Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSpeculative Container Scheduling for Deep Learning Applications in a Kubernetes Cluster
85
0
0.0
(
0
)
Added by
Ying Mao
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
In the past decade, we have witnessed a dramatically increasing volume of data collected from varied sources. The explosion of data has transformed the world as more information is available for collection and analysis than ever before. To maximize the utilization, various machine and deep learning models have been developed, e.g. CNN [1] and RNN [2], to study data and extract valuable information from different perspectives. While data-driven applications improve countless products, training models for hyperparameter tuning is still a time-consuming and resource-intensive process. Cloud computing provides infrastructure support for the training of deep learning applications. The cloud service providers, such as Amazon Web Services [3], create an isolated virtual environment (virtual machines and containers) for clients, who share physical resources, e.g., CPU and memory. On the cloud, resource management schemes are implemented to enable better sharing among users and boost the system-wide performance. However, general scheduling approaches, such as spread priority and balanced resource schedulers, do not work well with deep learning workloads. In this project, we propose SpeCon, a novel container scheduler that is optimized for shortlived deep learning applications. Based on virtualized containers, such as Kubernetes [4] and Docker [5], SpeCon analyzes the common characteristics of training processes. We design a suite of algorithms to monitor the progress of the training and speculatively migrate the slow-growing models to release resources for fast-growing ones. Specifically, the extensive experiments demonstrate that SpeCon improves the completion time of an individual job by up to 41.5%, 14.8% system-wide and 24.7% in terms of makespan.
rate research
Read More
With the prevalence of big-data-driven applications, such as face recognition on smartphones and tailored recommendations from Google Ads, we are on the road to a lifestyle with significantly more intelligence than ever before. For example, Aipoly Vision [1] is an object and color recognizer that helps the blind, visually impaired, and color blind understand their surroundings. At the back end side of their intelligence, various neural networks powered models are running to enable quick responses to users. Supporting those models requires lots of cloud-based computational resources, e.g. CPUs and GPUs. The cloud providers charge their clients by the amount of resources that they occupied. From clients perspective, they have to balance the budget and quality of experiences (e.g. response time). The budget leans on individual business owners and the required Quality of Experience (QoE) depends on usage scenarios of different applications, for instance, an autonomous vehicle requires realtime response, but, unlocking your smartphone can tolerate delays. However, cloud providers fail to offer a QoE based option to their clients. In this paper, we propose DQoES, a differentiate quality of experience scheduler for deep learning applications. DQoES accepts clients specification on targeted QoEs, and dynamically adjust resources to approach their targets. Through extensive, cloud-based experiments, DQoES demonstrates that it can schedule multiple concurrent jobs with respect to various QoEs and achieve up to 8x times more satisfied models compared to the existing system.
Kubernetes (k8s) has the potential to merge the distributed edge and the cloud but lacks a scheduling framework specifically for edge-cloud systems. Besides, the hierarchical distribution of heterogeneous resources and the complex dependencies among requests and resources make the modeling and scheduling of k8s-oriented edge-cloud systems particularly sophisticated. In this paper, we introduce KaiS, a learning-based scheduling framework for such edge-cloud systems to improve the long-term throughput rate of request processing. First, we design a coordinated multi-agent actor-critic algorithm to cater to decentralized request dispatch and dynamic dispatch spaces within the edge cluster. Second, for diverse system scales and structures, we use graph neural networks to embed system state information, and combine the embedding results with multiple policy networks to reduce the orchestration dimensionality by stepwise scheduling. Finally, we adopt a two-time-scale scheduling mechanism to harmonize request dispatch and service orchestration, and present the implementation design of deploying the above algorithms compatible with native k8s components. Experiments using real workload traces show that KaiS can successfully learn appropriate scheduling policies, irrespective of request arrival patterns and system scales. Moreover, KaiS can enhance the average system throughput rate by 14.3% while reducing scheduling cost by 34.7% compared to baselines.
Linux containers have gained high popularity in recent times. This popularity is significantly due to various advantages of containers over Virtual Machines (VM). The containers are lightweight, occupy lesser storage, have fast boot-up time, easy to deploy and have faster auto-scaling. The key reason behind the popularity of containers is that they leverage the mechanism of micro-service style software development, where applications are designed as independently deployable services. There are various container orchestration tools for deploying and managing the containers in the cluster. The prominent among them are Docker Swarm and Kubernetes. However, they do not address the effects of resource contention when multiple containers are deployed on a node. Moreover, they do not provide support for container migration in the event of an attack or increased resource contention. To address such issues, we propose C-Balancer, a scheduling framework for efficient placement of containers in the cluster environment. C-Balancer works by periodically profiling the containers and deciding the optimal container to node placement. Our proposed approach improves the performance of containers in terms of resource utilization and throughput. Experiments using a workload mix of Stress-NG and iPerf benchmark shows that our proposed approach achieves a maximum performance improvement of 58% for the workload mix. Our approach also reduces the variance in resource utilization across the cluster by 60% on average.
Businesses have made increasing adoption and incorporation of cloud technology into internal processes in the last decade. The cloud-based deployment provides on-demand availability without active management. More recently, the concept of cloud-native application has been proposed and represents an invaluable step toward helping organizations develop software faster and update it more frequently to achieve dramatic business outcomes. Cloud-native is an approach to build and run applications that exploit the cloud computing delivery models advantages. It is more about how applications are created and deployed than where. The container-based virtualization technology, such as Docker and Kubernetes, serves as the foundation for cloud-native applications. This paper investigates the performance of two popular computational-intensive applications, big data, and deep learning, in a cloud-native environment. We analyze the system overhead and resource usage for these applications. Through extensive experiments, we show that the completion time reduces by up to 79.4% by changing the default setting and increases by up to 96.7% due to different resource management schemes on two platforms. Additionally, the resource release is delayed by up to 116.7% across different systems. Our work can guide developers, administrators, and researchers to better design and deploy their applications by selecting and configuring a hosting platform.
Big data applications and analytics are employed in many sectors for a variety of goals: improving customers satisfaction, predicting market behavior or improving processes in public health. These applications consist of complex software stacks that are often run on cloud systems. Predicting execution times is important for estimating the cost of cloud services and for effectively managing the underlying resources at runtime. Machine Learning (ML), providing black box solutions to model the relationship between application performance and system configuration without requiring in-detail knowledge of the system, has become a popular way of predicting the performance of big data applications. We investigate the cost-benefits of using supervised ML models for predicting the performance of applications on Spark, one of todays most widely used frameworks for big data analysis. We compare our approach with textit{Ernest} (an ML-based technique proposed in the literature by the Spark inventors) on a range of scenarios, application workloads, and cloud system configurations. Our experiments show that Ernest can accurately estimate the performance of very regular applications, but it fails when applications exhibit more irregular patterns and/or when extrapolating on bigger data set sizes. Results show that our models match or exceed Ernests performance, sometimes enabling us to reduce the prediction error from 126-187% to only 5-19%.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
