Communication-intensive Applications Research Articles

A data quarantine model to secure data in edge computing

<p>Edge computing provides an agile data processing platform for latency-sensitive and communication-intensive applications through a decentralized cloud and geographically distributed edge nodes. Gaining centralized control over the edge nodes can be challenging due to security issues and threats. Among several security issues, data integrity attacks can lead to inconsistent data and intrude edge data analytics. Further intensification of the attack makes it challenging to mitigate and identify the root cause. Therefore, this paper proposes a new concept of data quarantine model to mitigate data integrity attacks by quarantining intruders. The efficient security solutions in cloud, ad-hoc networks, and computer systems using quarantine have motivated adopting it in edge computing. The data acquisition edge nodes identify the intruders and quarantine all the suspected devices through dimensionality reduction. During quarantine, the proposed concept builds the reputation scores to determine the falsely identified legitimate devices and sanitize their affected data to regain data integrity. As a preliminary investigation, this work identifies an appropriate machine learning method, linear discriminant analysis (LDA), for dimensionality reduction. The LDA results in 72.83% quarantine accuracy and 0.9 seconds training time, which is efficient than other state-of-the-art methods. In future, this would be implemented and validated with ground truth data.</p>

Optimal circulant graphs as low-latency network topologies

Communication latency has become one of the determining factors for the performance of parallel clusters. To design low-latency network topologies for high-performance computing clusters, we optimize the diameters, mean path lengths, and bisection widths of circulant topologies. We obtain a series of optimal circulant topologies of size $$2^5$$ through $$2^{10}$$ and compare them with torus and hypercube of the same sizes and degrees. We further benchmark on a broad variety of applications including effective bandwidth, FFTE, Graph 500 and NAS parallel benchmarks to compare the optimal circulant topologies and Cartesian products of optimal circulant topologies and fully connected topologies with corresponding torus and hypercube. Simulation results demonstrate superior potentials of the optimal circulant topologies for communication-intensive applications. We also find the strengths of the Cartesian products in exploiting global communication with data traffic patterns of specific applications or internal algorithms.

Real-time task scheduling for FPGA-based multicore systems with communication delay

Efficient scheduling is critical for achieving high performance in the multicore computing environment. This paper focuses on the scheduling problem for the real-time applications in the FPGA-based multicore structure with an objective to minimize the makespan under hardware resource constraints. To address this problem, we propose a priority-driven scheduling algorithm, called the Real-Time Priority-driven Algorithm (ReTPA). In this method, we first allocate the tasks into software and hardware computing units based on their hardware resource constraints. Then scheduling is performed to minimize the makespan. We further extend ReTPA with communication scheduling to improve the scheduling efficiency in the communication-intensive applications, which is called Communication-based Real-Time Priority-driven Algorithm (CReTPA). Our experimental results confirm the efficiency of the proposed algorithms in applications with different scales or different computation-communication ratios. Besides, we compare the two proposed algorithms with a classic priority-driven algorithm, HEFT, an advanced priority-driven algorithm, HETS, and two advanced genetic algorithms, GAA and MGAA, proposed by Abdallah et al. in 2019. The performance comparisons illustrate that our algorithms, ReTPA and CReTPA, both show the outstanding performance and CReTPA obtains the most reasonable solutions within the shortest scheduling time for communication-intensive applications.

Communication Cost Aware Resource Efficient Load Balancing (CARELB) Framework for Cloud Datacenter

Background: Load balancing of communication-intensive applications, allowing efficient resource utilization and minimization of power consumption is a challenging multi-objective virtual machine (VM) placement problem. The communication among inter-dependent VMs, raises network traffic, hampers cloud client’s experience and degrades overall performance, by saturating the network. Introduction: Cloud computing has become an indispensable part of Information Technology (IT), which supports the backbone of digitization throughout the world. It provides shared pool of IT resources, which are: always on, accessible from anywhere, at anytime and delivered on demand, as a service. The scalability and pay-per-use benefits of cloud computing has driven the entire world towards on-demand IT services that facilitates increased usage of virtualized resources. The rapid growth in the demands of cloud resources has amplified the network traffic in and out of the datacenter. Cisco Global Cloud Index predicts that by the year 2021, the network traffic among the devices within the datacenter will grow at Compound Annual Growth Rate (CAGR) of 23.4% Methods: To address these issues, a communication cost aware and resource efficient load balancing (CARE-LB) framework is presented, that minimizes communication cost, power consumption and maximize resource utilization. To reduce the communication cost, VMs with high affinity and inter-dependency are intentionally placed closer to each other. The VM placement is carried out by applying the proposed integration of Particle Swarm Optimization and non-dominated sorting based Genetic Algorithm i.e. PSOGA algorithm encoding VM allocation as particles as well as chromosomes. Results: The performance of proposed framework is evaluated by the execution of numerous experiments in the simulated datacenter environment and it is compared with the state-of-the-art methods like, Genetic Algorithm, First-Fit, Random-Fit and Best-Fit heuristic algorithms. The experimental outcome reveals that the CARE-LB framework improves 11% resource utilization, minimize 4.4% power consumption, 20.3% communication cost with reduction of execution time up to 49.7% over Genetic Algorithm based Load Balancing framework. Conclusion: The proposed CARE-LB framework provides promising solution for faster execution of data-intensive applications with improved resource utilization and reduced power consumption. Discussion: In the observed simulation, we analyze all the three objectives, after execution of the proposed multi-objective VM allocations and results are shown in Table 4. To choose the number of users for analysis of communication cost, the experiments are conducted with different number of users. For instance, for 100 VMs we choose 10, 20,...,80 users, and their request for VMs (number of VMs and type of VMs) are generated randomly, such that the total number of requested VMs do not exceed number of available VMs.

Performance comparison of multi-container deployment schemes for HPC workloads: an empirical study

The high-performance computing (HPC) community has recently started to use containerization to obtain fast, customized, portable, flexible, and reproducible deployments of their workloads. Previous work showed that deploying an HPC workload into a single container can keep bare-metal performance. However, there is a lack of research on multi-container deployments that partition the processes belonging to each application into different containers. Partitioning HPC applications has shown to improve their performance on virtual machines by allowing to set affinity to a non-uniform memory access (NUMA) domain for each of them. Consequently, it is essential to understand the performance implications of distinct multi-container deployment schemes for HPC workloads, focusing on the impact of the container granularity and its combination with processor and memory affinity. This paper presents a systematic performance comparison and analysis of multi-container deployment schemes for HPC workloads on a single-node platform, which considers different containerization technologies (including Docker and Singularity), two different platform architectures (UMA and NUMA), and two application subscription modes (exact subscription and over-subscription). Our results indicate that finer-grained multi-container deployments, on the one side, can benefit the performance of some applications with low interprocess communication, especially in over-subscribed scenarios and when combined with affinity, but, on the other side, they can incur some performance degradation for communication-intensive applications when using containerization technologies that deploy isolated network namespaces.

The transplantation technology of communication intensive applications on heterogeneous clusters

AbstractCompared with traditional clusters, heterogeneous clusters cherish the high performance to price ratio. However, compared with the rapid development of hardware technology, the current software technology is still lagging behind. This article introduces a parallel programming model to describe the multidimensional structure, arrangement, and communication modes of the data, threads, and fast Fourier transform. This model provides a new mechanism for transplantation between different types of heterogeneous systems. Using direct methods of turbulence simulation as an example, this article discusses the new software technologies of the parallel software theory, methods of program design, algorithm design, and system software support and provides experience for application transplantation on these heterogeneous systems.

An efficient multi-functional duplication-based scheduling framework for multiprocessor systems

Timing performance and energy consumption are the most critical performance indicators of schedules on multiprocessors. The timing performance of the schedule is limited by the inter-processor communication delay incurred by inter-task data dependence, especially for communication-intensive applications. To mitigate the negative impact of the communication delay on timing performance of the schedule, task duplication strategy, which is based on the fact that the local communication is delay-free, is often utilized. The duplication-based scheduling problem is NP-hard, in terms of optimizing both the schedule latency and energy consumption. However, obtaining the optimal solution is still useful, e.g., help to implement performance-critical system or give insights into the problem. While available works focus on limited performance metrics of the problem and have high time complexity, this paper proposes an efficient and multi-functional duplication-based scheduling framework based on the satisfiability modulo theory with boolean and integer theories, which enables solving the duplication-based scheduling problems quickly; besides, strategies for modeling the non-preemptive condition in static schedule are studied to improve the efficiency of the proposed framework. The proposed framework is applied to optimize different objectives, including the schedule length, the energy consumption, etc. The proposed framework is tested on a set of applications and platforms. The experimental results demonstrate that the proposed method is more than 10 times quicker than the available method and can reduce the energy consumption by up to 70%.

Multi-objective communication-aware optimization for virtual machine placement in cloud datacenters

This paper formulates a new multi-objective virtual machine placement (VMP) problem, which is a challenging task in cloud datacenters (DCs). In cloud environment, there are two stakeholders, namely, users and providers. Both sides try to take more benefit whereas a trade-off between conflicting benefits is crucial. From providers’ perspective, power consumption and resource wastage are two objectives to be optimized whereas gaining high quality of service (QoS) is a critical point for users. The unpleasant issue that a user endures in the cloud environment is network delay; this is affected by common bandwidth linkage which is shared between different users’ applications; the reason for considering bandwidth usage optimization as the third objective function in users’ viewpoint. However, inefficient network bandwidth usage has drastic impact on overall performance even makes network links to get saturated and throttles communication-intensive applications. Therefore, VMs with high affinity and traffic dependency must be physically placed as close as possible so less traffic is sent on network layers. To figure out this combinatorial multi-objective problem, we extend a hybrid multi-objective genetic-based optimization solution. To evaluate this work, we conducted extensive scenarios with variable correlation coefficients between resources in requested VMs. The simulation results prove that our proposed hybrid meta-heuristic algorithm outperforms against state-of-the-art ACO-based, well-known heuristic-based FFD algorithms, and random-based approach in terms of total power consumption, resource wastage, the total data transfer rate in network, and number of active servers in use. Also, the simulations in larger search space demonstrated proposed approach has high potential of scalability.

Performance drop at executing communication-intensive parallel algorithms

This work summarizes the results of a set of executions completed on three fat-tree network supercomputers: Stampede at TACC (USA), Helios at IFERC (Japan) and Eagle at PSNC (Poland). Three MPI-based, communication-intensive scientific applications compiled for CPUs have been executed under weak-scaling tests: the molecular dynamics solver LAMMPS; the finite element-based mini-kernel miniFE of NERSC (USA); and the three-dimensional fast Fourier transform mini-kernel bigFFT of LLNL (USA). The design of the experiments focuses on the sensitivity of the applications to rather different patterns of task location, to assess the impact on the cluster performance. The accomplished weak-scaling tests stress the effect of the MPI-based application mappings (concentrated vs. distributed patterns of MPI tasks over the nodes) on the cluster. Results reveal that highly distributed task patterns may imply a much larger execution time in scale, when several hundreds or thousands of MPI tasks are involved in the experiments. Such a characterization serves users to carry out further, more efficient executions. Also researchers may use these experiments to improve their scalability simulators. In addition, these results are useful from the clusters administration standpoint since tasks mapping has an impact on the cluster throughput.

Scheduling communication-intensive applications on Mesos

In recent years, the widespread use of container technologies has significantly altered the interactions between cloud service providers and their customers when developing and offering services. The shift away from Virtual Private Server scenarios in Infrastructure-as-a-Service environments requires drastic changes to the deployment strategies adopted by service providers. This also opens many questions to what information must be supplied by customers and how to improve the performance of user applications, especially in the case of communication-intensive applications. In this work, the authors propose the adoption of a new framework for Mesos clusters that aims to improve the deployment strategies of communication intensive applications. Coope4M is based on the partitioning of the user application graph via the Isolation Index parameter obtained through user-knowledge on the degree of the communication between its components.

Scheduling communication-intensive applications on Mesos

In recent years, the widespread use of container technologies has significantly altered the interactions between cloud service providers and their customers when developing and offering services. The shift away from Virtual Private Server scenarios in Infrastructure-as-a-Service environments requires drastic changes to the deployment strategies adopted by service providers. This also opens many questions to what information must be supplied by customers and how to improve the performance of user applications, especially in the case of communication-intensive applications. In this work, the authors propose the adoption of a new framework for Mesos clusters that aims to improve the deployment strategies of communication intensive applications. Coope4M is based on the partitioning of the user application graph via the Isolation Index parameter obtained through user-knowledge on the degree of the communication between its components.

Network Reconfiguration Algorithm (NRA) for scheduling communication-intensive graphs in heterogeneous computing environment

Distributed environments are widely used for computing complex applications modeled as task graphs. The computer network becomes more complex when the compute nodes are heterogeneous, however by choosing the appropriate network communication links for communication between a pair of compute tasks can enhance the computing efficiency(called network reconfiguration). One of the steps in heterogeneous network reconfiguration problem is mapping the application task graph edges on the network links. High Performance Computing (HPC) systems are usually heterogeneous, therefore mapping task graph edges on the communication links should consider the two factors: communication cost of task graph edges and the communication capability of network links. The system performance enhances if tasks are mapped on the compute nodes based on the computational costs of the tasks and the processing capability of compute nodes in addition to the edge scheduling on network links. In our earlier algorithm, Heterogeneous Edge and Task Scheduling (HETS) both edge and task mapping simultaneously improve the execution performance of task graphs. The proposed Network Reconfiguration Algorithm (NRA) minimizes the communication overhead and optimizes the schedule length with contention-aware model. NRA reduces an attribute Kirchhoff Index (KI) for optimal network reconfiguration providing minimum execution time. Both synthesized and task graphs of real applications are used for evaluation. The simulation results prove the efficiency of NRA in terms of average schedule length, schedule length ratio, speedup and system throughput. Comparisons with the baseline algorithms show that NRA provides 36% improved results specially for communication-intensive applications.

A Lifetime Reliability-Constrained Runtime Mapping for Throughput Optimization in Many-Core Systems

Due to technology scaling, lifetime reliability is becoming one of the major design constraints in the performance optimization of future many-core systems. Given a lifetime reliability constraint, the existing lifetime-constrained runtime mapping schemes often lead to low throughput because of the requirement to map all applications to compact regions. In this paper, we propose a runtime application mapping scheme that exploits a borrowing strategy to improve the throughput of many-core systems given a lifetime constraint. First, we propose using different strategies for mapping communication-intensive applications and computation-intensive applications. The lifetime reliability constraint can be relaxed in the local time scale when the communication requirement is high. The throughput is improved because the communication distance of communication-intensive applications is optimized while the waiting time of computation-intensive application is reduced. Then, we propose a method to effectively classify applications depending on the communication-to-computation ratio. A dynamic threshold is determined according to the current locations of available cores. Finally, we propose an improved neighborhood allocation scheme to reduce the communication cost in the task mapping. The experimental results show that compared to the state-of-the-art lifetime-constrained mapping, the proposed mapping scheme improves the throughput of many-core systems by 26% on average for synthetic task graphs and by 20% on average for realistic task graphs while the lifetime reliability is maintained within a constraint.

Experimenting SDN and Cloud Orchestration in Virtualized Testing Facilities: Performance Results and Comparison

Due to the impressive demand for communication-intensive applications deployed in the cloud, cross-layer orchestration solutions for Cloud data centers (DCs) are essential for the effective usage of both network and cloud resources while addressing service requirements and user expectations. In this regard, the software-defined networking (SDN) can play a key role thanks to programmable network control operations and to a more effective inter-working between network controllers and cloud management platforms. On the other hand, due to the relevant size of Cloud DCs, a significant evaluation of SDN-based orchestration solutions requires testing environments with proper scales, and significant level of fidelity. This paper discusses the opportunities of using virtual testbeds to carry out Cloud-related experimentations while addressing with significant scale requirements of DC infrastructures. First, we present a set of results of a test campaign we carried out using the Fed4FIRE experimental facility aiming at a comprehensive performance evaluation of an SDN-based orchestration solution for Cloud DCs. Second, we provide a comparison between virtual testbeds and other testing environments, i.e., laboratory testbeds, simulators and emulators, in terms of offered advantages, and effectiveness in experimentations. Moreover, we discuss the appropriateness and advisability of using virtualized testbeds for research on novel SDN and Cloud orchestration solutions at scale, by highlighting advantages of virtual testbeds in comparison with laboratory, simulation and emulation testing facilities in terms of effectiveness of experimentations and accuracy of results.

Pec: Proactive Elastic Collaborative Resource Scheduling in Data Stream Processing

In the Distributed Parallel Stream Processing Systems (DPSPS), elastic resource allocation allows applications to dynamically response to workload fluctuations. However, resource provisioning can be particularly challenging, due to the unpredictability of the workload. In addition, unlike CPU resources, bandwidth resources are often ignored in resource allocation. Moreover, resource allocation and resource placement are considered separately. In this paper, we investigate the proactive elastic resource scheduling problem for computation-intensive and communication-intensive applications, which aims at meeting the latency requirement with the minimal energy cost, and propose a dynamic collaborative strategy from the systemic perspective. Specifically, we first model a collaborative workload prediction pattern to accurately predict the upcoming workload, and construct a latency estimation model to estimate the latency of the application. Then, we design an energy-efficient resource pre-allocation method, in which the CPU frequency adjustment and the stability of resource reconfigurations are both considered. Finally, we present a communication-aware resource placement approach. Simulation results show that, compared with the reactive strategies, our strategy achieves an obviously better latency performance, and effectively avoids unnecessary resource adjustments. Meanwhile, the energy consumption is about saved by 50 percent on average, and the communication cost is maintained at a very low level of 4 percent.

MOPT: list-based heuristic for scheduling workflows in cloud environment

Cloud computing is a popular and widely adopted computing platform for the execution of scientific workflows as it provides flexible infrastructure and offers access to collection of autonomous heterogeneous resources. Effective scheduling of computationally complex workflows which contain many interconnected tasks is a complex problem and becomes more challenging in cloud environment. Optimal solutions can be obtained by considering not only the heterogeneity of computation costs involved, but also by taking into account the communication costs among the tasks in a way that schedule length of the application is reduced. In this paper, we propose a list scheduling heuristic, namely minimal optimistic processing time (MOPT), with optimized duplication approach. The additional feature is introduced for the entry task and is applied only in scenarios in which duplication is more practical and effective. The prioritization phase of the proposed work is based on an optimistic processing time matrix that is used for ranking of the tasks. The algorithm has same time complexity as state-of-the-art existing algorithms, but notable improvements are acquired in terms of makespan and other performance evaluation parameters. Extensive experimental analysis of the proposed algorithm is carried out using synthesized graphs and graphs from the real-world applications. The results prove that MOPT achieves quality schedules with reduced makespans. As communication cost among the tasks grows higher, performance of the proposed algorithm becomes more effective, thus providing the evidence that the MOPT algorithm is well-suited for communication-intensive applications.

Streaming graph partitioning

Graph partitioning is an essential yet challenging task for massive graph analysis in distributed computing. Common graph partitioning methods scan the complete graph to obtain structural characteristics offline, before partitioning. However, the emerging need for low-latency, continuous graph analysis led to the development of online partitioning methods. Online methods ingest edges or vertices as a stream, making partitioning decisions on the fly based on partial knowledge of the graph. Prior studies have compared offline graph partitioning techniques across different systems. Yet, little effort has been put into investigating the characteristics of online graph partitioning strategies. In this work, we describe and categorize online graph partitioning techniques based on their assumptions, objectives and costs. Furthermore, we employ an experimental comparison across different applications and datasets, using a unified distributed runtime based on Apache Flink. Our experimental results showcase that model-dependent online partitioning techniques such as low-cut algorithms offer better performance for communication-intensive applications such as bulk synchronous iterative algorithms, albeit higher partitioning costs. Otherwise, model-agnostic techniques trade off data locality for lower partitioning costs and balanced workloads which is beneficial when executing data-parallel single-pass graph algorithms.

Communication and cooling aware job allocation in data centers for communication-intensive workloads

Energy consumption is an increasingly important concern in data centers. Today, nearly half of the energy in data centers is consumed by the cooling infrastructure. Existing policies on thermally-aware workload allocation do not consider applications that include many tasks (or threads) running on a large set of nodes with significant communication among the tasks. Such jobs, however, constitute most of the cycles in high performance computing (HPC) domain, and have started to appear in other data centers as well. Job allocation strongly affects the performance of such communication-intensive applications. Communication-aware job allocation methods exist, but they focus solely on performance and do not consider cooling energy. This paper proposes a novel job allocation methodology to jointly minimize communication cost and cooling energy consumption in data centers. We formulate and solve the joint optimization problem using binary quadratic programming. Our joint optimization algorithm reduces cooling energy by 16.4% on average with only a 2.66% average increase in application running time compared to solely performance-aware allocations. To further optimize the communication cost, we develop a Charm++ based framework that extracts the communication behavior of applications. We then integrate our job allocation policy with recursive coordinate bisection (RCB) based task mapping method to place highly-communicating tasks in close proximity. Experimental results show that task mapping further decreases the communication cost by up to 20.9% compared to assuming all-to-all communication, a popular assumption in much of the prior work.

DwarfCode: A Performance Prediction Tool for Parallel Applications

We present DwarfCode, a performance prediction tool for MPI applications on diverse computing platforms. The goal is to accurately predict the running time of applications for task scheduling and job migration. First, DwarfCode collects the execution traces to record the computing and communication events. Then, it merges the traces from different processes into a single trace. After that, DwarfCode identifies and compresses the repeating patterns in the final trace to shrink the size of the events. Finally, a dwarf code is generated to mimic the original program behavior. This smaller running benchmark is replayed in the target platform to predict the performance of the original application. In order to generate such a benchmark, two major challenges are to reduce the time complexity of trace merging and repeat compression algorithms. We propose an O ( mpn ) trace merging algorithm to combine the traces generated by separate MPI processes, where m denotes the upper bound of tracing distance , p denotes the number of processes, and n denotes the maximum of event numbers of all the traces. More importantly, we put forward a novel repeat compression algorithm, whose time complexity is O ( nlogn ). Experimental results show that DwarfCode can accurately predict the running time of MPI applications. The error rate is below 10 percent for compute and communication intensive applications. This toolkit has been released for free download as a GNU General Public License v3 software.

Mobile device power models for energy efficient dynamic offloading at runtime

Spectacular advances in hardware and software technologies have resulted in powerful mobile devices, equipped with advanced processing, storage and network capabilities. Therefore, using resource-intensive applications has become a commodity in many contexts. However, the rapid evolution in hardware and software capabilities has not been paralleled by a similar advance in battery technology. A potential avenue to cope with the device energy resource limitation is to offload computational tasks to cloud infrastructure in the network. In order to offload tasks in an energy-aware manner, we present a detailed model of mobile device energy consumption, addressing the main power consuming subsystems, including CPU, display unit, wireless network interface and memory. Applying this model allows to estimate the power consumed by the application when executed locally, remotely or hybridly (i.e. partly on the device and partly in the cloud infrastructure). Offloading parts of the application can subsequently be decided at runtime based on these energy consumption estimates, also taking into account the power consumed by the device-to-cloud communication over the wireless network. The dynamic offloading has been validated with computational and communication intensive applications. Results show that 18–55% energy gains on the mobile device can be achieved, depending on different conditions.