Intelligent Pooling: Proactive Resource Provisioning in Large-scale Cloud Service

Effort estimation models have been long studied in software engineering research. Effort estimation models help organizations and individuals plan and track progress of their software projects and individual tasks to help plan delivery milestones better. Towards this end, there is a large body of work that has been done on effort estimation for projects but little work on an individual checkin (Pull Request) level. In this paper we present a methodology that provides effort estimates on individual developer check-ins which is displayed to developers to help them track their work items. Given the cloud development infrastructure pervasive in companies, it has enabled us to deploy our Pull Request Lifetime prediction system to several thousand developers across multiple software families. We observe from our deployment that the pull request lifetime prediction system conservatively helps save 44.61% of the developer time by accelerating Pull Requests to completion.

Cloud computing is provisioned with high flexibility with regard to on demand infrastructures, platforms and software as services through the Internet. The unique characteristics of cloud services such as dynamic and diverse services offering at different levels, as well as the lack of standardized description, are becoming important challenges in efficiently discovering cloud services for customers. In this paper, we propose a cloud service search engine that has the capability to automatically identify cloud services aiming at improving the accuracy when searching cloud services in real environments. Our search engine can detect cloud services effectively from the Web sources. Furthermore, we focus on learning the cloud service features, such as similarity function, semantic ontology and cloud service components to identify the cloud services. We use a real cloud service dataset to build an identifier. Our cloud service identifier can be used to automatically determine whether a given Web source is a cloud service with high accuracy.

As a core of the cloud computing, PaaS (Platform as a Service) provides a computing and software service platform to support large-scale cloud services. This paper presents one cloud comput...

Big, impactful efforts are full of challenges. Building new infrastructure in the public cloud brings six challenges: new technology, distributed data, integration, scale, achieving situational awareness, and new human processes. Teams need organizing principles to help them manage these risks and to successfully deliver results. Large teams, numbering in the hundreds and distributed around the globe, need clear and consistent methods to inventory, assess, and mitigate risks. Organizations have tried many different methods to manage risk, and they have experienced differing levels of success.

With the continuous evolution of microservice architecture and containerization technology, the challenge of efficiently and reliably scheduling large-scale cloud services has become increasingly prominent. In this paper, we present a cost-optimized scheduling approach with resource configuration for microservice workflows in container environments, taking into account deadline and reliability constraints. We introduce a graph deep learning model (DeepMCC) that automatically configures containers to meet various service quality (QoS) requirements. Additionally, we propose a reliability microservice workflow scheduling algorithm (RMWS), which incorporates heuristic leasing and deployment strategies to ensure reliability while reducing cloud resource leasing cost. Experiments on four scientific workflow datasets show that the proposed approach achieves an average cost reduction of 44.59% compared to existing reliability scheduling algorithms, with improvements of 26.63% in the worst case and 73.72% in the best case.

Remote Direct Memory Access (RDMA) has been widely deployed in datacenters for its high performance. Large-scale high performance cloud services built on geographically distributed datacenters require long-range RDMA for performance requirements. However, existing RDMA solutions can hardly satisfy the stringent requirements of the emerging large-scale high-performance cloud services built on geo-distributed datacenters in terms of throughput and delay. On the one hand, lossless RDMA suffers from a deep buffer and potential suboptimal throughput for inter-datacenter traffic due to delayed response to Priority Flow Control (PFC) messages. On the other hand, lossy RDMA with selective retransmissions suffers from poor performance when multiple flows with different round-trip times (RTTs) coexist in cross-datacenter scenarios. This article proposes <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Swing</small> , which expands the high-performance lossless RDMA to long-distance links through PFC-Relay. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Swing</small> ensures the throughput of long-distance links while minimizing the buffer requirement for long-range RDMA. It enables long-range RDMA without making any modifications to existing in-datacenter networks. The evaluation shows that <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Swing</small> can reduce the average flow completion time (FCT) by 14%-66% in a variety of traffic scenarios.

Mobile Cloud computing has shown its capability to support mobile devices for provisioning computing, storage and communication resources. Many existing research has proposed to offload computation tasks from mobile devices to clouds in order to reduce energy consumption, where the offloading service model is usually one-to-one. Due to the development of mobile sensing and location-based mobile cloud services, the cloud edge has been extended to the mobile devices and sensors. As a result, the one-to-one model is not sufficient to model the dynamic changes of mobile cloud-based services. Thus, a many-to-many (or multi-site) mobile cloud service composition is highly desired. In this research, MoSeC is presented to model the many-to-many mobile cloud service composition, where there are multiple surrogates, such as cloud computing nodes, mobile devices, or sensors, and their services (i.e., computation, storage, sensing, etc.) can be composed to fulfill functions required by a mobile service requestor. MoSeC takes into considerations the surrogates' changes due to their mobility and resource constraints. Moreover, MoSeC takes into considerations several service metrics for service mapping (or allocation) through a mobile cloud service topology reconfiguration process. A set of algorithms are presented to address the Service Topology Reconfiguration Problem (STRP) in several mobile cloud representative application scenarios, i.e., they are modeled as finite horizon scenarios, infinite horizon scenarios, and large state space scenarios to represent ad hoc, long-term, and large-scale mobile cloud service composition scenarios, respectively.

In recent years, physical sensor devices are used in many different applications, such as: Internet of Things (IoT), military, vehicle system, etc. Sensor devices are widely utilized in the real world so that a large-scale process becomes an important issue in such a cyber physical system. Large scale cyber physical system (LCPS) must have enough computing power for processing all real-time sensed data from sensor devices. LCPS can deploy a cloud platform to increase its computing power. The cloud based LCPS then serves these requests at the same time by well arranging sensor tasks based on the arrival time. Hence, cloud service can obtain better performance and more efficiency when good scheduling algorithms are implemented. In this paper, apply the early deadline first until zero laxity (EDZL) scheduling algorithm to optimize the cloud service scheduler for ensuring sensor jobs do not miss deadline in the LCPS.

Replicating data across geo-distributed datacenters is usually necessary for large scale cloud services to achieve high locality, durability and availability. One of the major challenges in such geo-replicated data services lies in consistency maintenance, which usually suffers from long latency due to costly coordination across datacenters. Among others, transaction chopping is an effective and efficient approach to address this challenge. However, existing chopping is conducted statically during programming, which is stubborn and complex for developers. In this article, we propose Dynamic Transaction Chopping (DTC), a novel technique that does transaction chopping and determines piecewise execution in a dynamic and automatic way. DTC mainly consists of two parts: a dynamic chopper to dynamically divide transactions into pieces according to the data partition scheme, and a conflict detection algorithm to check the safety of the dynamic chopping. Compared with existing techniques, DTC has several advantages: transparency to programmers, flexibility in conflict analysis, high degree of piecewise execution, and adaptability to data partition schemes. A prototype of DTC is implemented to verify the correctness of DTC and evaluate its performance. The experiment results show that our DTC technique can achieve much better performance than similar work.

Blockchain model for environment/infrastructure monitoring in cloud-enabled high-altitude platform systems

Large-scale interactive services usually divide requests into multiple sub-requests and distribute them to a large number of server components for parallel execution. Hence the tail latency (i.e. The slowest component's latency) of these components determines the overall service latency. On a cloud platform, each component shares and competes node resources such as caches and I/O bandwidths with its co-located jobs, hence inevitably suffering from their performance interference. In this paper, we study the short-running jobs in a 12k-node Google cluster to illustrate the dynamic resource demands of these jobs, resulting in both individual components' latency variability over time and across different nodes and hence posing a major challenge to maintain low tail latency. Given this motivation, this paper introduces a dynamic and interference-aware scheduler for large-scale, parallel cloud services. At each scheduling interval, it collects workload and resource contention information of a running service, and predicts both the component latency on different nodes and the overall service performance. Based on the predicted performance, the scheduler identifies straggling components and conducts near-optimal component-node allocations to adapt to the changing workloads and performance interferences. We demonstrate that, using realistic workloads, the proposed approach achieves significant reductions in tail latency compared to the basic approach without scheduling.

In recent years, cloud computing has emerged as a successful model to offer computing resources in an on-demand manner for large-scale Internet services and applications. However many companies are still reluctant to embrace the cloud. This chapter provides a comprehensive study of fundamental concepts and techniques related to survivability and reliability in cloud computing environments. It first lays out key concepts of the cloud computing model and concepts related to survivability. Then it presents an overview of the outcomes of recent analyses of failures in the cloud. The chapter further reviews and discusses existing techniques aimed at improving fault tolerance and availability of cloud services. The ultimate goal is to develop a comprehensive understanding of state-of-the-art solutions for improving cloud survivability and reliability, and to provide insights into the critical challenges to be addressed in the future.

By leveraging the Internet, cloud computing allows users to have on-demand access to large pools of configurable computing resources. PaaS (Platform as a Service), IaaS (Infrastructure as a Service), and SaaS (Software as a Service) are three basic categories for the services provided by cloud the computing environments. Quality of service (QoS) metrics like reliability, availability, performance, and cost determine which resources and services are available in a cloud computing scenario. Provider and the user-specified performance characteristics, such as, rejection rate, throughput, response time, financial cost, and energy consumption, form the basis for QoS. To fulfil the needs of its customers, cloud computing must ensure that its services are given with the appropriate quality of service QoS. A “A legally enforceable agreement known as a “Service Level Agreement” (SLA) between a service provider and a customer that outlines service objectives, quality of service requirements, and any associated financial penalties for falling short. We, therefore, presented “A Proactive Resource Supply based Run-time Monitoring of SLA in Cloud Computing”, which allows for the proactive management of SLAs during run-time via the provisioning of cloud services and resources. Within the framework of the proposed work, SLAs are negotiated between cloud users and providers at run-time utilizing SLA Manager. Resources are proactively allocated via the Resource Manager to cut down on SLA violations and misdetection costs. As metrics of performance, we looked at the frequency with which SLAs were broken and the money lost due to false positives. We compared the proposed PRP-RM-SLA model’s simulated performance to the popular existing SLA-based allocation strategy SCOOTER. According to simulation data, the suggested PRP-RM-SLA model is 25% more effective than the current work SCOOTER at reducing SLA breaches and the cost of misdetection.

Cloud service provider, infrastructure vendor and clients/Cloud user’s are main actors in any cloud enterprise like Amazon web service’s cloud or Google’s cloud. Now these enterprises take care in infrastructure deployment and cloud services management (IaaS/PaaS/SaaS). Cloud user ‘s need to provide correct amount of services needed and characteristic of workload in order to avoid over – provisioning of resources and it’s the important pricing factor. Cloud service provider need to manage the resources and as well as optimize the resources to maximize the profit. To manage the profit we consider the M/M/m queuing model which manages the queue of job and provide average execution time. Resource Scheduling is one of the main concerns in profit maximization for which we take HYBRID PSO-MOBA as it resolves the global convergence problem, faster convergence, less parameter to tune, easier searching in very large problem spaces and locating the right resource. In HYBRID PSO-MOBA we are combining the features of PSO and MOBA to achieve the benefits of both PSO and MOBA and have greater compatibility.

Due to development in cloud services, lots of data is transferred between users and servers of the cloud. This transmission of data consumes huge amounts of energy. This energy consumption occurs during the operation of network infrastructure, the conversion of electrical to optical signals to travel long distances, and signal amplification. As Green computing is the use of computing devices in an environmentally friendly way, i.e., using electrical energy efficiently as Data centers require a significant amount of electricity to operate and cool the servers, leading to carbon emissions from the burning of fossil fuels. Green computing in cloud services is about optimizing energy consumption and by incorporating deep learning algorithms, we can enhance the energy efficiency of cloud infrastructure. These algorithms can analyze real-time data from sensors, optimize resource allocation, and dynamically adjust power usage. Through intelligent workload scheduling, server consolidation, and power management, deep learning enables the reduction of energy waste and carbon emissions. The integration of deep learning in cloud services not only improves energy efficiency but also enhances performance and cost-effectiveness. Here we are using a deep learning model which can be used for workload prediction and resource provisioning. By analyzing historical workload patterns and user behavior, deep learning algorithmscan predict future resource demands and allocate resources accordingly, leading to more efficient resource utilization and energy savings