J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/band-price.html  Home/Search   Document Details and Download   Summary   Related Articles   Check

....shown in subsequent analysis, careful selection of these constants leads to stable network behavior. There are several characteristics inherent to the polynomial pricing function which make it preferable to other schemes explored in the literature, such as the barrier pricing function examined in [19] and provided below: P (n(t) a n(t) C # b (5) These advantages of a polynomial pricing function include: The price stamped by a barrier function asymptotically approaches infinity as the bu#er approaches its full capacity. Since it is possible for a burst of tra#c to arrive ....

J. Murphy and L. Murphy. Bandwidth Allocation by Pricing in ATM Networks. In IFIP Transactions C-24: Broadband Communications II, pages 333--351, Paris, France, March 1994.

....Implementation varies based on the underlying technology. Proposed schemes for ATM networks include use of pricing at connection admission to encourage users to correctly declare QOS parameters [Kelly97b] and use of prices to cause users to exert control over their cell transmission rate [Murphy94] In the Internet, schemes have been proposed that would allow users to place the price that they would be willing to pay for a packet s transmission into the packet s header [MacKie Mason95] The network would then make a decision as to whether the packet should be transmitted or dropped, and ....

J. Murphy and L. Murphy. Bandwidth Allocation By Pricing In ATM Networks. In Second International IFIP Conference on Broadband Communications, BB94,

....are interdependent, it is important to address these issues simultaneously. However, previous microeconomicbased research has only investigated these issues in isolation. It has been demonstrated that pricing is an e#ective method for achieving fair allocations as well as revenue generation [1, 4, 6, 10, 14, 18, 23]. However, these methods do not consider how to provision resources. Other work has investigated resource provisioning [3, 5, 12, 11, 19] but not resource allocation to individual users. In [8] a hierarchical model was introduced to provision and allocate 1 Di#Serv bandwidth. Bandwidth was ....

J. Murphy and L. Murphy. Bandwidth Allocation by Pricing in ATM Networks. In ITC, June 1995.

....by consumers, and seek to maximize their satisfaction by selling or renting their resources. Using this framework, microeconomics can be used to define how network resources are allocated. One approach of applying microeconomics to computer networks involves a maximization of utility functions [9, 10, 11, 12, 14, 17]. A utility function maps a resource amount to a satisfaction value. Using this function, one can compare the satisfaction levels of di#erent resource amounts. The maximization process determines the optimal resource allocation such that the utility of a group of users is maximized subject to ....

J. Murphy and L. Murphy. Bandwidth Allocation by Pricing in ATM Networks. In ITC, June 1995.

....and processor time) and maximize their utility by renting or selling. Using this framework, microeconomics can 2 be used to define how network resources are allocated. One approach of applying microeconomics to computer networks involves a maximization of utility functions [8] 9] 13] 14] [17] [21] A utility function maps a resource amount to a satisfaction value. Using this function, one can compare the satisfaction levels of different resource amounts. The maximization process determines the optimal resource allocation such that utility is maximized subject to budget and resource ....

J. Murphy and L. Murphy. Bandwidth Allocation by Pricing in ATM Networks. In ITC, June 1995.

....determine the order in which connections are established. Since CAC requires a number of parameters the addition of another parameter will not produce a serious overhead. We do not propose that the bid should replace the CAC process but rather be used as another parameter in this process. See [5] for related work on allocating bandwidth according to a pricing scheme. An important aspect is what happens to existing connections when the network gets congested. It is important to remember that once a connection has been set up it can t be broken down again. CBR and VBR are guaranteed ....

J. Murphy and L. Murphy, Bandwidth Allocation by Pricing in ATM Networks, , Technical report, EECS Department, University of California, Berkeley, 1994

....seen in the growth of Spam email or more subtly in the competition among browsers which can open multiple TCP connections to increase performance. Thus, it is of increasing importance to understand incentive issues on the Internet and there has been a growing body of work on this topic, e.g. [31, 37, 5, 41, 42, 24, 25, 39, 38, 4, 3, 30, 29, 36, 32]. To address these issues, many theorists have begun to model the Internet as a noncooperative game (see, e.g. 7, 6, 17, 19, 26, 25] Most of these analyses assume that the outcome of a game is a Nash equilibrium, the standard outcome analyzed in economics [16] Given the decentralized and ....

John Murphy and Liam Murphy. Bandwidth allocation by pricing in ATM networks. Technical report, Dublin City University, Ireland, July 1994.

....equals the total amount available. Most of the results on decentralized resource allocation currently available in the literature are based on the price directed approach #Cocchi et al. 1993; MacKie Mason and Varian, 1995b; Jordan and Jiang, 1995; Jiang and Jordan, 1995; Wang et al. 1997; Murphy and Murphy, 1994; Murphy et al. 1994; Parris et al. 1992; Parris and Ferrari, 1992; Kelly, 1994; Kelly et al. 1998; Gupta et al. 1997; Courcoubetis et al. 1997; MacKie Mason and Varian, 1995a; Low and Varaiya, 1993; de Veciana and Baldick, 1998; Thomas and Teneketzis, 1997#. Since in this paper we follow the ....

....available. Most of the results on decentralized resource allocation currently available in the literature are based on the price directed approach #Cocchi et al. 1993; MacKie Mason and Varian, 1995b; Jordan and Jiang, 1995; Jiang and Jordan, 1995; Wang et al. 1997; Murphy and Murphy, 1994; Murphy et al. 1994; Parris et al. 1992; Parris and Ferrari, 1992; Kelly, 1994; Kelly et al. 1998; Gupta et al. 1997; Courcoubetis et al. 1997; MacKie Mason and Varian, 1995a; Low and Varaiya, 1993; de Veciana and Baldick, 1998; Thomas and Teneketzis, 1997#. Since in this paper we follow the price directed ....

[Article contains additional citation context not shown here]

Murphy, L. and Murphy, J. #1994#. Bandwidth allocation by pricing in ATM Networks. In Proc. of the IFIP Broadband Communications.

....the extent to which they can adjust prices to fit the particular applications being used. See [28] for a discussion of the implications of this layering for content provision. There have been many pricing proposals in the recent literature and we do not attempt to review them all here; see [2, 3, 14, 25, 24, 23, 21, 27, 32, 34, 35, 37, 36, 42, 18, 45] for a few representative samples. The most ambitious pricing proposal for best effort traffic is the smart market proposal of MacKie Mason and Varian described in [29] see also ( 27, 26] In this scheme, each packet carries a bid in the packet header; packets are given service at each ....

John Murphy and Liam Murphy. Bandwidth allocation by pricing in ATM networks. In Proc. of IFIP Broadband Communications '94, Paris, France, March 1994. Available from URL: ftp: //gopher. econ. lsa. umich. edu/pub/Archive/Murphy Bandwidth Pricing.ps.Z.

....seek to maximize their satisfaction by selling or renting their resources. Using this framework, microeconomics can be used to define how network resources are allocated. One approach of applying microeconomics to computer networks involves a maximization of utility functions [9] 10] 11] 12] [14] [17] A utility function maps a resource amount to a satisfaction value. Using this function, one can compare the satisfaction levels of different resource amounts. The maximization process determines the optimal resource allocation such that the utility of a group of users is maximized subject ....

J. Murphy and L. Murphy. Bandwidth Allocation by Pricing in ATM Networks. In ITC, June 1995.

....that as congestion grows, the price that users are willing to pay for higher priority services also grows. While these results are purely theoretical, they do confirm our intuition about pricing and they provide a basic analytical and simulation framework for studying pricing and DS issues. In [19], a closed loop pricing scheme is developed in which a per packet price is determined by the buffer length at a local gateway the longer the buffer, the higher the price. The price is fixed for a predetermined period, after which the buffer is sampled again and a new price is assigned, if ....

J. Murphy and L. Murphy, "Bandwidth Allocation by Pricing in ATM Networks," IFIP Transactions C: Communication Systems, Vol. C-24, 1994.

....allocated. In this paper, we apply microeconomics theory only to the task of flow control (i.e. we do not suggest its use for revenue generation or usage based billing) One approach of applying microeconomics to computer networks uses maximization techniques to maximize utility [4] 5] 6] 7] [8]. A utility function maps a resource amount to a satisfaction value. Using this function, one can compare the satisfaction levels of different resource amounts. The maximization process determines the optimal resource allocation such that the utility of a group of users is maximized subject to ....

J. Murphy and L. Murphy, "Bandwidth allocation by pricing in atm networks," in ITC, June 1995.

....networks presents additional challenges. Since these networks rely on statistical multiplexing, the reservation of resources is ideally kept to a minimum, and measuring usage is more difficult and costly. The current debates on how to charge for Internet services [1, 8, 10] and ATM services [11, 12, 15] have sparked renewed interest in the field of network pricing. We contend that a simple but relatively general pricing policy would assign a price P to service type j according to three main factors: resource allocation, usage and fixed connection costs. This relationship can be expressed as: P j ....

John Murphy and Liam Murphy. Bandwidth Allocation by Pricing in ATM Networks. Technical report, Dublin City University, Ireland, July 1994.

....addresses the issue of demand elasticity (since users can choose to postpone their call based on the price being currently charged) and preliminary results show positive impacts on the control of congestion. Different implementations of pricing as feedback have been suggested in [12] and [14]. Responsive pricing possesses many desirable characteristics from a theoretical point of view; however, its implementation seems unlikely due to significant complexity. A compromise would be time of day pricing, where rates are higher at times when the network is expected to be more congested. At ....

....can lead users to make choices that optimize overall network performance. Only priority based services were considered, with no resource reservation allowed. It is reasonable that in a network where resource reservation is allowed, it will be a major factor in pricing for a service. The work in [14] and [18] recognize the importance of charging for bandwidth allocation. We propose in [4] a pricing scheme with three components: a fixed connection charge according to the service; an allocation sensitive charge; and a usage sensitive charge. We focus on the question of how to design a pricing ....

[Article contains additional citation context not shown here]

John Murphy and Liam Murphy. Bandwidth Allocation by Pricing in ATM Networks. Available from http: //www.sims.berkeley.edu/ resources/ infoecon/ Pricing.html.

....the extent to which they can adjust prices to fit the particular applications being used. See [28] for a discussion of the implications of this layering for content provision. There have been many pricing proposals in the recent literature and we do not attempt to review them all here; see [3, 4, 15, 25, 24, 23, 22, 27, 32, 34, 35, 37, 36, 42, 19, 44] for a few representative samples. The most ambitious pricing proposal for best effort traffic is the smart market proposal of MacKie Mason and Varian described in [29] see also ( 27, 26] In this scheme, each packet carries a bid in the packet header; packets are given service at each ....

John Murphy and Liam Murphy. Bandwidth allocation by pricing in ATM networks. In Proc. of IFIP Broadband Communications '94, Paris, France, March 1994. Available from URL: ftp: //gopher. econ. lsa. umich. edu/pub/Archive/Murphy Bandwidth Pricing.ps.Z.

....and Tribble (1993) describe a low cost, bilateral transfer arrangement for routing packets based on trust and repeated interaction. Cocchi, Estrin, Shenker, and Zhang (1992) provide motivation for pricing, describe a general framework for dealing with pricing issues and conduct some simulations. Murphy and Murphy (1994) describe some simulations with pricing ATM traffic. What about pricing multiple qualities of service Most studies to date have considered pricing for a single service dimension (priority in a firstcome, first served network, or call admission to an ATM network) However, with the growing ....

Murphy, J., and Murphy, L. (1994). Bandwidth allocation by pricing in ATM networks. Tech. rep., EECS Department, University of California, Berkeley.

No context found.

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/band-price.html

No context found.

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', Proc. of IFIP Broadband Communications '94,Paris, France, March 1994.

No context found.

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/band-price.html

No context found.

John Murphy and Liam Murphy, "Bandwidth Allocation By Pricing In ATM Networks", to be presented at Second International IFIP Conference on Broadband Communications, BB-94, Paris, France, March 1994.

No context found.

John Murphy and Liam Murphy, "Bandwidth Allocation By Pricing In ATM Networks", to be presented at Second International IFIP Conference on Broadband Communications, BB-94, Paris, France, March 1994.

....adjusts the prices based on current network conditions. It turns out that it is not necessary for the network operator to know the user benefit functions# therefore this pricing scheme is suitable for both public and private networks. A distributed iterative pricing algorithm has been developed [3] ( see Figure 3 ) The distributed nature of the pricing algorithm suggests that it may be possible to meet the real time feedback requirement. In addition, the computation required per iteration at each user and ATM access switch is simple, which suggests that inexpensive processing elements may ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', Proc. of IFIP Broadband Communications '94,Paris, France, March 1994.

....deviation is similarly chosen. 6. DISCUSSION AND CONCLUSIONS The models presented in this paper adhere to the H.261 video coding standard and have been implemented by a discrete eventsimulator. This will allow the sources to be used on manyATM network issues such as congestion control strategies [6]. The accuracy of the models in representing the real codecs is shown for the one layer codec by comparing the moments of the bit rates and these are seen to be compatible. The accuracy of the models is only as good as the accuracy of the codec algorithms that were implemented and the length of ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', Broadband Communications '94, IFIP TC-6, Telecom Paris,Paris, France, 2-4 March1994.

....within the tolerance for a reasonable call admission algorithm. However, sometimes the amountofactive best efforts traffic will be large, and the network will suffer some performance difficulties (packet delays and losses) 4. 2 Closed loop feedback Our first feedback network uses a simple scheme [9], 12] The network state is measured by buffer occupancy at the gateway. This occupancy is converted intoapriceperpacket, which is then transmitted back to each active adaptive application. The applications then decide on howmany packets to send during the next interval, as described above. In ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/band-price.html

....is within the tolerance for a reasonable call admission algorithm. However, sometimes the amount of active elastic traffic will be large, and the network will suffer some performance difficulties (packet delays andlosses) 4.2. Closed loopfeedback Our first feedback network uses a simple scheme [Murphy and Murphy 1994], Murphy et al. 1994] The network state is measured by buffer occupancy at the gateway. This occupancy is converted into a price per packet, which is then transmitted back to each active adaptive application. The applications then decide on how many packets to send during the next interval, as ....

....a reasonable call admission algorithm. However, sometimes the amount of active elastic traffic will be large, and the network will suffer some performance difficulties (packet delays andlosses) 4.2. Closed loopfeedback Our first feedback network uses a simple scheme [Murphy and Murphy 1994] [Murphy et al. 1994]. The network state is measured by buffer occupancy at the gateway. This occupancy is converted into a price per packet, which is then transmitted back to each active adaptive application. The applications then decide on how many packets to send during the next interval, as describedabove. In this ....

J. Murphy and L. Murphy. 1994. Bandwidth Allocation By Pricing In ATM Networks. IFIP Transactions C: Communication Systems, No. C-24, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/bandprice. html.

....is within the tolerance for a reasonable call admission algorithm. However, sometimes the amount of active elastic traffic will be large, and the network will suffer some performance difficulties (packet delays and losses) 4. 2 Closed loop feedback Our first feedback network uses a simple scheme [9], 12] The network state is measured by buffer occupancy at the gateway. This occupancy is converted into a price per packet, which is then transmitted back to each active adaptive application. The applications then decide on how many packets to send during the next interval, as described above. ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/ murphyj/band-price/band-price.html

....such as bandwidth or buffer space. 4.1 Economic framework In this Section we outline one possible method for computing adaptive prices. The users discussed below are thus adaptive users ( either adaptive best effort, inelastic, or elastic ) A more detailed description can be found in [11] [12]. Other pricing schemes for communication networks have been suggested, e.g. 13] 14] The network and its users are considered to form an economy or economic system. The system has various resources such as link bandwidths and buffer spaces that can be used to meet user demands for service. ....

....adjusts the prices based on current network conditions. It turns out that it is not necessary for the network operator to know the user benefit functions; therefore this pricing scheme is suitable for both public and private networks. A distributed iterative pricing algorithm has been developed [12]. The distributed nature of the pricing algorithm suggests that it may be possible to meet the real time feedback requirement. In addition, the computation required per iteration at each user and ATM access switch is simple, which suggests that inexpensive processing elements may be sufficient in ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', Proc. of IFIP Broadband Communications '94, Paris, France, March 1994.

....a set of admitted user connections. The only room for interaction with the users in such a setting is through the connection setup negotiation. Therefore, as a baseline, we simulate a network that does not provide feedback. 4.1. 2 Closed loop feedback This feedback network uses a simple scheme [7], 10] The network state is measured by buffer occupancy at the network access points. This occupancy is converted into a price per cell, which is then transmitted back to each active adaptive application. The applications then decide on how many cells to send during the next period. In this ....

J. Murphy and L. Murphy, `Bandwidth Allocation By Pricing In ATM Networks', IFIP Transactions C : Communication Systems, No. C-24, 1994, p. 333-351. Available from URL http://www.eeng.dcu.ie/¸murphyj/bandprice /band-price.html

No context found.

J.Murphy and L.Murphy, "Bandwidth Allocation by Pricing in ATM Networks," Technical Report, Dublin City University, Ireland, 1994.

No context found.

J. Murphy and L. Murphy, "Bandwidth Allocation by Pricing in ATM Networks." Technical Report, Dublin City University, Ireland, July 1994.

No context found.

John Murphy and Liam Murphy. Bandwidth allocation by pricing in ATM networks. Technical report, Dublin City University, Ireland, July 1994.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC