Manos Dramitinos

GridEcon: A Market Place for Computing Resources

This paper discusses the rationales for a Grid market and, in particular, the introduction of a market place for trading commoditized computing resources. The market place proposed makes computing resources from different providers substitutable through virtualization. This includes the definition of a spot and future market as well as the parameters that a market mechanism for computing resources should consider. The above market place is complemented by a set of value-added services (e.g. insurance against resource failures, capacity planning, resource quality assurance, stable price offering) that ensure quality for Grid users over time. The market place technology for all of the above services has been designed by the GridEcon project, contributing to a broader adoption of Grid technology and enabling a service-oriented knowledge utility environment.

An auction mechanism for bandwidth allocation over paths

Abstract The demand for bandwidth contracts over networks has recently been growing rapidly. Moreover, due to increased competition among providers, customized short-term bandwidth contracts are now preferred by users to static, long-term contracts of the past. To this end, auctions appear to be a proper trading mechanism. When purchasing bandwidth over a path of a network, it is only meaningful for a user to reserve the same quantity at all the constituent links. We have developed a simple, yet efficient auction for allocating bandwidth on a network basis to users who wish to utilize it for the same time period. This mechanism (referred to as MIDAS) consists of a set of simultaneous multi-unit Dutch (i.e. descending-price) auctions, one per link of the network. In order to win bandwidth over a certain path, it suffices for a user to simultaneously bid for the quantity desired at all relevant auctions. Thus, instant allocation of bandwidth is attained. An important feature of our approach is that prices at the various links are reduced at different rates, so that prices reflect the demand exhibited so far for each link. We have evaluated experimentally two price reduction policies, in terms of the social welfare associated with the resulting allocation, and argue (both theoretically and by means of experimental results) that it is indeed efficient to introduce such rules rather than reduce all prices at the same rate. We have also briefly addressed the issue of incentive compatible pricing.

An auction mechanism for allocating the bandwidth of networks to their users

We present a mechanism for auctioning bandwidth on a network-wide basis to end users or ISPs that will utilize it for the same time period. This mechanism consists of a set of simultaneous multi-unit descending-price (i.e. Dutch) auctions, one per link of the network. The per unit prices of bandwidth at the various links are asymmetric, thus reflecting the asymmetry of demand for these links. A user can be instantly allocated bandwidth over a certain path, by simultaneously bidding for the quantity desired at all relevant auctions. This winner determination rule is complemented by a payment rule of the VCG (Vickrey-Clarke-Groves) type, which provides users with the incentive to bid truthfully, thus simplifying bidding. Also, the mechanism enables the auctioneer to use his prior information on market demand anticipated and its spreading among the various links in order to set effectively the auctions parameters. We argue that our mechanism attains nearly efficient allocation of the networks bandwidth (i.e. the resulting social welfare is close to the respective maximum for the quantity decided to be sold by the auctioneer), while it is simple, scalable and applicable to real networks, even for auctioning the capacity of links owned by multiple providers and then splitting the revenue among them. Alternatively, the mechanism offers the provider the opportunity to optimize his revenue, rather than the social welfare. Since our mechanisms computational complexity is low it can serve as a fast, practical, and near-optimal solution to a generally NP-hard optimization problem.

Auction-based resource reservation in 2.5/3G networks

We consider UMTS networks in which users request services other than telephony that last for long time intervals: e.g., video clips that last for several minutes. The duration of network time-slots over which resource units are allocated is much shorter. This complicates consistent reservation of resources over longer time scales, where consistent reservation is required to ensure that service quality is constant throughout the entire service session. In this paper, we define an auction-based mechanism for nearly consistent reservation of the resources of a UMTS (or GPRS) network by the users that value them the most, in order to satisfy the longer time scale requirements of their service sessions. Each of these sessions has a fixed target bit-rate. The mechanism is based on a series of Generalized Vickrey Auctions and a set of predefined user utility functions that we propose. Bidding is performed automatically on behalf of the users on the basis of each users selection of one of these utility functions and his declaration of a total willingness to pay. We argue that under our mechanism the user does not have a clear incentive of not performing a truthful selection of a bidding function according to his own utility. The utility functions we define express appropriately the preferences of the users with respect to the resource allocation pattern in the cases where perfectly consistent allocation cannot be attained. We also provide a mapping of these functions to the UMTS service classes. The effectiveness of our resource reservation mechanism is demonstrated by means of experiments. It appears that most of the users either are served very satisfactorily or essentially are not served at all. The mechanism is implemented at the network base station, and is applicable in practical cases of networks with large numbers of users whose sessions last for many slots.

Challenges to support edge-as-a-service

A new era in telecommunications is emerging. Virtualized networking functions and resources will offer network operators a way to shift the balance of expenditure from capital to operational, opening up networks to new and innovative services. This article introduces the concept of edge as a service (EaaS), a means of harnessing the flexibility of virtualized network functions and resources to enable network operators to break the tightly coupled relationship they have with their infrastructure and enable more effective ways of generating revenue. To achieve this vision, we envisage a virtualized service access interface that can be used to programmatically alter access network functions and resources available to service providers in an elastic fashion. EaaS has many technically and economically difficult challenges that must be addressed before it can become a reality; the main challenges are summarized in this article.

Auction-based resource allocation in UMTS high speed downlink packet access (HSDPA)

Data, audio and video services over 3G networks possess certain requirements in QoS and thus in resources. Accommodating such service requests constitutes a challenge for UMTS networks. To this end, 3GPP Release 5 introduces the high speed downlink packet access (HSDPA). In this paper, we propose: i) an auction-based HSDPA resource allocation mechanism, which consists of a series of Generalized Vickrey Auctions that are conducted in the short time-scale over which the network resources are allocated and ii) a set of innovative user utility functions. These functions provide a quantification of both the users willingness to pay for reserving resources in the short time-scale auctions and his attained utility from the pattern of resources allocated during his long time-scale service session. The utility functions are additive and are used as bidding functions in the series of auctions the user participates. Each user selects one of these functions, which is scaled by his total willingness to pay. Then, the network runs the auctions by bidding on behalf of each user. The effectiveness of our mechanism has been assessed experimentally and it appears that most of the users either are served very satisfactorily or are allocated very limited quantities of resources (if at all) at a low total charge.

To subscribe, or not to subscribe: Modeling and analysis of service paradigms in cellular markets

Traditionally customers subscribe to specific providers and are served by accessing base stations (BSs) of the network of their provider. Inevitably subscribers with relatively “high” usage pattern and data-rate requirements are subsidized by the ones with lower usage and data-rates. As the wireless technology advances, a diverse set of services will be available. This paper introduces the “flex service” paradigm that allows a customer to dynamically access BSs of different providers based on various criteria, such as profile, network conditions, and offered prices. “Flex users” can select the appropriate provider and BS on a per-session basis. This work considers a diverse customer population with respect to their demand, their preference on data-rate over price, their tolerance on the blocking probabilities of their sessions, and their willingness to pay for certain services. Users can dynamically decide to buy a long-term subscription or become flex users. In this paper, we develop a rich framework for modeling and analysis of such markets in different spatio-temporal scales. We analyze the evolution of markets with the flex service paradigm, focusing on whether it can improve the quality-of-service (QoS), social welfare, flexibility and further enhance the competition among providers. The main contribution of this paper is detailed modeling and indepth performance analysis of such complex markets, in different spatial and temporal scales. It considers the perspective of clients, providers, and regulators. It demonstrates the benefits of markets with the flex service paradigm and compares them with the ones that only offer subscription contracts.

Market Mechanisms for Trading Grid Resources

There has been recently an increasing interest in Grid services and economic-aware Grid systems both in the industry and the academia. In this paper we specify a market for hardware providers and consumers interested in leasing Grid resources for a time period. Our approach comprises a stock-market like mechanism that enables the trading of computational power on the basis of a spot and a futures market. The spot market comprises a pair of bid and ask queues. This grid market is more complicated than the standard spot/futures markets of storable commodities, because the computational service traded in our case comprises of resources that are perishable, and has both quantity and duration specified in terms of a time interval. This is an important feature of our market mechanism, complicating considerably the trading algorithms that we develop and assess in this paper.

Manufactured by Software: SDN-Enabled Multi-Operator Composite Services with the 5G Exchange

Foreseen 5G verticals hold the promise of being true value-added services, hence bringing significant income to their respective providers. However, the nature of these verticals are very demanding in terms of both economic and technical requirements, such as multi-operator cooperation, end-to-end quality assurance, and the unified orchestration of network and cloud resources. Existing systems fall short of satisfying these requirements, but emerging network softwarization and resource virtualization technologies, such as SDN and NFV, show promise for being key enablers in this context. In this article, we introduce the 5G Exchange (5GEx) concept that builds on SDN and NFV, and facilitates the provisioning of multi-operator 5G services by means of inter-operator management and orchestration of virtualized network, compute, and storage resources. We present potential 5GEx use cases, conceptual architecture, and value proposition. We also outline open research questions on how to exchange information in such a coopetitive environment, and provide an outlook on the impact of 5GEx on a network service providers business and operation.

Inter-carrier interconnection services: QoS, economics and business issues

The Internet has evolved towards a unique technology base for creating value-added services with worldwide connectivity. These services, enabled by the increasing bandwidth of access networks, result in new high-performance applications (e.g. e-health, high definition video streaming, network gaming etc.) To support and materialize the value of these emerging services, it is important for the operators to be able to provide some form of Quality of Service (QoS) assurance. This paper presents the main economic issues regarding the efficient provisioning of such services in inter-domain level and overviews some candidate economic mechanisms and game-theoretic tools that could be adopted.