Modern Cyber-physical Systems (CPS) include applications like smart traffic, smart agriculture, smart power grid, etc. Commonly, these systems are distributed and composed of end-user applications and microservices that typically run in the cloud. The connection with the physical world, which is inherent to CPS, brings the need to operate and respond in real-time. As the cloud becomes part of the computation loop, the real-time requirements have to be also reflected by the cloud. In this paper, we present an approach that provides soft real-time guarantees on the response time of services running in cloud and edge-cloud (i.e., cloud geographically close to the end-user), where these services are developed in high-level programming languages. In particular, we elaborate a method that allows us to predict the upper bound of the response time of a service when sharing the same computer with other services. Importantly, as our approach focuses on minimizing the impact on the developer of such services, it does not require any special programming model nor limits usage of common libraries, etc.
Managing cloud services is a fundamental challenge in todays virtualized environments. These challenges equally face both providers and consumers of cloud services. The issue becomes even more challenging in virtualized environments that support mobile clouds. Cloud computing platforms such as Amazon EC2 provide customers with flexible, on demand resources at low cost. However, they fail to provide seamless infrastructure management and monitoring capabilities that many customers may need. For instance, Amazon EC2 doesn't fully support cloud services automated discovery and it requires a priva…
This work introduces a runtime model for managing communication with support for latency-hiding. The model enables non-computer science researchers to exploit communication latency-hiding techniques seamlessly. For compiled languages, it is often possible to create efficient schedules for communication, but this is not the case for interpreted languages. By maintaining data dependencies between scheduled operations, it is possible to aggressively initiate communication and lazily evaluate tasks to allow maximal time for the communication to finish before entering a wait state. We implement a h…
Interactions between cloud services result in service dependencies. Evaluating and managing the cascading impacts caused by service dependencies is critical to the reliability of cloud systems. This paper summarizes the dependency types in cloud systems and demonstrates the design of the Dependency Management System (DMS), a platform for managing the service dependencies in the production cloud system. DMS features full-lifecycle support for service reliability (i.e., initial service deployment, service upgrade, proactive architectural optimization, and reactive failure mitigation) and refined…
Many scientific high-throughput applications can benefit from the elastic nature of Cloud resources, especially when there is a need to reduce time to completion. Cost considerations are usually a major issue in such endeavors, with networking often a major component; for data-intensive applications, egress networking costs can exceed the compute costs. Dedicated network links provide a way to lower the networking costs, but they do add complexity. In this paper we provide a description of a 100 fp32 PFLOPS Cloud burst in support of IceCube production compute, that used Internet2 Cloud Connect…
Multi-cloud computing is a promising paradigm to support very large scale world wide distributed applications. Multi-cloud computing is the usage of multiple, independent cloud environments, which assumed no priori agreement between cloud providers or third party. However, multi-cloud computing has to face several key challenges such as portability, provisioning, elasticity, and high availability. Developers will not only have to deploy applications to a specific cloud, but will also have to consider application portability from one cloud to another, and to deploy distributed applications span…
Ensuring data correctness over partitioned distributed database systems is a classical problem. Classical solutions proposed to solve this problem are mainly adopting locking or blocking techniques. These techniques are not suitable for cloud environments as they produce terrible response times; due to the long latency and faultiness of wide area network connections among cloud datacenters. One way to improve performance is to restrict access of users-bases to specific datacenters and avoid data sharing between datacenters. However, conflicts might appear when data is replicated between datace…
Management of data in education sector particularly management of data for big universities with several employees, departments and students is a very challenging task. There are also problems such as lack of proper funds and manpower for management of such data in universities. Education sector can easily and effectively take advantage of cloud computing skills for management of data. It can enhance the learning experience as a whole and can add entirely new dimensions to the way in which education is imbibed. Several benefits of Cloud computing such as monetary benefits, environmental benefi…
Outsourcing jobs to a public cloud is a cost-effective way to address the problem of satisfying the peak resource demand when the local cloud has insufficient resources. In this paper, we study on managing deadline-constrained bag-of-tasks jobs on hybrid clouds. We present a binary nonlinear programming (BNP) problem to model the hybrid cloud management where the utilization of physical machines (PMs) in the local cloud/cluster is maximized when the local resources are enough to satisfy the deadline constraints of jobs, while when not, the rent cost from the public cloud is minimized. To solve…
This chapter presents: (i) a layered peer-to-peer Cloud provisioning architecture; (ii) a summary of the current state-of-the-art in Cloud provisioning with particular emphasis on service discovery and load-balancing; (iii) a classification of the existing peer-to-peer network management model with focus on extending the DHTs for indexing and managing complex provisioning information; and (iv) the design and implementation of novel, extensible software fabric (Cloud peer) that combines public/private clouds, overlay networking and structured peer-to-peer indexing techniques for supporting scal…
Energy consumption of Cloud data centers has been a major concern of many researchers, and one of the reasons for huge energy consumption of Clouds lies in the inefficient utilization of computing resources. Besides energy consumption, another challenge of data centers is the unexpected loads, which leads to the overloads and performance degradation. Compared with VM consolidation and Dynamic Voltage Frequency Scaling that cannot function well when the whole data center is overloaded, brownout has shown to be a promising technique to handle both overloads and energy consumption through dynamic…
Rapid adoption of Cloud computing for hosting services and its success is primarily attributed to its attractive features such as elasticity, availability and pay-as-you-go pricing model. However, the huge amount of energy consumed by cloud data centers makes it to be one of the fastest growing sources of carbon emissions. Approaches for improving the energy efficiency include enhancing the resource utilization to reduce resource wastage and applying the renewable energy as the energy supply. This work aims to reduce the carbon footprint of the data centers by reducing the usage of brown energ…
This is a survey paper that explores six Cloud-based deployment archetypes for Cloud applications and the tradeoffs between them to achieve high availability, low end-user latency, and acceptable costs. These are (1) Zonal, (2) Regional, (3) Multi-Regional, (4) Global, (5) Hybrid, and (6) Multi-Cloud deployment archetypes. The goal is to classify cloud applications into a set of deployment archetypes and deployment models that tradeoff their needs around availability, latency, and geographical constraints with a focus on serving applications. This enables application owners to better examine t…
Some mission critical systems, e.g., fraud detection, require accurate, real-time metrics over long time sliding windows on applications that demand high throughput and low latencies. As these applications need to run 'forever' and cope with large, spiky data loads, they further require to be run in a distributed setting. We are unaware of any streaming system that provides all those properties. Instead, existing systems take large simplifications, such as implementing sliding windows as a fixed set of overlapping windows, jeopardizing metric accuracy (violating regulatory rules) or latency (b…
With the wide adoption of large-scale Internet services and big data, the cloud has become the ideal environment to satisfy the ever-growing storage demand, thanks to its seemingly limitless capacity, high availability and faster access time. In this context, data replication has been touted as the ultimate solution to improve data availability and reduce access time. However, replica placement systems usually need to migrate and create a large number of data replicas over time between and within data centers, incurring a large overhead in terms of network load and availability. In this paper,…
Various Cloud layers have to work in concert in order to manage and deploy complex multi-cloud applications, executing sophisticated workflows for Cloud resource deployment, activation, adjustment, interaction, and monitoring. While there are ample solutions for managing individual Cloud aspects (e.g. network controllers, deployment tools, and application security software), there are no well-integrated suites for managing an entire multi cloud environment with multiple providers and deployment models. This paper presents the CYCLONE architecture that integrates a number of existing solutions …
In this paper, the fundamental problem of distribution and proactive caching of computing tasks in fog networks is studied under latency and reliability constraints. In the proposed scenario, computing can be executed either locally at the user device or offloaded to an edge cloudlet. Moreover, cloudlets exploit both their computing and storage capabilities by proactively caching popular task computation results to minimize computing latency. To this end, a clustering method to group spatially proximate user devices with mutual task popularity interests and their serving cloudlets is proposed.…
Duplication can be a powerful strategy for overcoming stragglers in cloud services, but is often used conservatively because of the risk of overloading the system. We present duplicate-aware scheduling or DAS, which makes duplication safe and easy to use, by leveraging the two well-known primitives of prioritization and purging. To support DAS across diverse layers of a cloud system (e.g., network, storage, etc), we propose the D-Stage abstraction, which decouples the duplication policy from the mechanism, and facilitates working with legacy layers of a system. Using this abstraction, we evalu…
Training a Convolutional Neural Network (CNN) model typically requires significant computing power, and cloud computing resources are widely used as a training environment. However, it is difficult for CNN algorithm developers to keep up with system updates and apply them to their training environment due to quickly evolving cloud services. Thus, it is important for cloud computing service vendors to design and deliver an optimal training environment for various training tasks to lessen system operation management overhead of algorithm developers. To achieve the goal, we propose PROFET, which …
We present a framework for performance optimization in serverless edge-cloud platforms using dynamic task placement. We focus on applications for smart edge devices, for example, smart cameras or speakers, that need to perform processing tasks on input data in real to near-real time. Our framework allows the user to specify cost and latency requirements for each application task, and for each input, it determines whether to execute the task on the edge device or in the cloud. Further, for cloud executions, the framework identifies the container resource configuration needed to satisfy the perf…
Edge-assisted Parallel Uncertain Skyline Processing for Low-latency IoE Analysis
National Chung Cheng University（国立中正大学）；Trend Micro Inc（趋势科技）；Taiwan Semiconductor Manufacturing Company, Ltd (TSMC)（台湾积体电路制造股份有限公司）；National Taipei University of Technology（台北科技大学）
Recent years have witnessed a rapid growth of deep-network based services and applications. A practical and critical problem thus has emerged: how to effectively deploy the deep neural network models such that they can be executed efficiently. Conventional cloud-based approaches usually run the deep models in data center servers, causing large latency because a significant amount of data has to be transferred from the edge of network to the data center. In this paper, we propose JALAD, a joint accuracy- and latency-aware execution framework, which decouples a deep neural network so that a part…
Low latency is highly desirable for cloud services. To achieve low response time, stringent timing requirements are needed for task scheduling in a large-scale server farm spanning thousands of servers. In this paper, we conduct an in-depth analysis for distributed Join-the-Idle-Queue (JIQ), a promising new approximation of an idealized task-scheduling algorithm. In particular, we derive semi-closed form expressions for the delay performance of distributed JIQ, and we propose a new variant of distributed JIQ that offers clear advantages over alternative algorithms for large systems.
Modern latency-critical online services often rely on composing results from a large number of server components. Hence the tail latency (e.g. the 99th percentile of response time), rather than the average, of these components determines the overall service performance. When hosted on a cloud environment, the components of a service typically co-locate with short batch jobs to increase machine utilizations, and share and contend resources such as caches and I/O bandwidths with them. The highly dynamic nature of batch jobs in terms of their workload types and input sizes causes continuously cha…
The Grid and Cloud User Support Environment (gUSE) enables users convenient and easy access to grid and cloud infrastructures by providing a general purpose, workflow-oriented graphical user interface to create and run workflows on various Distributed Computing Infrastructures (DCIs). Its arrangements for creating and modifying existing workflows are, however, non-intuitive and cumbersome due to the technologies and architecture employed by gUSE. In this paper, we outline the first integrated web-based workflow editor for gUSE with the aim of improving the user experience for those with indust…
Edge computing has enabled a large set of emerging edge applications by exploiting data proximity and offloading latency-sensitive and computation-intensive workloads to nearby edge servers. However, supporting edge application users at scale in wide-area environments poses challenges due to limited point-of-presence edge sites and constrained elasticity. In this paper, we introduce Armada: a densely-distributed edge cloud infrastructure that explores the use of dedicated and volunteer resources to serve geo-distributed users in heterogeneous environments. We describe the lightweight Armada ar…
Existing approaches to tolerate Byzantine faults in geo-replicated environments require systems to execute complex agreement protocols over wide-area links and consequently are often associated with high response times. In this paper we address this problem with Spider, a resilient replication architecture for geo-distributed systems that leverages the availability characteristics of today's public-cloud infrastructures to minimize complexity and reduce latency. Spider models a system as a collection of loosely coupled replica groups whose members are hosted in different cloud-provided fault d…
The in-memory cache system is an important component in a cloud for the data access performance. As the tenants may have different performance goals for data access depending on the nature of their tasks, effectively managing the memory cache is a crucial concern in such a shared computing environment. Two extreme methods for managing the memory cache are unlimited sharing and complete isolation, both of which would be inefficient with the expensive storage complexity to meet the per-tenant performance requirement. In this paper, we present a new cache model that incorporates global caching (b…
In cloud storage, the digital data is stored in logical storage pools, backed by heterogeneous physical storage media and computing infrastructure that are managed by a Cloud Service Provider (CSP). To balance the tradeoff between service performance and cost, CSPs often employ different storage tiers, for instance, cold storage and hot storage. Storing data in hot storage incurs high storage cost yet delivers low access latency, whereas cold storage is able to inexpensively store massive amounts of data and thus provides lower cost with higher latency. In this paper, we address a major challe…
Modern latency-critical online services such as search engines often process requests by consulting large input data spanning massive parallel components. Hence the tail latency of these components determines the service latency. To trade off result accuracy for tail latency reduction, existing techniques use the components responding before a specified deadline to produce approximate results. However, they may skip a large proportion of components when load gets heavier, thus incurring large accuracy losses. This paper presents AccuracyTrader that produces approximate results with small accur…
Cloud-computing shares a common pool of resources across customers at a scale that is orders of magnitude larger than traditional multi-user systems. Constituent physical compute servers are allocated multiple "virtual machines" (VM) to serve simultaneously. Each VM user should ideally be unaffected by others' demand. Naturally, this environment produces new challenges for the service providers in meeting customer expectations while extracting an efficient utilization from server resources. We study a new cloud service metric that measures prolonged latency or delay suffered by customers. We m…
Generation Quality-Latency Tradeoff-Aware Inference Offloading for Multimodal LLMs in Cloud-Edge Continuum
School of Computer Science and Technology, Xidian University（西安电子科技大学计算机科学与技术学院）；School of Computer Science and Technology, Xi’an Jiaotong University（西安交通大学计算机科学与技术学院）；School of Software Technology, Zhejiang University（浙江大学软件学院）；School of Cyber Science and Engineering, Xi’an Jiaotong University（西安交通大学网络空间安全学院）；School of Computer Science and Engineering, University of New South Wales（新南威尔士大学计算机科学与工程学院）
Speculative Policy Orchestration: A Latency-Resilient Framework for Cloud-Robotic Manipulation
Department of Computing Science, Umeå University（乌梅大学计算科学系）；School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology（皇家理工学院电气工程与计算机科学学院）
Function as a Service (FaaS) permits cloud customers to deploy to cloud individual functions, in contrast to complete virtual machines or Linux containers. All major cloud providers offer FaaS products (Amazon Lambda, Google Cloud Functions, Azure Serverless); there are also popular open-source implementations (Apache OpenWhisk) with commercial offerings (Adobe I/O Runtime, IBM Cloud Functions). A new feature of FaaS is function composition: a function may (sequentially) call another function, which, in turn, may call yet another function - forming a chain of invocations. From the perspective …
The increase in data storage and power consumption at data-centers has made it imperative to design energy efficient Distributed Storage Systems (DSS). The energy efficiency of DSS is strongly influenced not only by the volume of data, frequency of data access and redundancy in data storage, but also by the heterogeneity exhibited by the DSS in these dimensions. To this end, we propose and analyze the energy efficiency of a heterogeneous distributed storage system in which $n$ storage servers (disks) store the data of $R$ distinct classes. Data of class $i$ is encoded using a $(n,k_{i})$ erasu…
Achieving high performance in virtualized data centers requires both deploying high throughput storage clusters, i.e. based on Solid State Disks (SSDs), as well as optimally consolidating the workloads across storage nodes. Nowadays, the only practical solution for cloud storage providers to offer guaranteed performance is to grossly over-provision the storage nodes. The current workload scheduling mechanisms used in production do not have the intelligence to optimally allocate block storage volumes based on the performance of SSDs. In this paper, we introduce Serifos, an autonomous performanc…
In cloud computing systems, assigning a job to multiple servers and waiting for the earliest copy to finish is an effective method to combat the variability in response time of individual servers. Although adding redundant replicas always reduces service time, the total computing time spent per job may be higher, thus increasing waiting time in queue. The total time spent per job is also proportional to the cost of computing resources. We analyze how different redundancy strategies, for eg. number of replicas, and the time when they are issued and canceled, affect the latency and computing cos…
University of Maryland, Baltimore County（马里兰大学巴尔的摩郡分校）；DEVCOM Army Research Laboratory（DEVCOM陆军研究实验室）
With the popularity of Internet of Things (IoT), edge computing and cloud computing, more and more stream analytics applications are being developed including real-time trend prediction and object detection on top of IoT sensing data. One popular type of stream analytics is the recurrent neural network (RNN) deep learning model based time series or sequence data prediction and forecasting. Different from traditional analytics that assumes data are available ahead of time and will not change, stream analytics deals with data that are being generated continuously and data trend/distribution coul…
Computational resource provisioning that is closer to a user is becoming increasingly important, with a rise in the number of devices making continuous service requests and with the significant recent take up of latency-sensitive applications, such as streaming and real-time data processing. Fog computing provides a solution to such types of applications by bridging the gap between the user and public/private cloud infrastructure via the inclusion of a "fog" layer. Such approach is capable of reducing the overall processing latency, but the issues of redundancy, cost-effectiveness in utilizing…
In cloud computing systems, assigning a task to multiple servers and waiting for the earliest copy to finish is an effective method to combat the variability in response time of individual servers, and reduce latency. But adding redundancy may result in higher cost of computing resources, as well as an increase in queueing delay due to higher traffic load. This work helps understand when and how redundancy gives a cost-efficient reduction in latency. For a general task service time distribution, we compare different redundancy strategies in terms of the number of redundant tasks, and time when…
