检索结果-内蒙古大学图书馆

On Performance VNF Load Prediction Models in Service function Chaining 11

On Performance VNF Load Prediction Models in Service Functio...

11th International Conference on Information and Communication Technology Convergence (ICTC) - Data, network, and AI in the age of Untact (ICTC)

作者： Cho, Yunyoung Jang, Seokwon Pack, Sangheon Korea Univ Sch Elect & Engn Seoul South Korea

ISBN: (纸本)9781728167589

One of the fundamental challenges in network function virtualization (NFV) is how to autonomously and dynamically allocate resources to virtualized network functions (VNFs) whose resource requirements frequently changed. To address this issue, several machine learning (ML)-based studies have been proposed, which predict the future loads of VNF in service function chaining (SFC) enabled networks. In this paper, we compare two prediction models: 1) basic long short term memory (LSTM) model that uses historical resource utilization of VNF to predict and 2) context and aspect embedded attentive target dependent LSTM (CAT-LSTM) model that leverages both historical utilization of VNF and its neighbor VNFs. Simulation results demonstrate that performance of basic LSTM is not affected by the number of service chain (SC) allocated in VNF. Meanwhile, the performance of CAT-LSTM is apparently degraded when multiple SCs are allocated in VNFs, which means prediction loss of CAT-LSTM increases from 4% to 6%.

关键词： virtualized network function service function chaining machine learning load prediction

来源：评论

学校读者我要写书评

暂无评论

VBaaS: VNF Benchmark-as-a-Service 4

VBaaS: VNF Benchmark-as-a-Service

引用

Fourth European Workshop on Software Defined networks - EWSDN 2015

作者： Rosa, Raphael Vicente Rothenberg, Christian Esteve Szabo, Robert Univ Campinas UNICAMP Sch Elect & Comp Engn FEEC Campinas SP Brazil Ericsson Res Budapest Hungary

ISBN: (纸本)9781509001804

When rolling out network function Virtualization (NFV) services, resource monitoring becomes a critical task subject to different cost-accuracy tradeoffs depending on whether continuous monitoring or more static infrastructure resource views are taken. In this context, we propose virtualized network functions (VNF) Benchmark-as-a-Service (VBaaS) to enable not only run-time resource evaluation but also test-before-deploy opportunities for VNFs and NFV Infrastructures. We describe the motivation behind VBaaS and its main value proposition for a number of use cases around the orchestration tasks of VNF Forwarding Graphs We present the main components of VBaaS along their system interactions and interfaces, discussing the main benefits of adopting VBaaS and open research issues. Addressing the identified challenges and finalizing our proof of concept VBaaS are our main ongoing work activities.

关键词： benchmark network function virtualization virtualized network function

来源：评论

学校读者我要写书评

暂无评论

VNF Service Chaining on SAVI SDI 1st

VNF Service Chaining on SAVI SDI

引用

1st EAI International Conference on Future Access Enablers of Ubiquitous and Intelligent Infrastructures (FABULOUS)

作者： Yasrebi, Pouya Bemby, Spandan Bannazadeh, Hadi Leon-Garcia, Alberto Univ Toronto St George Campus40 St George St Toronto ON M5S 2E4 Canada

ISBN: (纸本)9783319270722

Managing computational resources and networking elements over today's heterogeneous infrastructure has become very challenging. A need for virtualizing network functions has emerged to reduce infrastructure operating costs. In this paper we consider using software-defined infrastructure (SDI) resource management system (RMS) to achieve service chaining of virtualized network functions (VNFs). SDI allows for the integrated control and management of heterogenous resources. In an SDI environment, the end user has access to interfaces that allow programmatic management of the resources. The user can define their own service graph (SG), which determines the path that traffic must take through various VNFs. The ability to dynamically realize the SG is what is referred to by service chaining. Use cases of service chaining include adding a firewall in front of web server and multicasting. Furthermore, we tested the firewall use case in two scenarios to verify validity of our service chaining implementation.

关键词： Software defined infrastructure network function virtualization virtualized network function Software defined networking OpenFlow Smart application on virtual infrastructure

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-based Optimal VNF Deployment 21

Machine Learning-based Optimal VNF Deployment

引用

21st Asia-Pacific network Operations and Management Symposium (APNOMS)

作者： Park, Suhyun Kim, Hee-Gon Hong, Jibum Lange, Stanislav Yoo, Jae-Hyoung Hong, James Won-Ki Pohang Univ Sci & Technol Comp Sci & Engn Pohang South Korea Norwegian Univ Sci & Technol Informat Secur & Commun Technol Trondheim Norway

ISBN: (纸本)9788995004388

network function Virtualization (NFV) environment can deal with dynamic changes in traffic status with appropriate deployment and scaling of virtualized network function (VNF). However, determining and applying the optimal VNF deployment in consideration of the cost and Quality of Service (QoS) is a complicated and difficult task. In particular, it is necessary to predict the situation at a future point when the deployment decision is applied because it takes processing time to apply the deployment decision to the actual NFV environment. In this paper, we randomly generate service requests in Multi-access Edge Computing (MEC) topology, then obtain optimal VNF deployment and Service function Chaining (SFC) result from an Integer Linear Programming (ILP) solution. We use the simulation data to train a machine learning model which predicts the optimal VNF deployment at a predefined future point. The prediction model shows the accuracy over 90% compared to the ILP solution for the 5-minute future time point.

关键词： network function Virtualization virtualized network function VNF Deployment Machine Learning

来源：评论

学校读者我要写书评

暂无评论

Provably Efficient Algorithms for VNF Routing Optimization 24

Provably Efficient Algorithms for VNF Routing Optimization

引用

24th IEEE International Conference on Parallel and Distributed Systems (ICPADS)

作者： Zhang, Shike Liu, Yuxiang Gao, Xiaofeng Zheng, Jiaqi Chen, Guihai Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai Key Lab Scalable Comp & Syst Shanghai Peoples R China Nanjing Univ State Key Lab Novel Software Technol Nanjing Jiangsu Peoples R China

ISBN: (纸本)9781538673089

Previous researches on (virtualized network function)VNF deployment mainly focus on resource and VNF distribution given several candidate paths, whereas routing, though more practical, is not under consideration. Especially, resource distribution does not remain optimal given continuously coming flows with different characteristics. No quantitative approach is proposed to determine the timing to redistribute resources and functions while real-time adjusting network distribution is too expensive. In this paper, we claim an alternative approach to handle the resource-and-flow matching problem using the method of routing. We find a routing and evaluate its costs using our VNFs Routing Evaluation Model. We prove the NP-hardness of VNF-RE. While previous works focus on heuristic algorithms, whose performance cannot be proven, we propose a two phase approach and prove the approximation ratio. The problem is formulated as an optimization problem which jointly considers VNF covering, multi-resource consumption, and routing overhead. We prove that the problem is NP-hard and our solution O((1 (kD (f max)/D- 1)beta)) log(3) N log m) approximates the optimal where k, D-f (max) and beta are variables that depend on the input flow. The simulation compares our work with several representative heuristic algorithms and shows that VNF-RE has better performance.

关键词： network function Virtualization virtualized network function routing Group Steiner Tree

来源：评论

学校读者我要写书评

暂无评论

Data access performance impact of a memory-centric radio network

Data access performance impact of a memory-centric radio net...

引用

作者： Tero Lindfors Aalto University

学位级别：硕士

Future 5G-based mobile networks will be largely defined by virtualized network functions (VNF). The related computing is being moved to cloud where a set of servers is provided to run all the software components of the VNFs. Such software component can be run on any server in the mobile network cloud infrastructure. The servers conventionally communicate via TCP/IP -network. To realize planned low-latency use cases in 5G, some servers are placed to data centers near the end users (edge clouds). Many of these use cases involve data accesses from one VNF to another, or to other network elements. The accesses are desired to take as little time as possible to stay within the stringent latency requirements of the new use cases. As a possible approach for reaching this, a novel memory-centric platform was studied. The main ideas of the memory-centric platform are to collapse the hierarchy between volatile and persistent memory by utilizing non-volatile memory (NVM) and use memory-semantic communication between computer components. In this work, a surrogate memory-centric platform was set up as a storage for VNFs and the latency of data accesses from VNF application was measured in different experiments. Measurements against a current platform showed that memory-centric platform was significantly faster to access than the current, TCP/IP using platform. Measurements for accessing RAM with different memory bandwidths within the memory-centric platform showed that the order of latency was roughly independent of the available memory bandwidth. These results mean that memory-centric platform is a promising alternative to be used as a storage system for edge clouds. However, more research is needed to study how other service qualities, such as low latency variation, are fulfilled in memory-centric platform in a VNF environment.

关键词： memory-centric virtualized network function Gen-Z memory-semantic SHM storage back-end access latency

来源：评论

学校读者我要写书评

暂无评论

An Efficient Adaptive Meta Learning Model Based VNFs Affinity for Resource Prediction Optimization in virtualized networks

引用

JOURNAL OF network AND SYSTEMS MANAGEMENT 2023年第2期31卷 40-40页

作者： Bellili, Asma Kara, Nadjia Ecole Technol Super Ets Software & Informat Technol Engn Dept 1100 Rue Notre Dame O Montreal PQ H3C 1K3 Canada

In today's cloud environments, diverse and massive data is growing exponentially. This requires efficient prediction mechanisms to support automatic resource allocation and management. These mechanisms enable anticipating resource consumption variation in complex and distributed systems. To allow the service provider fulfilling the service level objective (SLO), these mechanisms must deal with the heterogeneity and the non-stability of different resource demands. To enable the prediction mechanism adapting to different resource demands, we propose a selector model that chooses among several prediction methods, the best suiTable one to trigger at a given workload situation. The proposed model is a multitask selector based on meta learning strategy MT-MLS. The MT-MLS introduces a novel concept by analyzing similarities of multidimensional resource consumption between virtual network functions (VNFs) of a service function chain (SFC). Attention mechanism is used to identify a weight for each VNF based on the similarity analysis. The MT-MLS is designed as a multitask classifier, that assigns simultaneously and independently for each VNF the corresponding best predictor label to forecast multidimensional resource needs. We developed four deep learning base predictors: a CNN model, a GNN model, an LSTM model, and a hybrid model. These base predictors were used to generate meta data to train the MT-MLS. Then compared the prediction results of each model to the prediction results using the MT-MLS. Various real world SFC resource consumption datasets have been used to evaluate the proposed MT-MLS using three performance metrics for the evaluation of classification and five others for the evaluation of prediction. The performance analysis reveals that the MT-MLS enhance considerably the efficiency of the resource usage prediction.

关键词： virtualized network function Deep learning VNF dependencies Service function chain Cloud computing Multitask classification Meta learning VNF affinity

来源：评论

学校读者我要写书评

暂无评论

Hardware Acceleration Resource Allocation Mechanism for VNF

引用

Procedia Computer Science 2018年 131卷 746-755页

作者： Hongwei Fan Yuxiang Hu Shaojun Zhang Quan Ren National Digital Switching System Engineering & Technological R&D Center Kim Street No. 7 Zhengzhou 450000 China

Improving the performance for VNF (virtualized network function) by hardware acceleration is a hot research topic in SDN/NFV architecture. After introducing the acceleration resource for VNF, how to implement uniform deployment and allocation of the acceleration resources is an urgent problem to be solved. Thus in this paper, a unified deployment architecture of acceleration resource for VNF is proposed firstly. On the basis of architecture, the problem of acceleration resource allocation is modeled. By analyzing the influence of acceleration resource for VNF on deployment of service chains, we propose the evaluation index of acceleration resource allocation algorithm for VNF. Finally, we design a two-stage acceleration resource allocation algorithm to solve the problem. The experimental results show that the proposed algorithm can rationally allocate acceleration resources to nodes in the entire network and improve the efficiency of acceleration resource utilization.

关键词： SDN/NFV virtualized network function Hardware Acceleration Resource Allocation

来源：评论

学校读者我要写书评

暂无评论

Joint IT and Spectrum Resource Load Balancing for VNF Selection in Inter-Datacenter Elastic Optical networks 16

Joint IT and Spectrum Resource Load Balancing for VNF Select...

引用

16th International Conference on Optical Communications and networks (ICOCN)

作者： Li, Bo Yu, Xiaosong Zhao, Yongli Li, Yajie Ou, Qinghai Liu, Zhu Liao, Xiao Zhang, Jie BUPT Beijing 100876 Peoples R China KTH Royal Inst Technol Kista Sweden State Grid Informat & Telecommun Grp Co LTD Beijing Peoples R China

ISBN: (纸本)9781538632734

The selection of virtualized network functions (VNFs) in service provisioning is a critical issue in inter-datacenter elastic optical networks. In this paper, considering load balancing of IT and spectrum resources, we propose two joint balancing factors, i.e., Global Balancing factor (GB factor) and Local Balancing factor (LB-factor) for VNFs selection. Then a Joint-Optimization Selection (JOS) algorithm is designed based on the joint balancing factors. Simulation results show that GB-factor based JOS can achieve better performance compared with LB-factor based JOS and benchmark algorithm.

关键词： Inter-datacenter elastic optical networks joint balancing factor virtualized network function

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：