Forecasting the Energy Consumption of Cloud Data Centers Based on Container Placement with Ant Colony Optimization and Bin Packing

Bouaouda, Amine; Afdel, Karim; Abounacer, Rachida

doi:10.1109/ciot53061.2022.9766522

Cited by 9 publications

(17 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is being approached in the literature from many points of view, both in the workload specification and in the forecasting model itself. In the following, we will review some representative recent work to illustrate this variety of approaches [44][45][46][47][48].…”

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

“…how to reproduce the history of resource allocation, usage and release when assigning tasks to VMs, two approaches stand out. The first one consists of describing a synthetic workload, either in a static form [44], or from estimated resource life-cycle probabilities [47]. The second, more widespread, considers time series recorded in real data centers annotated with the relevant events [45,46,48].…”

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

“…• Bouaouda et al compare two algorithms taken from the area of operational research to estimate the energy consumed by a cloud data center, namely Ant-Colony Optimization, a population-based metaheuristics, and First Fit Decreasing, an algorithm to solve the Bin Packing problem [44]. The relevant workload events are generated by Cloudsim, a federated cloud data center simulator [49].…”

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

See 2 more Smart Citations

L2C2: Last-level compressed-contents non-volatile cache and a procedure to forecast performance and lifetime

et al. 2023

View full text Add to dashboard Cite

Several emerging non-volatile (NV) memory technologies are rising as interesting alternatives to build the Last-Level Cache (LLC). Their advantages, compared to SRAM memory, are higher density and lower static power, but write operations wear out the bitcells to the point of eventually losing their storage capacity. In this context, this paper presents a novel LLC organization designed to extend the lifetime of the NV data array and a procedure to forecast in detail the capacity and performance of such an NV-LLC over its lifetime. From a methodological point of view, although different approaches are used in the literature to analyze the degradation of an NV-LLC, none of them allows to study in detail its temporal evolution. In this sense, this work proposes a forecasting procedure that combines detailed simulation and prediction, allowing an accurate analysis of the impact of different cache control policies and mechanisms (replacement, wear-leveling, compression, etc.) on the temporal evolution of the indices of interest, such as the effective capacity of the NV-LLC or the system IPC. We also introduce L2C2, a LLC design intended for implementation in NV memory technology that combines fault tolerance, compression, and internal write wear leveling for the first time. Compression is not used to store more blocks and increase the hit rate, but to reduce the write rate and increase the lifetime during which the cache supports near-peak performance. In addition, to support byte loss without performance drop, L2C2 inherently allows N redundant bytes to be added to each cache entry. Thus, L2C2+N, the endurance-scaled version of L2C2, allows balancing the cost of redundant capacity with the benefit of longer lifetime. For instance, as a use case, we have implemented the L2C2 cache with STT-RAM technology. It has affordable hardware overheads compared to that of a baseline NV-LLC without compression in terms of area, latency and energy consumption, and increases up to 6-37 times the time in which 50% of the effective capacity is degraded, depending on the variability in the manufacturing process. Compared to L2C2, L2C2+6 which adds 6 bytes of redundant capacity per entry, that means 9.1% of storage overhead, can increase up to 1.4-4.3 times the time in which the system gets its initial peak performance degraded.

show abstract

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

Section: Forecasting Workload Behavior In Cloud Data Centers: a Seemi...mentioning

confidence: 99%

See 1 more Smart Citation

L2C2: Last-level compressed-contents non-volatile cache and a procedure to forecast performance and lifetime

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Because it is essential to identify improvement points and optimize the system's functioning. The study by Bouaouda et al [42] discusses energy consumption performance in a data centre using the CloudSim simulator, emphasizing placing containers on hosts, comparing the performance of two algorithms to minimize energy consumption in Cloud systems.…”

Section: Macro Metricsmentioning

confidence: 99%

Energy Consumption in Microservices Architectures: A Systematic Literature Review

Araújo,

Sabino,

Lima

et al. 2023

Preprint

View full text Add to dashboard Cite

Cloud computing is a paradigm that delivers technology resources on demand through service virtualization. Virtualization allows sharing of hardware by applications compartmentalized into multiple independent modules. Microservices architecture uses the ability of virtualization to adopt a software architecture approach based on building applications composed of several interdependent and loosely coupled modules. However, microservices architectures can incur large energy costs to achieve the required levels of performance and availability. Several research topics related to energy consumption in microservice architectures have already been explored, such as elasticity, reliability, performance, and availability. However, given the variety of problems and solutions, identifying research trends and gaps related to microservices is complex. We aim to systematically identify, rank, and compare existing research trends and gaps around energy consumption in microservices. This paper presents a systematic review of energy consumption in microservices architectures. This work also provides a list of references to find the mapped works, thus being able to guide searches more safely. paper selection started with 3625 papers, followed by three refinement processes in which 37 papers were selected for a full review. The remaining 37 papers were cataloged and classified according to some criteria, such as metrics, evaluation method and type of architecture. Thus, the review revealed research gaps and trends in energy consumption in microservices architectures. In addition, significant research challenges and directions were identified. The research challenges are presented based on the metrics that guide each work in this systematic review: performance, elasticity, scalability, reliability, sustainability, and availability.

show abstract

“…This paper is an extension of work originally presented in the Fifth Conference on Cloud and Internet of Things (CIoT) [1]. By adding other methods such as the Genetic Algorithm, First-Fit, Random-Fit, and Simulated Annealing, our approach will predict the energy consumed by the data centers and the workload placement (containers) in the servers, offering best and optimal solutions to reduce energy consumption, waste of cloud resources, and have an energy-efficient container placement policy.…”

Section: Introductionmentioning

confidence: 99%

Meta-heuristic and Heuristic Algorithms for Forecasting Workload Placement and Energy Consumption in Cloud Data Centers

Bouaouda¹,

Afdel²,

Abounacer³

2023

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

The increase of servers in data centers has become a significant problem in recent years that leads to a rise in energy consumption. The problem of high energy consumed by data centers is always related to the active hardware especially the servers that use virtualization to create a cloud workspace for the users. For this reason, workload placement such as virtual machines or containers in servers is an essential operation that requires the adoption of techniques that offer practical and best solutions for the workload placement that guarantees an optimization in the use of material resources and energy consumption in the cloud. In this article, we propose an approach that uses heuristics and meta-heuristics to predict cloud container placement and power consumption in data centers using a Genetic Algorithm (GA) and First Fit Decreasing (FFD). Our algorithms have been tested on CloudSim and the results showed that our methods gave better and more efficient solutions, especially the Genetic Algorithm after comparing them with Ant Colony Optimization (ACO) and Simulated Annealing (SA).

show abstract

Forecasting the Energy Consumption of Cloud Data Centers Based on Container Placement with Ant Colony Optimization and Bin Packing

Cited by 9 publications

References 18 publications

L2C2: Last-level compressed-contents non-volatile cache and a procedure to forecast performance and lifetime

L2C2: Last-level compressed-contents non-volatile cache and a procedure to forecast performance and lifetime

Energy Consumption in Microservices Architectures: A Systematic Literature Review

Meta-heuristic and Heuristic Algorithms for Forecasting Workload Placement and Energy Consumption in Cloud Data Centers

Contact Info

Product

Resources

About