Maximize Application Performance Research Articles

Towards constrained optimization of cloud applications: A hybrid approach

A database system is one of the most common cloud applications where elasticity allows to dynamically allocate resources in response to changing workload demands. In such systems, users usually configure resources based on empirical decisions, for example by boosting computational resources with the expectation of improving database throughput. The latter also suggests that users manually scale resources and tune the database configurations offline to meet the application demands. However, this approach could rapidly increase infrastructure expenditures while this is also time consuming. In this paper, we propose SONA, a framework to support constrained performance optimization of cloud applications using a hybrid approach of artificial neural networks and genetic algorithms. The proposed framework monitors the source system to identify the optimal configurations that maximize application performance based on genuine workload executions. As a result, the experimental analysis presents a novel dataset collected from TPC-C runs on MySQL server. The optimization process is being constrained to satisfy user and application requirements including the cloud infrastructure expenditures, the cost-performance ratio, the baseline performance and the average percentage of idle CPU resources. Furthermore, SONA uses a cloned containerized environment that replicates the main application to avoid system overhead during the optimization process. Our results demonstrate the effectiveness of SONA framework to optimize the performance of OLTP applications deployed on cloud.

BarMan: A run-time management framework in the resource continuum

Over the last years, the number of IoT devices has grown exponentially, highlighting the current Cloud infrastructure limitations. In this regard, Fog and Edge computing began to move part of the computation closer to data sources by exploiting interconnected devices as part of a single heterogeneous and distributed system in a computing continuum viewpoint. Since these devices are typically heterogeneous in terms of performance, features, and capabilities, this perspective should encompass programming models and run-time management layers. This work presents and evaluates the BarMan open-source framework by implementing a Fog video surveillance use-case. BarMan leverages a task-based programming model combined with a run-time resource manager and the novel BeeR framework to deploy the application’s tasks transparently. This enables the possibility of considering aspects related to the energy and power dissipated by the devices and the single application. Moreover, we developed a task allocation policy to maximize application performance, considering run-time aspects, such as load and connectivity, of the time-varying available devices. Through an experimental evaluation performed on a real cluster equipped with heterogeneous embedded boards, we evaluated different execution scenarios to show the framework’s functionality and the benefit of a distributed approach, leading up to an improvement of 66% on the frame processing latency w.r.t. a monolithic solution.

Runtime power allocation approach for GAMESS hybrid CPU‐GPU implementation

SummaryTo improve power consumption of applications at the runtime, modern processors provide frequency scaling capabilities, which along with workload optimization, are also available on GPU accelerators. In this work, a runtime strategy is proposed to distribute a given power allocation among the host components and the GPU according to the current application performance and power usage, such that GPU execution is prioritized over CPU for power allocation to maximize application performance. Next, the strategy is tailored to an application, a quantum‐chemistry package GAMESS for ab initio electronic structure calculations. Specifically, GAMESS hybrid CPU–GPU implementation as provided in the Libcchem library is considered. Experiments, performed on a 28‐core node with a Kepler GPU, resulted in performance gains of up to 50% under the proposed strategy and the largest power allocation considered here as compared with the scenario when this allocation was equally distributed among the computing‐platform components.

Trust-Based Service Management for Mobile Cloud IoT Systems

We propose and analyze a 3-tier cloud-cloudlet-device hierarchical trust-based service management protocol called IoT-HiTrust for large-scale mobile cloud Internet of Things (IoT) systems. Our mobile cloud hierarchical service management protocol allows an IoT customer to report its service experiences and query its subjective service trust score toward an IoT service provider following a scalable report-and-query design. We conduct a formal scalability analysis along with an ns-3 simulation performance analysis demonstrating that IoT-HiTrust not only achieves scalability without compromising accuracy, convergence, and resiliency properties against malicious attacks but also outperforms contemporary distributed and centralized IoT trust management protocols. We test the feasibility by applying IoT-HiTrust to two case studies: 1) a smart city travel service composition and binding application and 2) an air pollution detection and response application. The results demonstrate that IoT-HiTrust outperforms contemporary distributed and centralized trust-based IoT service management protocols in selecting trustworthy nodes to maximize application performance, while achieving scalability.

Efficient Performance-Centric Bandwidth Allocation with Fairness Tradeoff

Fair bandwidth allocation in datacenter networks has received a substantial amount of research attention, as multiple tenants are hosted by virtual machines in a public cloud. In the context of private datacenters, link bandwidth is shared among applications running data parallel frameworks, such as MapReduce, instead. In this paper, we introduce the rigorous definition of performance-centric fairness , with the guiding principle that the performance that data parallel applications will enjoy should be proportional to their weights. We first investigate the problem of maximizing application performance while maintaining strict performance-centric fairness. We then present an inherent tradeoff between fairness and efficiency, which is interpreted from the perspectives of bandwidth utilization and social welfare, respectively. From the first perspective, we propose an algorithm to improve bandwidth utilization by introducing an extended version of fairness. From the second perspective, we formulate an optimization problem of bandwidth allocation that maximizes the social welfare across all the applications, allowing a tunable degree of relaxation on performance-centric fairness. A distributed algorithm is then presented to solve the problem, based on dual based decomposition. With extensive simulations, we demonstrate the effectiveness of our algorithms in improving efficiency and application performance (by up to $1.4X$ ), with flexible degree of relaxation on the performance-centric fairness.

Sensor Network Configuration Learning for Maximizing Application Performance.

Numerous applications rely on data obtained from a wireless sensor network where application performance is of utmost importance. However, energy usage is also important, and oftentimes, a subset of sensors can be selected to maximize application performance. We cast the problem of sensor selection as a local search optimization problem and solve it using a variant of stochastic hill climbing extended with novel heuristics. This paper introduces sensor network configuration learning, a feedback-based heuristic algorithm that dynamically reconfigures the sensor network to maximize the performance of the target application. The proposed algorithm is described in detail, along with experiments conducted and a scalability study. A quick method for launching the algorithm from a better starting point than random is also detailed. The performance of the algorithm is compared to that of two other well-known algorithms and randomness. Our simulation results obtained from running sensor network configuration learning on a number of scenarios show the effectiveness and scalability of our approach.

FoToNoC: A Folded Torus-Like Network-on-Chip Based Many-Core Systems-on-Chip in the Dark Silicon Era

Dark silicon refers to the phenomenon that a fraction of a many-core chip has to become “dark” or “dim” in order to guarantee the system to be kept in a safe temperature range and allowable power budget. Techniques have been developed to selectively activate non-adjacent cores on many-core chip to avoid temperature hotspot, while resulting unexpected increase of communication overhead due to the longer average distance between active cores, and in turn affecting application performance and energy efficiency, when Network-on-Chip (NoC) is used as a scalable communication subsystem. To address the brand-new challenges brought by dark silicon, in this paper, we present FoToNoC, a Folded Torus-like NoC, coupled with a hierarchical management strategy for heterogeneous many-core systems. On top of it, objectives of maximizing application performance, energy efficiency and chip reliability are isolated and well achieved by hardware-software co-design in several different phases, including application mapping and scheduling, cluster management and DVFS control. Evaluations on PARSEC benchmark applications demonstrate the significance of the entire strategy. Compared with state-of-the-art approaches, the proposed FoToNoC organization can achieve on average 35.4 and 35.2 percent on communication efficiency and application performance improvement, respectively, when maintaining the safe chip temperature. The hierarchical cluster-based management strategy can further reduce an average 34.6 percent of the total energy consumption with a notable reduction on the chip peak temperature. The significant achievements on system energy efficiency and the reduction on chip temperature of H.264 decoder and DSP-stone benchmarks additionally verify the effectiveness of the proposed methods.

Trace: a high-throughput tomographic reconstruction engine for large-scale datasets

BackgroundModern synchrotron light sources and detectors produce data at such scale and complexity that large-scale computation is required to unleash their full power. One of the widely used imaging techniques that generates data at tens of gigabytes per second is computed tomography (CT). Although CT experiments result in rapid data generation, the analysis and reconstruction of the collected data may require hours or even days of computation time with a medium-sized workstation, which hinders the scientific progress that relies on the results of analysis.MethodsWe present Trace, a data-intensive computing engine that we have developed to enable high-performance implementation of iterative tomographic reconstruction algorithms for parallel computers. Trace provides fine-grained reconstruction of tomography datasets using both (thread-level) shared memory and (process-level) distributed memory parallelization. Trace utilizes a special data structure called replicated reconstruction object to maximize application performance. We also present the optimizations that we apply to the replicated reconstruction objects and evaluate them using tomography datasets collected at the Advanced Photon Source.ResultsOur experimental evaluations show that our optimizations and parallelization techniques can provide 158× speedup using 32 compute nodes (384 cores) over a single-core configuration and decrease the end-to-end processing time of a large sinogram (with 4501 × 1 × 22,400 dimensions) from 12.5 h to <5 min per iteration.ConclusionThe proposed tomographic reconstruction engine can efficiently process large-scale tomographic data using many compute nodes and minimize reconstruction times.

Hierarchical trust management of community of interest groups in mobile ad hoc networks

In mission-critical applications deployed in mobile ad hoc networks, very frequently a commander will need to assemble and dynamically manage Community of Interest (COI) mobile groups to achieve the mission assigned despite failure, disconnection or compromise of COI members. In this paper, we present a dynamic hierarchical trust management protocol that can learn from past experiences and adapt to changing environment conditions (e.g., increasing misbehaving node population, evolving hostility and node density, etc.) to enhance agility and maximize application performance. With trust-based misbehaving node detection as an application, we demonstrate how our proposed COI trust management protocol is resilient to node failure, disconnection and capture events, and can help maximize application performance in terms of minimizing false negatives and positives in the presence of mobile nodes exhibiting vastly distinct QoS and social behaviors.

Virtual machine placement with two-path traffic routing for reduced congestion in data center networks

Virtualization-based Data Centers are increasingly becoming the hosting platform for a wide range of applications. The communication patterns in Data Center networks show the trend towards increasing bandwidth usage between virtual machines (VMs) within the Data Center resulting in higher chance of occurrence of network congestion. Thus, VM placement and routing algorithms are increasingly important to maximize application performance, provide fault tolerance, and reduce network loads. A less-than optimal placement of communicating VMs can cause inter-VM traffic to traverse bottlenecked network paths leading to large cross network traffic. The core network oversubscription and unbalanced workload placement could lead to long-lived congestion in Data Center networks. Multipath routing with traffic distributed in an appropriate proportion helps balance the load and decrease the possibility of congestion. Furthermore, by routing traffic on multiple link-disjoint paths, traffic can be protected against failures. The use of link-disjoint paths ensures the availability of at least one path for the traffic upon a link failure, thus guaranteeing a certain bandwidth (associated with the surviving paths). In this paper, we study the problem of VM placement with traffic routing on multiple paths for reduced occurrence of congestion while satisfying a certain protection grade which is defined as the fraction of rate (or bandwidth) guaranteed to be available in the event of single link failures. We develop an efficient algorithm based on a greedy technique for placing VMs onto servers satisfying the computing and memory resource requirements, taking into account the amount of inter-VM traffic and network load. In addition, we develop a two-path routing algorithm to satisfy the bandwidth and protection grade requirements so as to reduce the network congestion. Our simulation results show the effectiveness of the proposed algorithms in balancing the load and resilient when compared to other first-fit and random algorithms.

Hierarchical Trust Management for Wireless Sensor Networks and its Applications to Trust-Based Routing and Intrusion Detection

We propose a highly scalable cluster-based hierarchical trust management protocol for wireless sensor networks (WSNs) to effectively deal with selfish or malicious nodes. Unlike prior work, we consider multidimensional trust attributes derived from communication and social networks to evaluate the overall trust of a sensor node. By means of a novel probability model, we describe a heterogeneous WSN comprising a large number of sensor nodes with vastly different social and quality of service (QoS) behaviors with the objective to yield "ground truth" node status. This serves as a basis for validating our protocol design by comparing subjective trust generated as a result of protocol execution at runtime against objective trust obtained from actual node status. To demonstrate the utility of our hierarchical trust management protocol, we apply it to trust-based geographic routing and trust-based intrusion detection. For each application, we identify the best trust composition and formation to maximize application performance. Our results indicate that trust-based geographic routing approaches the ideal performance level achievable by flooding-based routing in message delivery ratio and message delay without incurring substantial message overhead. For trust-based intrusion detection, we discover that there exists an optimal trust threshold for minimizing false positives and false negatives. Furthermore, trust-based intrusion detection outperforms traditional anomaly-based intrusion detection approaches in both the detection probability and the false positive probability.

ADFT: An Adaptive Framework for Fault Tolerance on Large Scale Systems using Application Malleability

Exascale systems of the future are predicted to have mean time between failures (MTBF) of less than one hour. Malleable applications, where the number of processors on which the applications execute can be changed during executions, can make use of their malleability to better tolerate high failure rates. We present AdFT, an adaptive fault tolerance framework for long running malleable applications to maximize application performance in the presence of failures. AdFT framework includes cost models for evaluating the benefits of various fault tolerance actions including checkpointing, live-migration and rescheduling, and runtime decisions for dynamically selecting the fault tolerance actions at different points of application execution to maximize performance. Simulations with real and synthetic failure traces show that our approach outperforms existing fault tolerance mechanisms for malleable applications yielding up to 23% improvement in application performance, and is effective even for petascale systems and beyond.

Optimizing a virtualized data center

Many data centers extensively use virtual machines (VMs), which provide the flexibility to move workload among physical servers. VMs can be placed to maximize application performance, power efficiency, or even fault tolerance. However, VMs are typically repositioned without considering network topology, congestion, or traffic routes. In this demo, we show a system, Virtue, which enables the comparison of different algorithms for VM placement and network routing at the scale of an entire data center. Our goal is to understand how placement and routing affect overall application performance by varying the types and mix of workloads, network topologies, and compute resources; these parameters will be available for demo attendees to explore.