Elastic Cloud Resources Research Articles

FedAdaSS: Federated Learning with Adaptive Parameter Server Selection Based on Elastic Cloud Resources

FedAdaSS: Federated Learning with Adaptive Parameter Server Selection Based on Elastic Cloud Resources

Showcasing Data Management Challenges for Future IoT Applications with NebulaStream

Data management systems will face several new challenges in supporting IoT applications during the coming years. These challenges arise from managing large numbers of heterogeneous IoT devices and require combining elastic cloud and fog resources in unified fog-cloud environments. In this demonstration, we introduce a smart city simulation called IoTropolis and use it to create interactive eHealth and Smart Grid application scenarios. We use these scenarios to showcase three key challenges of unified fog-cloud environments. Furthermore, we demonstrate how our recently proposed data management system for the IoT NebulaStream addresses these challenges. Visitors to our demonstration can configure and interact with the scenarios to manage electricity usage in IoTropolis or to distribute patients across different hospitals. Thereby, visitors can actively engage with the challenges showcased by IoTropolis and utilize NebulaStream to address them. As a result, our demonstration enables visitors to experience data management for future IoT applications.

Fast DRL-based scheduler configuration tuning for reducing tail latency in edge-cloud jobs

Edge-cloud applications are rapidly prevailing in recent years and pose the challenge of using both resource-strenuous edge devices and elastic cloud resources under dynamic workloads. Efficient resource allocation on edge-cloud jobs via cluster schedulers (e.g. Kubernetes/Volcano scheduler) is essential to guarantee their performance, e.g. tail latency, and such allocation is sensitive to scheduler configurations such as applied scheduling algorithms and task restart/discard policy. Deep reinforcement learning (DRL) is increasingly applied to optimize scheduling decisions. However, DRL faces the conundrum of achieving high rewards at a dauntingly long training time (e.g. hours or days), making it difficult to tune the scheduler configurations online in accordance to dynamically changing edge-cloud workloads and resources. For such an issue, this paper proposes EdgeTuner, a fast scheduler configuration tuning approach that efficiently leverages DRL to reduce tail latency of edge-cloud jobs. The enabling feature of EdgeTuner is to effectively simulate the execution of edge-cloud jobs under different scheduler configurations and thus quickly estimate these configurations’ influence on job performance. The simulation results allow EdgeTuner to timely train a DRL agent in order to properly tune scheduler configurations in dynamic edge-cloud environment. We implement EdgeTuner in both Kubernetes and Volcano schedulers and extensively evaluate it on real workloads driven by Alibaba production traces. Our results show that EdgeTuner outperforms prevailing scheduling algorithms by achieving much lower tail latency while accelerating DRL training speed by an average of 151.63x.

Scheduling Periodical Multi-Stage Jobs With Fuzziness to Elastic Cloud Resources

We investigate a workflow scheduling problem with stochastic task arrival times and fuzzy task processing times and due dates. The problem is common in many real-time and workflow-based applications, where tasks with fixed stage number and linearly dependency are executed on scalable cloud resources with multiple price options. The challenges lie in proposing effective, stable, and robust algorithms under stochastic and fuzzy tasks. A triangle fuzzy number-based model is formulated. Two metrics are explored: the cost and the degree of satisfaction. An iterated heuristic framework is proposed to periodically schedule tasks, which consists of a task collection and a fuzzy task scheduling phases. Two task collection strategies are presented and two task prioritization strategies are employed. In order to achieve a high satisfaction degree, deadline constraints are defined at both job and task levels. By designing delicate experiments and applying sophisticated statistical techniques, experimental results show that the proposed algorithm is more effective and robust than the two existing methods.

Effectiveness of Segment Routing Technology in Reducing the Bandwidth and Cloud Resources Provisioning Times in Network Function Virtualization Architectures

Network Function Virtualization is a new technology allowing for a elastic cloud and bandwidth resource allocation. The technology requires an orchestrator whose role is the service and resource orchestration. It receives service requests, each one characterized by a Service Function Chain, which is a set of service functions to be executed according to a given order. It implements an algorithm for deciding where both to allocate the cloud and bandwidth resources and to route the SFCs. In a traditional orchestration algorithm, the orchestrator has a detailed knowledge of the cloud and network infrastructures and that can lead to high computational complexity of the SFC Routing and Cloud and Bandwidth resource Allocation (SRCBA) algorithm. In this paper, we propose and evaluate the effectiveness of a scalable orchestration architecture inherited by the one proposed within the European Telecommunications Standards Institute (ETSI) and based on the functional separation of an NFV orchestrator in Resource Orchestrator (RO) and Network Service Orchestrator (NSO). Each cloud domain is equipped with an RO whose task is to provide a simple and abstract representation of the cloud infrastructure. These representations are notified of the NSO that can apply a simplified and less complex SRCBA algorithm. In addition, we show how the segment routing technology can help to simplify the SFC routing by means of an effective addressing of the service functions. The scalable orchestration solution has been investigated and compared to the one of a traditional orchestrator in some network scenarios and varying the number of cloud domains. We have verified that the execution time of the SRCBA algorithm can be drastically reduced without degrading the performance in terms of cloud and bandwidth resource costs.

Cost-Efficient Server Provisioning for Cloud Gaming

Cloud gaming has gained significant popularity recently due to many important benefits such as removal of device constraints, instant-on, and cross-platform. The properties of intensive resource demands and dynamic workloads make cloud gaming appropriate to be supported by an elastic cloud platform. Facing a large user population, a fundamental problem is how to provide satisfactory cloud gaming service at modest cost. We observe that the software storage cost could be substantial compared to the server running cost in cloud gaming using elastic cloud resources. Therefore, in this article, we address the server provisioning problem for cloud gaming to optimize both the server running cost and the software storage cost. We find that the distribution of game software among servers and the selection of server types both trigger tradeoffs between the software storage cost and the server running cost in cloud gaming. We formulate the problem with a stochastic model and employ queueing theory to conduct a solid theoretical analysis of the system behaviors under different request dispatching policies. We then propose several classes of algorithms to approximate the optimal solution. The proposed algorithms are evaluated by extensive experiments using real-world parameters. The results show that the proposed Ordered and Genetic algorithms are computationally efficient, nearly cost-optimal, and highly robust to dynamic changes.

JAVRE: A Joint Asymmetric Video Rendering and Encoding Approach to Enable Optimized Cloud Mobile 3D Virtual Immersive User Experience

With the growing adoption of mobile devices as the primary device for information and media consumption, and recent growth in availability of mobile devices with auto-stereoscopic 3D displays, the time has come to enable rich 3D virtual immersive experiences on mobile devices, like 3D display gaming, 3D virtual classroom, 3D virtual art gallery, etc., which are currently enabled on PC platforms. Instead of developing such compute-intensive applications on mobile devices, a scalable and portable approach is to take advantage of the elastic cloud resources. In this paper, we explore the possibility of developing a cloud based mobile 3D virtual immersive application architecture, where the 3D video rendering and encoding are performed on cloud servers, with the resulting 3D video streamed to mobile 3D devices through wireless networks. However, with the significantly higher bit rate requirement for 3D videos, ensuring user experience is a challenge considering the inherent dynamic variability of mobile networks. In this paper, in order to address this challenge, we propose a joint asymmetric graphics rendering and video encoding approach called JAVRE, where both the encoding quality and rendering richness of the left view and the right view can be asymmetric. Specifically, we first develop two mathematical models (user experience model and video bit rate model) through subjective and objective tests. Then we propose an optimization algorithm that can automatically choose the video encoding settings and graphics rendering settings for the left view and the right view to ensure the best user experience given the network conditions. Experiments conducted using real 4G-LTE network profiles and a commercial cloud service demonstrate significant improvement in user experience when JAVRE is used.

A coral-reefs and Game Theory-based approach for optimizing elastic cloud resource allocation

Elasticity is a key feature in cloud computing, which distinguishes this paradigm from other ones, such as cluster and grid computing. On the other hand, dynamic resource reallocation is one of the most important and complex issues in cloud scenarios, which can be expressed as a multi-objective optimization problem with the opposing objectives of maximizing demand satisfaction and minimizing costs and resource consumptions. In this paper, we propose a meta-heuristic approach for cloud resource allocation based on the bio-inspired coral-reefs optimization paradigm to model cloud elasticity in a cloud-data center, and on the classic Game Theory to optimize the resource reallocation schema with respect to cloud provider’s optimization objectives, as well as customer requirements, expressed through Service Level Agreements formalized by using a fuzzy linguistic method.

Cloud Mobile 3D Display Gaming User Experience Modeling and Optimization by Asymmetric Graphics Rendering

With the arrival of auto-stereoscopic 3D displays for mobile devices, and emergence of more 3D content, there is much anticipation for 3D mobile multimedia experiences, including 3D display gaming. Simultaneously, with the emergence of cloud computing, more mobile applications are being developed to take advantage of the elastic cloud resources. In this paper, we explore the possibility of developing Cloud Mobile 3D Display Gaming, where the 3D video rendering and encoding is performed on cloud servers, with the resulting 3D video streamed over wireless networks to mobile devices with 3D displays for a true 3D mobile gaming experience. However, with the significantly higher bitrate requirement for 3D video, ensuring user experience may be a challenge, both in terms of 3D video quality and network delay (response time), considering the bandwidth constraints and fluctuations of wireless networks. In this paper, we propose a new asymmetric graphics rendering approach which can significantly reduce the video encoding bitrate needed for a certain video quality, thereby making it easier to transmit the video over wireless network. However, since asymmetric graphics rendering may also impair the graphics quality, we need to be able to understand and measure its impact. We conduct subjective tests to study and model the impairments due to asymmetric graphics rendering and network delay, thereby developing a user experience model for cloud based mobile 3D display gaming. By conducting subsequent subjective tests, we prove the correctness of the impairment functions and the resulting user experience model. Furthermore, given any network condition, we propose to solve the problem of selecting the optimal graphics rendering factors for the left and right views so as to maximize user experience of cloud mobile 3D display gaming. In order to solve this problem, we first develop a model to estimate the resulting video bitrate of the rendered 3D video when certain graphics rendering factors are used. Next, we derive a model to predict the delay given the available network bandwidth and the video bitrate of the rendered 3D video. We use the above two models together with a branch and bound algorithm to solve the optimization problem and determine the optimal values for the left and right view rendering factors. Experiments conducted using real 4G-LTE network profiles on commercial cloud service demonstrate the feasibility of significant improvement in user experience when the proposed optimization algorithm is used to dynamically select optimal rendering factors according to changing network conditions.

Reactive Resource Provisioning Heuristics for Dynamic Dataflows on Cloud Infrastructure

The need for low latency analysis over high-velocity data streams motivates the need for distributed continuous dataflow systems. Contemporary stream processing systems use simple techniques to scale on elastic cloud resources to handle variable data rates. However, application QoS is also impacted by variability in resource performance exhibited by clouds and hence necessitates autonomic methods of provisioning elastic resources to support such applications on cloud infrastructure. We develop the concept of “dynamic dataflows” which utilize alternate tasks as additional control over the dataflow’s cost and QoS. Further, we formalize an optimization problem to represent deployment and runtime resource provisioning that allows us to balance the application’s QoS , value , and the resource cost . We propose two greedy heuristics, centralized and sharded , based on the variable-sized bin packing algorithm and compare against a Genetic Algorithm (GA) based heuristic that gives a near-optimal solution. A large-scale simulation study, using the linear road benchmark and VM performance traces from the AWS public cloud, shows that while GA-based heuristic provides a better quality schedule, the greedy heuristics are more practical, and can intelligently utilize cloud elasticity to mitigate the effect of variability, both in input data rates and cloud resource performance, to meet the QoS of fast data applications.

Elastic Business Process Management: State of the art and open challenges for BPM in the cloud

With the advent of cloud computing, organizations are nowadays able to react rapidly to changing demands for computational resources. Not only individual applications can be hosted on virtual cloud infrastructures, but also complete business processes. This allows the realization of so-called elastic processes, i.e., processes which are carried out using elastic cloud resources. Despite the manifold benefits of elastic processes, there is still a lack of solutions supporting them.In this paper, we identify the state of the art of elastic Business Process Management with a focus on infrastructural challenges. We conceptualize an architecture for an elastic Business Process Management System and discuss existing work on scheduling, resource allocation, monitoring, decentralized coordination, and state management for elastic processes. Furthermore, we present two representative elastic Business Process Management Systems which are intended to counter these challenges. Based on our findings, we identify open issues and outline possible research directions for the realization of elastic processes and elastic Business Process Management.

Clouds + Games: A Multifaceted Approach

The computer game landscape is changing: people play games on multiple computing devices with heterogeneous form-factors, capability, and connectivity. Providing high playability on such devices concurrently is difficult. To enhance the gaming experience, designers could leverage abundant and elastic cloud resources, but current cloud platforms aren't optimized for highly interactive games. Existing studies focus on streaming-based cloud gaming, which is a special case for the more general cloud game architecture. The authors explain how to integrate techniques from the cloud and game research communities into a complete architecture for enhanced online gaming quality. They examine several open issues that appear only when clouds and games are put together.

A family of heuristics for agent-based elastic Cloud bag-of-tasks concurrent scheduling

The scheduling and execution of bag-of-tasks applications (BoTs) in Clouds is performed on sets of virtualized Cloud resources that start being exhausted right after their allocation disregarding whether tasks are being executed. In addition, BoTs may be executed in potentially heterogeneous sets of Cloud resources, which may be either previously allocated for a different and fixed number of hours or dynamically reallocated as needed. In this paper, a family of 14 scheduling heuristics for concurrently executing BoTs in Cloud environments is proposed. The Cloud scheduling heuristics are adapted to the resource allocation settings (e.g., 1-hour time slots) of Clouds by focusing on maximizing Cloud resource utilization based on the remaining allocation times of Cloud resources. Cloud scheduling heuristics supported by information about BoT tasks (e.g., task size) and/or Cloud resource performances are proposed. Additionally, scheduling heuristics that require no information of either Cloud resources or tasks are also proposed. The Cloud scheduling heuristics support the dynamic inclusion of new Cloud resources while scheduling and executing a given BoT without rescheduling. Furthermore, an elastic Cloud resource allocation mechanism that autonomously and dynamically reallocates Cloud resources on demand to BoT executions is proposed. Moreover, an agent-based Cloud BoT scheduling approach that supports concurrent and parallel scheduling and execution of BoTs, and concurrent and parallel dynamic selection and composition of Cloud resources (by making use of the well-known contract net protocol) from multiple and distributed Cloud providers is designed and implemented. Empirical results show that BoTs can be (i) efficiently executed by attaining similar (in some cases shorter) makespans to commonly used benchmark heuristics (e.g., Max–min), (ii) effectively executed by achieving a 100% success execution rate even with high BoT execution request rates and executing BoTs in a concurrent and parallel manner, and that (iii) BoTs are economically executed by elastically reallocating Cloud resources on demand.