compare the wall temperature of the four combustors under various hydrogen mass flow rates, hydrogen/air equivalence ratios and solid materials. Two variables, namely, mean wall temperature and nonuniformity coefficient of wall temperature are defined for comparing the wall temperature of the four meso-scale combustors. It is found that the purely counterflow four-channel meso-scale planar combustor D has more uniform and higher wall temperature compared with other combustors. Furthermore, the detailed heat transfer mechanisms of four different four-channel meso-scale planar combustors are analyzed and presented. In addition, some critical values are proposed for keeping high and uniform wall temperature on the purely counterflow four-channel meso-scale planar combustor. This work offers us great reference value for the design of multi-channel meso-scale planar combustor. [ABSTRACT FROM AUTHOR]

A new optimized size exclusion chromatography small‐angle X‐ray scattering (SEC‐SAXS) system for biomolecular SAXS at the Australian Synchrotron SAXS/WAXS beamline has been developed. The compact configuration reduces sample dilution to maximize sensitivity. Coflow sample presentation allows an 11‐fold increase in flux on sample without capillary fouling, improving throughput and data quality, which are now primarily limited by the full flux available on the beamline. Multi‐wavelength fibre optic UV analysis in close proximity to the X‐ray beam allows for accurate concentration determination for samples with known UV extinction coefficients and thus estimation of the molecular weight of the scattering particle from the forward X‐ray scattering intensity. Fast‐flow low‐volume SEC columns provide sample throughput competitive with batch concentration series measurements, albeit with a concomitant reduction of potential resolution relative to lower flow rates and larger SEC columns. The performance of the system is demonstrated using a set of model proteins, and its utility to solve various challenges is illustrated with a diverse suite of protein samples. These developments increase the quality and rigor of SEC‐SAXS analysis and open new avenues for biomolecular solution SEC‐SAXS studies that have been challenged by low sample yields, temporal instability, radiation sensitivity and complex mixtures. [ABSTRACT FROM AUTHOR]

The coflow air effect on the ignition process of a methane jet diffusion flame has been investigated using high speed colour/schlieren imaging and image processing techniques experimentally. The methane flow rate is kept at constant ( Re = 55.4), while the coflow air flow rate changes ( Re from 171 to 5985), creating a wide range of the air/fuel velocity ratios varying from 0.36 to 12.5. Special digital image processing techniques are applied to visualise the weak blue flame and weak yellow flame, which is often difficult to view in the presence of the bright orange diffusion flame, during ignition process. The processed images have shown clearly that a sooty diffusion flame is initially formed inside a blue flame pocket at low air velocities. When the coflow air flow rate exceeds 75 l/min, only blue flame can be observed. The equivalence ratio of blue flame has been evaluated based on colour characteristics, which is close to 1 during the ignition process for all the cases. Moreover, the fuel flow, flame and hot gas interactions with the cold air flow are investigated by visualising the schlieren images. It is found that a hot gas bulge is formed due to the excessive fuel exiting before ignition and a hot laminar central jet is formed with the help of coflow effect. The hot gas bulge tip and bottom moving velocities are found to increase with the coflow air flow rates. Besides flow visualisation based on high-speed schlieren imaging sequences, the velocity fields during ignition process have been evaluated quantitatively using optical flow method. [ABSTRACT FROM AUTHOR]

In order to extend the output power of the micro-thermophotovoltaic system, it is necessary to assemble multiple single-channel micro combustors into a multiple-channel micro combustor. As a result, how to assemble multiple single-channel micro combustors into a multiple-channel micro combustor comes to be an interesting issue. In this work, the counterflow combustion concept is used in a double-layer four-channel micro combustor, and three representative double-layer four-channel micro combustors are designed. Extensive numerical investigations are conducted to compare the thermal performance of the three different double-layer four-channel micro combustors under various hydrogen mass flow rates, hydrogen/air equivalence ratios and solid materials. Moreover, the purely counterflow double-layer four-channel micro combustor is compared with the counterflow single-layer double-channel micro combustor in our previous work under various H 2 mass flow rates, hydrogen/air equivalence ratios and solid materials. It is found that the purely counterflow double-layer four-channel micro combustor can keep more uniform wall temperature. And the purely counterflow double-layer four-channel micro combustor has better thermal performance compared with that of the counterflow single-layer double-channel micro combustor. Moreover, some critical values are given for keeping the merit of the purely counterflow double-layer four-channel micro combustor. [ABSTRACT FROM AUTHOR]

The communication between data-intensive applications in data centers often involves a collection of parallel flows, usually referred to as a coflow. A coflow between two groups of machines can capture diverse communication patterns observed in data centers. The static coflow routing and spectrum assignment (CofRSA) problem in optical orthogonal frequency division multiplexed data center networks is investigated, when the coflow traffic demands are given. The static CofRSA problem considers the spectrum constraints of the flows between different coflows and those within the same coflow. We formulate the static CofRSA problem as an integer linear programming (ILP) model. The objective of the ILP model is to minimize the used spectrum slots. However, the ILP model cannot achieve an optimal solution within tolerable time for large networks. To solve the problem, two highly efficient heuristic algorithms, the most cofSize first (MCSF) ordering algorithm and the greedy inserting (GI) algorithm, are proposed to achieve suboptimal solutions. The simulation results indicate that ILP provides an optimal solution for small networks, whereas GI and MCSF yield suboptimal solutions in large networks. The results also show that GI provides more efficient solutions than MCSF. [ABSTRACT FROM AUTHOR]

In current data centers, an application (e.g., MapReduce, Dryad, search platform, etc.) usually generates a group of parallel flows to complete a job. These flows compose a coflow and only completing them all is meaningful to the application. Accordingly, minimizing the average Coflow Completion Time (CCT) becomes a critical objective of flow scheduling. However, achieving this goal in today's Data Center Networks (DCNs) is quite challenging, not only because the schedule problem is theoretically NP-hard, but also because it is tough to perform practical flow scheduling in large-scale DCNs. In this paper, we find that minimizing the average CCT of a set of coflows is equivalent to the well-known problem of minimizing the sum of completion times in a concurrent open shop. As there are abundant existing solutions for concurrent open shop, we open up a variety of techniques for coflow scheduling. Inspired by the best known result, we derive a 2-approximation algorithm for coflow scheduling, and further develop a decentralized coflow scheduling system, D-CAS, which avoids the system problems associated with current centralized proposals while addressing the performance challenges of decentralized suggestions. Trace-driven simulations indicate that D-CAS achieves a performance close to Varys, the state-of-the-art centralized method, and outperforms Baraat, the only existing decentralized method, significantly. [ABSTRACT FROM PUBLISHER]

We study numerically nonuniform quantum turbulence of coflow in a square channel by the vortex filament model. Coflow means that superfluid velocity $$\varvec{v}_\mathrm{s}$$ and normal fluid velocity $$\varvec{v}_\mathrm{n}$$ flow in the same direction. Quantum turbulence for thermal counterflow has been long studied theoretically and experimentally. In recent years, experiments of coflow have been performed to observe different features from thermal counterflow. By supposing that $$\varvec{v}_\mathrm{s}$$ is uniform and $$\varvec{v}_\mathrm{n}$$ takes the Hagen-Poiseuille profile, simulations find that quantized vortices are distributed inhomogeneously. Vortices like to accumulate on the surface of a cylinder with $$\varvec{v}_\mathrm{s} \simeq \varvec{v}_\mathrm{n}$$ . Consequently, the vortex configuration becomes degenerate from three-dimensional to two-dimensional. [ABSTRACT FROM AUTHOR]

Thermoelectric generators (TEGs) present good potential applications for the conversion of low level thermal energy into electrical power, especially applied to waste heat recovery systems. This paper presents an advanced mathematical model that uses multi-element thermoelectric components connected in series, and considers internal and external irreversibility and also the temperature gradients along the flow direction. A Fortran computation program was used for the numerical calculations. This paper focusses mainly on the thermoelectric performance study under four different types of cooling methods with temperature gradient modeling. The results showed that it is important in TEG system design to choose an optimal module area that is appropriate to the mass flow rate of the fluids, but this is not affected by the intake temperature of fluids. With the water cooling method, the coflow and counterflow methods did not need to be distinguished because of their small difference, but they should be distinguished for the air cooling method. The counterflow arrangement generally produced a small higher maximum power output, but needed a much larger module area compared to the coflow method. [ABSTRACT FROM AUTHOR]

This paper reports an experimental investigation on the flow characteristics upstream of a lifted turbulent diffusion flame in the presence of a coflow. Three fuel nozzles made of a long pipe with different outlet geometry were examined. One pair of these nozzles has the same orifice diameter but different normalized lip thickness, and another pair has the same normalized lip thickness but different orifice diameter. The strength of the co-airflow was also varied to assess its impact on the liftoff height of the jet diffusion flame. Previously published studies reported the existence of a hysteresis phenomenon in the liftoff height of a turbulent diffusion flame in the presence of a high co-airflow. That is, as the fuel velocity decreases, the lifted flame base would first move upstream (toward the burner) to a local minima followed by a downstream movement before its reattachment. The results of the present study, however, showed that such a phenomenon does not appear for a fuel pipe having a very small lip thickness. The present results also revealed that in the hysteresis region, the flame base sits where the turbulence intensity experiences its local maxima in the upcoming unburnt mixture. This corroborates the premixed stability theory which is based on turbulence intensity. Based on this, a correlation was found between the flame liftoff height in the hysteresis region and the fuel and co-airflow velocity at the nozzle exit. This relationship predicts successfully the liftoff height trend as a function of the fuel jet and co-airflow velocity and nozzle geometry. Away from the hysteresis region, however, the flame base location tends more toward the outside of the local turbulence intensity maxima. This indicates the limitations of the premixed stability theory in predicting the flame behavior in this region where the effect of the flow large-scale structures becomes important. [ABSTRACT FROM AUTHOR]

This paper presents a new advanced mathematical model for thermoelectric conversion using multi-element thermoelectric components connected in series, considering internal and external irreversibility along with temperature gradients along the flow direction. Combined with computer programming, the theoretical optimizations on the size of the thermoelectric generator (TEG) area are studied with two different flow directions for both hot and cold fluids, namely the coflow and counterflow directions, and the effects of the changing hot fluid parameters are analyzed. The results show that an optimal TEG area exists corresponding to the maximum output power for both the coflow and counterflow directions. The counterflow direction can produce more power output with a larger optimal model area than the coflow direction, yielding greater thermoelectric advantages. [ABSTRACT FROM AUTHOR]

Toward the goal of understanding and controlling instability in combustion systems, we present a fundamental characterization of the interaction of the buoyancy-induced instability in flickering flames with forced excitation of fuel supply. Laminar coflow diffusion flames were acoustically forced, whose frequency responses were recorded as a function of excitation frequency and amplitude. The evolving structure of such flames was also examined through the use of video analysis and particle imaging velocimetry (PIV). For specific combinations of excitation frequency and amplitude, the frequency response of the flames was found to couple to that of the forcing, where the contribution of natural puffing frequency disappears. Such instances of coupling exhibited many harmonics of the excitation frequency, related indirectly to the natural puffing frequency. We showed how such harmonics form, through application of PIV, and furthermore unveiled insight into the physics of how the flame couples to the forcing under certain conditions. Our frequency response characterization provides quantitative results, which are of utility for both modeling studies and active-control strategies. [ABSTRACT FROM PUBLISHER]

Abstract: Effects of nozzle pressure ratios on the coflow during mixing of circular supersonic inner jets with subsonic outer jets were experimentally studied for different area ratios such as 1.00, 1.25 and 1.76 (Aexit/Athroat). The same were numerically computed using CFD. The analyses for inner jet NPR values of 2.75 and 5 reveal that in downstream, the jet approaches symmetry at an axial distance of 1.7 times and 3.0 times the nozzle exit diameter. In circular coflow jet with different area ratios the dependence on the outer structures was also observed. At a greater NPR and area ratio the merging of the inner and outer flow occurs earlier in the downstream. Moreover an elongation in potential core length due to higher concentration of turbulent kinetic energy in the inner shear layer was established. Also, the experimental results of coaxial jets have good correlation with the CFD simulation. Overall, a greater area ratio nozzle exhibits distinct flow characteristics in mixing region. These flow characteristics produce more forceful mixing which is constructive and can be used effectively in combustion processes and propulsion applications. [Copyright &y& Elsevier]

Abstract: In this paper, high speed direct/schlieren imaging system together with Particle Image Velocimetry (PIV) system is applied to investigate the vortex dynamics and structures of methane–air coflow diffusion flames. The schlieren and PIV images show that the dynamics of the vortices outside the visible flame are strongly affected by the coflow air velocity. The coflow air is observed to push the initiation point of toroidal vortices from the nozzle exit to downstream. When the vortices are pushed beyond the visible flame height, the typical flame flickering behavior disappears and the flame becomes stable. The critical air flow rate at which the flame stops oscillating is observed to increase with the fuel flow rate. The visible flame shows no obvious change when the coflow flow rate exceeds a critical value. However, the flow pattern and vortices outside the visible flame keep changing at different air flow rates. The velocity vectors and vorticity contours at different air flow rates are presented and analyzed. The shedding frequency of the toroidal vortex is in good agreement with the flame flickering frequency obtained from a photomultiplier. [Copyright &y& Elsevier]

Abstract: Enzyme microreactors are used as research tools because of many advantages including low reagent consumption and the general notion that mass transfer restrictions are reduced. The use of microchannels can indeed shorten the characteristic mass transfer time but may also affect the productivity of the microreactor. To what extent mass transfer restrictions affect the reaction rate and the productivity is determined by parameters such as the enzyme properties, operating conditions, and dimensions of the microreactor. This article provides the correlations between these parameters for coflow enzyme microreactors obeying Michaelis–Menten kinetics. These correlations outline the design space based on reduced mass transfer restrictions and maximum productivity, respectively. The methodology that yields the design space provides a generic hands-on approach to optimally design coflow enzyme microreactors. We conclude that effectiveness is an important parameter for designing microreactors, but – depending on the reactor configuration – does not necessarily lead to optimized throughput. We demonstrate design windows to illustrate this effect and how to take advantage of it. [Copyright &y& Elsevier]