Application-Agnostic Batch Workload Management in Cloud Environments

Javadi, Seyyed Ahmad; Bhaskara, Shalini; Doshi, Rahul N.; Soundarapandian, Prashanth; Wajahat, Muhammad; Gandhi, Anshul

doi:10.1145/3267809.3275446

Cited by 5 publications

(16 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding virtualization, only three of the listed works use VMs to contain the tenant applications. Two of these approaches [6,12] also use Linux KVM to deploy VMs. Some approaches use containers, which allow better performance at the expense of worse isolation.…”

Section: Related Workmentioning

confidence: 99%

“…Additionally, the manager is adapted to allow implementing and applying QoS policies. For instance, policies that manage resource sharing among VMs [7,11,12], predict interference among VMs [8,9,13,15,52] or schedule VMs [53].…”

Section: )mentioning

confidence: 99%

“…The development of a representative experimental framework is challenging since such a platform must provide three key attributes: i) it should include the main types of hardware nodes (server, client, and storage), ii) it should offer VM-based isolation for tenant applications, and iii) it should support the partitioning of major shared resources, including main memory, last level cache (LLC), network components, etc. However, most frameworks in existing research works present shortcomings such as: i) being composed of a single machine [6][7][8][9][10], ii) lack of virtualization support [6][7][8], and iii) neglecting the management of important components such as the network infrastructure [6][7][8] and the remote storage [7,8,[11][12][13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Modular Approach to Design an Experimental Framework for Resource Management Research

Pons,

Petit,

Pons

et al. 2023

Preprint

View full text Add to dashboard Cite

Research on resource management focuses on optimizing system performance and energy efficiency by distributing shared resources like processor cores, caches, and main memory among competing applications. This research spans a wide range of applications, including those from high-performance computing, machine learning, and mobile computing. Existing research frameworks often simplify research by concentrating on specific characteristics, such as the architecture of the computing nodes, resource monitoring, and representative workloads. For instance, this is typically the case with cloud systems, which introduce additional complexity regarding hardware and software requirements. To avoid this complexity during research, experimental frameworks are being developed. Nevertheless, proposed frameworks often fail regarding the types of nodes included, virtualization support, and management of critical shared resources. This paper presents Stratus, an experimental framework that overcomes these limitations. Stratus includes different types of nodes, a comprehensive virtualization stack, and the ability to partition the major shared resources of the system. Even though Stratus was originally conceived to perform cloud research, its modular design allows Stratus to be extended, broadening its research use on different computing domains and platforms, matching the complexity of nowadays cloud environments.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: )mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Modular Approach to Design an Experimental Framework for Resource Management Research

Pons,

Petit,

Pons

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…3 Nonetheless, issues about job assignment for asynchronous workloads or batch jobs, a particular instance of these issues, also hold in the contemporaneous systems. 4,5 Modern cloud infrastructures provide users with great flexibility in allocating computing resources for distributed systems, but they still need to be concerned about better operating the available services. 6 In this context, using microservices has been proven to be more than a trend for optimized resource management and system delivery.…”

Section: Introductionmentioning

confidence: 99%

“…At the same time, service management in cloud environments has also been a continuous area of investigation as software development and delivery techniques advanced 3 . Nonetheless, issues about job assignment for asynchronous workloads or batch jobs, a particular instance of these issues, also hold in the contemporaneous systems 4,5 . Modern cloud infrastructures provide users with great flexibility in allocating computing resources for distributed systems, but they still need to be concerned about better operating the available services 6 .…”

Section: Introductionmentioning

confidence: 99%

Managing asynchronous workloads in a multi‐tenant microservice enterprise environment

Batista,

Morais,

Cavalcante

et al. 2023

Softw Pract Exp

View full text Add to dashboard Cite

A multi‐tenant microservice architecture involving components with asynchronous interactions and batch jobs requires efficient strategies for managing asynchronous workloads. This article addresses this issue in the context of a leading company developing tax software solutions for many national and multi‐national corporations in Brazil. A critical process provided by the company's cloud‐based solutions encompasses tax integration, which includes coordinating complex tax calculation tasks and needs to be supported by asynchronous operations using a message broker to guarantee order correctness. We explored and implemented two approaches for managing asynchronous workloads related to tax integration within a multi‐tenant microservice architecture in the company's context: (i) a polling‐based approach that employs a queue as a distributed lock (DL) and (ii) a push‐based approach named single active consumer (SAC) that relies on the message broker's logic to deliver messages. These approaches aim to achieve efficient resource allocation when dealing with a growing number of container replicas and tenants. In this article, we evaluate the correctness and performance of the DL and SAC approaches to shed light on how asynchronous workloads impact the management of multi‐tenant microservice architectures from delivery and deployment perspectives.

show abstract

Genetic-Based Virtual Machines Consolidation Strategy With Efficient Energy Consumption in Cloud Environment

2023

View full text Add to dashboard Cite

In cloud computing environments, virtualization is used to share physical machine (PM) resources among multiple users by creating virtual machines (VMs). Running the PM consumes a large amount of energy. Additionally, the PM will be overloaded when the demand for resources exceeds the PM capacity. This overload on the PM leads to violations of Service Level Agreements (SLAs). Dynamic VM consolidation techniques use live migration of VMs to optimize resource utilization and minimize energy consumption. However, excessive migration of VMs impacts negatively the application performance due to the incurred overhead at the runtime. This paper presents a modified genetic-based VM consolidation (MGVMC) strategy that aims to replace VMs in an online manner taking into account energy consumption, SLA violations, and the number of VM migrations. The MGVMC strategy utilizes the genetic algorithm to migrate VMs to the appropriate PM in a way that minimizes the number of over-utilized and under-utilized physical machines (PMs) as low as possible. The performance of the MGVMC strategy was evaluated using the CloudSim Plus framework with a large number of VMs and workload traces from the PlanetLab platform. The experimental results revealed that the MGVMC strategy achieved a significant improvement in energy consumption, SLA violations, and the number of VM migrations compared to other recent approaches. These results demonstrate the effectiveness of the MGVMC strategy in optimizing VM consolidation in the cloud environment.

show abstract

Application-Agnostic Batch Workload Management in Cloud Environments

Cited by 5 publications

References 3 publications

A Modular Approach to Design an Experimental Framework for Resource Management Research

A Modular Approach to Design an Experimental Framework for Resource Management Research

Managing asynchronous workloads in a multi‐tenant microservice enterprise environment

Genetic-Based Virtual Machines Consolidation Strategy With Efficient Energy Consumption in Cloud Environment

Contact Info

Product

Resources

About