Learning-Based Reservation of Virtualized Network Resources

نویسندگان

چکیده

Network slicing markets have the potential to increase significantly utilization of virtualized network resources and facilitate low-cost deployment over-the-top services. However, their success is conditioned on service providers (SPs) being able bid effectively for resources. In this paper, we consider a hybrid advance-reservation spot slice market study how SPs should reserve maximize services’ performance while not violating time-average budget threshold. We problem in its general form where SP demand prices are time-varying revealed only after reservations decided. develop learning-based framework, using theory online convex optimization, that allows employ no-regret reservation policy, i.e., achieve same with an oracle has full access all future prices. extend framework scenario decides dynamically orchestration hence needs learn performance-maximizing resource composition; further mixed-time scale scheme leverage spot-market information between successive reservations. The proposed learning evaluated representative simulation scenarios highlight efficacy as well impact key system algorithm parameters.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Advance Reservation of Network Resources for Multimedia Applications

To guarantee Quality of Services for continuous media like audio and video, intermediate nodes in networks must be able to reserve resources like buffer capacity, CPU time and bandwidth. Several methods have been introduced to verify that new connections can be accepted by the network without violating guarantees made for already established connections. One drawback of all proposed admission c...

متن کامل

Response Time-based Self-tuning Fuzzy Control of Virtualized Resources

Cloud elasticity allows dynamic resource provisioning in concert with actual application demands. Feedback control approaches have been applied with success to resource allocation in physical servers. However, cloud dynamics make the design of an accurate and stable resource controller more challenging, especially when response time is considered as the measured output. Response time is highly ...

متن کامل

Truthful Market-based Trading of Cloud Resources with Reservation Price

With the rapidly growing demand for the cloud services, a need for efficient methods to trade computing resources increases. Commonly used fixed-price model is not always the best approach for trading cloud resources, because of its inflexible and static nature. Dynamic trading systems, which make use of market mechanisms, show promise for more efficient resource allocation and pricing in the c...

متن کامل

Advance Reservation of Resources in Workflow System

This is related to reserve the resource in advance in “Workflow scheduling” of cloud computing. The problem with the traditional allocation of resource is that the resource may be not available when needed and the requested application will be rejected. To overcome this problem, an advance reservation technique is used. Various techniques are used to reserve the resource in advance such as ECA ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Transactions on Network and Service Management

سال: 2022

ISSN: ['2373-7379', '1932-4537']

DOI: https://doi.org/10.1109/tnsm.2022.3144774