Cloud service negotiation framework for real-time E-commerce application using game theory decision system

Abstract

A major demanding issue is developing a Service Level Agreement (SLA) based negotiation framework in the cloud. To provide personalized service access to consumers, a novel Automated Dynamic SLA Negotiation Framework (ADSLANF) is proposed using a dynamic SLA concept to negotiate on service terms and conditions. The existing frameworks exploit a direct negotiation mechanism where the provider and consumer can directly talk to each other, which may not be applicable in the future due to increasing demand on broker-based models. The proposed ADSLANF will take very less total negotiation time due to complicated negotiation mechanisms using a third-party broker agent. Also, a novel game theory decision system will suggest an optimal solution to the negotiating agent at the time of generating a proposal or counter proposal. This optimal suggestion will make the negotiating party aware of the optimal acceptance range of the proposal and avoid the negotiation break off by quickly reaching an agreement.

Keywords

Service level agreement broker-based negotiation framework game theory decision system E-commerce application

1 Introduction

Cloud Computing provides the real-time on-demand service provisioning over the internet. In order to keep mutual trust between Service Provider (SP) and Service Consumer (SC), an SLA signed between the involved parties, which describe the agreed quality of service deliverables and the penalty associated with violation of the agreement by both parties. Current cloud business model supports semi-personalized service access mechanism, which may not be feasible to adapt in the future due to increased demand of users and the dramatic shift of web-based application deployment on the cloud, which may lead to demand for personalized service access in the cloud, by making changes in the terms and conditions of cloud service SLA. It requires the change in current cloud service SLA model, which follows the static or dynamic SLA. Either the consumer can accept or reject the agreement, but there is no provision to modify the terms and conditions set by the provider. So in order to insist on the modifiable terms and conditions in the agreement, a dynamic SLA concept needs to initiate through the effective negotiation between the parties. Buyya et al. have addressed the issue of developing an application level negotiation framework in the scenario of SLA oriented resource allocation models [1]. Therefore, the development of negotiation framework addressed using various decision-making approaches like Adaptive Probabilistic Behavioral Learning [2], Probabilistic Decision Making [3], Bulk Negotiation Behavioral Learning [4], Similarity & Gale Shapely Stable Matching [5] and Adaptive Neuro-fuzzy Behavioral Learning [6]. None of these research works concentrated on the optimization of negotiation solutions in SLA oriented resource management [7]. Here, the optimization of the negotiation process using the decision support system is the major research gap identified during the research study. This scenario has motivated the research work towards the development of a novel negotiation framework using the game theory decision system approach.

The objective of this research work is to develop an efficient SLA negotiation framework for providing personalized service access to the SC which can take less Total Negotiation Time (TNT) during the negotiation process. Here the negotiation is to be made between the parties concerning the SLA parameters and because of this process, the relevant service provider processes consumer requests after establishing an agreement. This negotiation can broadly classify into two types based on the way of communication between the parties namely direct and indirect negotiation. Direct negotiation otherwise called a straightforward negotiation where the SC can talk or negotiate to the SP directly. In paper [8], a generic direct negotiation model with the one-to-one scenario is derived using a game theory approach, which is applicable for web, grid and cloud services. This model takes less TNT to communicate with one SP, where it does not require an indirect negotiation model to spend extra delay in the intermediate part.

In the case of a one-to-many scenario, a direct negotiation model takes more TNT, because the SC needs to negotiate sequentially with multiple numbers of SPs to identify the best provider in the market with less cost and time. Therefore, there is a need for a new negotiation model, which can handle multiple SPs negotiation with less TNT. So an indirect negotiation model was introduced to negotiate with multiple SPs where the Third-Party Broker (TB) used as an intermediate, to negotiate on behalf of SC. The TB contains an ‘n’ number of agents, which are capable of negotiating in parallel with ‘n’ number of SPs. Therefore, introducing an indirect negotiation model with a multi-agent broker will significantly reduce the TNT involved in the negotiation process. The existing negotiation frameworks without decision-making models do not guarantee that the negotiation party would result in the signing SLA. Hence a novel decision-making model is introduced in the third-party broker which exploits the Nash bargaining solution to suggest the optimal solution for SC before the final round of the negotiation process, which makes SC conscious about sending the final proposal with high utility for either party may result in the signing of SLA.

The key contribution of this research study includes as follows: (a) a novel Game Theory based Automated Dynamic SLA Negotiation Framework is introduced for initiating the cooperative negotiation among the participants; and (b) a game theory decision system is proposed to generate proposal or counter-proposal by the negotiation parties. The primary idea of this research work includes the exploitation of co-operative game theory approach to negotiation framework for the scenario of multiple service providers and consumers. This research work extends the negotiation model of Xianrong Zheng, et al., where the 1-to-m, m-to-1 and m-to-m model is addressed as the future work [8]. In order to justify the proposed approach mathematical models derived for both the direct and indirect negotiation framework. Here the corollary and its proof are defined for the negotiation process with a set of players PY = {SC, TB, SP} where SC, TB, and SP denotes the service consumer, broker and service providers respectively. So the optimal solutions suggested by the TB (with proposed decision-making model) will provide high value to both SC and SP.

2 Related works

2.1 Negotiation framework

To negotiate with a physical resource of different cloud infrastructure, a resource negotiator addressed as future work for integrating into a real cloud test bed [9]. Current negotiation models support only time and price-based negotiation without any decision-making model [10]. In order to support negotiation with advanced reservation, Current Price and Timeslot Negotiation (CPTN) is proposed using alternating offer protocol, decision-making model and trade-off algorithm to increase total utility and negotiation speed, and reduces computational load [11]. The decision-making algorithm and new policies are required for complex and advanced resource allocation in a time-critical application to overcome the existing best effort policy [12]. This multilateral negotiation framework in the cloud market consists of multiple one-to-many negotiations, which identify the best fit of user requirement [13]. For successful negotiation to take place between provider and consumer, a negotiation framework and protocol has to define and bargaining solutions are to be optimized [14].

2.2 Negotiation protocols

A conversation protocol between the parties can be either bilateral or multilateral. The bilateral protocol used in one-to-one negotiation model, and multilateral protocol used in case of one-to-many, many-to-one and many-to-many negotiation models [15]. None of them addressed the negotiation specific protocol, architecture and comparison of its requirement elaborately [16]. The framework developed using the Contract Net Protocol (CNP) makes effective web service negotiation between the insurance and repair companies. The CNP specifies the communication and control for problem-solving in the distributed nodes using resource allocation and effective selection of tasks respectively [17].

In the cloud, dynamic resource allocation uses automated negotiation, which extends the Alternate Offer Protocol (AOP) for supporting bilateral negotiation [18]. In the multi-agent system, an Alternate Offer Protocol extended for improving the efficiency of protocol concerning the percentage of failure, length of bargaining, computation cost, participation rate and fairness. As a test bed, agent-based simulation is used for protocol design and “Pareto Optimum” is suggested as a favorable solution for selecting payoff [19]. Logic-based alternate offer protocol introduced to handle this conflicting information with a suitable agreement where each agent knows its utility function [20]. In the agent-based electronic market, fuzzy multi-agent negotiation proposed with two types of agents where the agents employ time, resource and behavior dependent tactics [21].

2.3 Negotiation solutions

According to negotiation theory, approaches are classified into axiomatic, structural, strategic, procession, behavioral and integrative approaches [22]. The ultimate goal of SLA negotiation is to get a better payoff for both negotiating parties. In the context of game theory, bargaining solutions can be obtained through an axiomatic or strategic approach.

Nash bargaining solution (NBS): Concept of two-person bargaining problem was first introduced by John Nash to establish agreement from the set of possible outcomes obtained from the player’s co-operation [23]. To provide the unique solution to the above problem, Nash proposed the following axioms: Pareto optimality, Symmetry, Invariant concerning Affine Transformations of Utility, and Independence of Irrelevant Alternatives, which describes the behavior of each player [24].

Kalai-Smorodinsky bargaining solution (KSBS): Kalai and Smorodinsky proposed a new solution for bargaining problem that is entirely different from Nash bargaining solution with a limited set of axioms such as Pareto optimality, Symmetry and Invariant. To overcome the difficulty addressed in the paper [25], and to avoid the criticism of the fourth axiom an alternative axiom called Monotonicity is proposed in addition to the above three axioms. The extension of KSBS is proposed where the single-valued solution is identified in the large domain and two axiomatic characterizations such as crucial axiom and three weaker axioms including monotonicity are also presented on the subset of the domain [26].

Egalitarian bargaining solution (EBS): The egalitarian bargaining solution is a third solution introduced by Ehud Kalai for awarding equal gain to both the parties. This solution follows the axiom of Monotonicity and Independence of Irrelevant Alternatives by dropping the condition of scale invariance. An ‘n’ person egalitarian solution is the only bargaining solution presented in the existing research work [27], which is characterized using the symmetric decomposition. The axiomatic approach of bargaining should satisfy the list of properties (Pareto optimality) which requires the available set as comprehensive [28]. Finally, the Table 1 comparison will show the actual difference between the existing and proposed negotiation techniques.

Table 1
Comparison of techniques exploited in negotiation frameworks

Negotiation Model Negotiation Framework Negotiation Protocol Optimization Solution

Direct Negotiation Without Agent Trust Negotiation Contract Net NIL

Direct Negotiation With Agent Price and Timeslot Negotiation, Distributed Negotiation Alternate Offer NIL

Indirect Negotiation Without Agent Multilateral Negotiation Contract Net NIL

Indirect Negotiation With Agent Proposed Automated Dy-namic SLA Negotiation Alternate Offer Game Theory Decision

Negotiation Model	Negotiation Framework	Negotiation Protocol	Optimization Solution
Direct Negotiation Without Agent	Trust Negotiation	Contract Net	NIL
Direct Negotiation With Agent	Price and Timeslot Negotiation, Distributed Negotiation	Alternate Offer	NIL
Indirect Negotiation Without Agent	Multilateral Negotiation	Contract Net	NIL
Indirect Negotiation With Agent	Proposed Automated Dy-namic SLA Negotiation	Alternate Offer	Game Theory Decision

3 Automated dynamic SLA negotiation framework

The architecture of the proposed Automated Dynamic SLA Negotiation Framework illustrated as Fig. 1 for providing personalized service access in the cloud environment. This framework supports the indirect cloud service negotiation for the classification of one-to-one, one-to-many, many-to-one and many-to-many models. In order to reduce the delay on the negotiating parties, the negotiation process of ADSLANF automated by using the agents in the intermediate and negotiating end. The Service Consumer Agent (SCA) and Service Provider Agent (SPA) are incorporated in the negotiating end to negotiate on behalf of SC and SP respectively. Next to reduce the TNT of SCA with multiple SPAs; a trusted TB is introduced to negotiate on behalf of SCA, which consists of multiple autonomous agents called Third-party Broker Agent (TBA) to negotiate simultaneously with all the SPAs. This research work implements the one-to-many negotiation model with mathematical derivation for the selection of best Proposal (P) or Counter Proposal (CP), by using bargaining game theory approach in TB (decision-making model).

Fig. 1

Game Theory based Automated Dynamic SLA Negotiation Framework.

To follow standard procedure during the negotiation process, an SLA negotiation strategy is proposed using other offer protocols for making communication among the SC, TB and SP. During the negotiation process, TNT value may increase or decrease based on the Expected Communication Time (ECT) among the negotiating parties and occurrence of delay in the negotiating (SC and SP) and intermediate (TB) end. In the negotiating end, Proposal Generation Delay (PGD) and Counter Proposal Generation Delay (CPGD) delay occur due to the respective SC and SP. On the intermediate end, Third-party Broker Forwarding Delay (TBFD) occurs while forwarding the P and CP to SP and SC. To measure the performance of negotiation, TNT of direct and indirect negotiation methods can compute differently. In the direct negotiation, Total Negotiation Time (TNT) between the negotiation parties concerning one negotiation round can be computed as shown in Equation (1).

$TNT (SC, SPi) = \sum_{i = 1 to n} {PGD + CPGD + 2 * ECT (SP}_{i})$ (1)

In the case of indirect negotiation, Total Negotiation Time (TNT) among the negotiation parties (SC, TB and SPs) concerning one negotiation round can be computed as shown in Equation (2).

$\begin{matrix} {TNT (SC, TB, SP}_{i}) = PGD + CPGD + 2 * \sum_{i = 1 ton} TBFD \\ + ECT (SPi) \end{matrix}$ (2)

3.1 Representation of bargaining game

An SLA in the multiplayer bargaining game G is represented as septuplet (P, ρ, U, X, δ, φ, η).

PY = {SCA, TBA_q, SPA_q} is a set of players in the bargaining game where (q = 1, 2,.., n).

ρ = {ρ₁, ρ₂,.., ρ_α, ρ_α +1,.., ρ_m} be the P (ρ^P) or CP (ρ^CP) consist of set of functional parameters ρ₁, ρ₂,.., ρ_α and non-functional parameters {ρ_α +1,.., ρ_m} of the service. U = {U_SCA, U_SPA} is the payoff or utility value in the perspective of SC and SP. X = {x₁,x₂,.., x_n} is the set of negotiable parameters of ρ where X is the proper subset of ρ and n <m.

δ = {δ_i, δ_k} is the responsibility of the negotiation parties where δ_i specifies the conditions on ρ that SCA promised to satisfy and δ_k specifies the conditions on X that SPA promised to satisfy.

φ = {ψ_i, φ_i} is the set of reservation value for the SCA and SPA respectively. Here ψ_i be the upper threshold value that SCA wish to pay and φ_i be the lower threshold value that SPA prepares to accept.

η = {η₁, η₂,..,η_n} is the set of negotiation constraints present in the ρ.

The above multiplayer bargaining game G is modeled using the follows assumptions: (a) All the SPAs are independent and do not cooperate, (b) Both SPAs and SCA are selfish and aim to maximize their revenue, (c) The behavior of the SPA and SCA are determined by the utility function.

3.2 Utility function

The total utility function represents the level of satisfaction for the Negotiation Party (NP) concerning negotiable attributes (X) of the ρ^P or ρ^CP. A utility function estimated for a single and multi-attribute by modeling in two-dimensional spaces as linear and monotone. In the proposed research work, multi-attribute utility function of proposal and counter-proposal is estimated in the perspective of NP, that is, either SCA or SPA which may give different preferences over the utility value U_NP(x_i) and weight W(x_i) of each attribute x_i. The total utility function of the negotiable parameters (X ∈ ρ^P or ρ^CP) of negotiation party $U_{\cdot}^{t} NP (X)$ computed as shown in the Equation (3). ${\underset{\cdot}{U}}^{t} NP (X) = \sum W (xi) * U_{NP} {(x}_{i}) i = 1, 2, . ., m$ (3)

Here the total weight of the attribute should be equal to one i.e. [W(x₁)+ W(x₂)+...+W(x_n)] = 1 and the utility value should be ranges from zero to one i.e. U_NP(x_i) ∈ (0,1). The utility value ‘0’ denotes dissatisfaction and ‘1’ denotes full satisfaction of NP.

In this research work, time-slot, cost, policy, license, device and region scalability are the negotiation attributes considered as SLA parameters or attributes. The utility value of the attribute xi may be high in the perspective of SPA, low in the perspective of SCA and sometimes vice versa. This is because, higher the price will give high utility value for the SPA, lower utility value for SCA, and average utility value for both the parties. Table 2 shows the utility value of negotiable attributes represented in the perspective of both SPA (U_SPA) and SCA (U_SCA).

Table 2

Conceptual analysis of CCSLAF with existing frameworks

Negotiable Attributes	Sub-attributes	U_SCA	U_SPA
Time-slot (x₁)	Peak time (a₁)	1	0
	Non-peak time (b₁)	0.5	0.5
	Free time (c₁)	0	1
Cost (x₂)	Free (a₂)	1	0
	Concession (b₂)	0.5	0.5
	Non-concession (c₂)	0	1
Policy (x₃)	User (a₃)	1	0
	Broker (b₃)	0.5	0.5
	Provider (c₃)	0	1
License (x₄)	Gold (a₄)	1	0
	Silver (b₄)	0.5	0.5
	Bronze (c₄)	0	1
Device (x₅)	Wired device (a₅)	1	0
	High-end wireless device (b₅)	0.5	0.5
	Low-end wireless device (c₅)	0	1
Region scalability (x₆)	Within Data Centre (a₆)	1	0
	Within Region (b₆)	0.5	0.5
	Remote Region (c₆)	0	1

3.3 SLA negotiation strategy

The negotiation strategy tells the process, how the SCA, TBA and SPA negotiate among themselves using the standard protocol and algorithms. In the proposed ADSLANF, a bilateral negotiation is used between the SCA and TBA, and a multilateral (multiple bilateral) negotiation is used between the TBA and SPA. An alternate offer protocol is used for making communication among the parties, which constitutes the negotiation operation. Initially, the ρ^P and the corresponding ρ^CP is generated by SCA in the first round of the negotiation process. This process must iterate until the negotiating parties agree. In this research work, concession and trade-off algorithms are extended with multi-attribute in the SCA and SPA end respectively. The SCA exploits the concession-making algorithm for generating ρ^P by conceding their previous proposal by giving enough discounts on the total utility of the possible proposal. In the opponent side, SPA exploits the trade-off making algorithm for generating ρ^CP by altering the utility of the entire negotiable attribute without making much change in the total utility value of CP.

The SCA exploits Algorithm 1 for handling communication with the TBA and initiates the negotiation process by sending initiate(Negotiation) messages to the broker who in turn responds with the Negotiation ID. Then the Proposal is submitted to the broker and the consumer waits for the reply(Accepted ∥ Rejected ∥ Counter Proposal) message. In case of Accepted reply, SCA will send confirm(Accepted ∥ Rejected) message to the broker and wait for the notification(Accepted ∥ Rejected) message. In case of a Rejected reply, SCA will send confirm(Rejected) message. However, for the reply of Counter Proposal, SCA will checks whether the negotiable parameters p_i value ≤ upper thresholds ψ_i, if yes then it send the confirm(Accepted) message otherwise modifies the Proposal for resubmission and waits for the reply(Accepted ∥ Rejected ∥ Counter Proposal) message.

Algorithm 1 Service Consumer Negotiation Algorithm

Create the SLA Template

Fill template with user expected quality of service requirement

Set the upper thresholds ψ_i for the negotiable parameters x_i

Select the best TB for negotiation

Repeat

Send negotiation request to TBA_q as initiate(Negotiation)

Get the Negotiation ID as a response message

Send the request to TBA_q as submit(Proposal) message

Receive reply(Accepted ∥ Rejected ∥ Counter Proposal) of TBA_q

R1: if (getReply() is Accepted) then

if (negotiable parameters x_i value ≤ ψ_i) then

Send confirm(Accepted) message to TBA_q

Get notification message from TBA_q

if (getNotification () is Accepted) then

Goto: N1

end if

else

Send confirm(Rejected) message to TBA_q

end else

end if

R2: else if (getReply() is Rejected) then

Send confirm(Rejected) message to TBA_q

end else if

R3: else (getReply() is Counter Proposal) then

repeat

if (negotiable parameters p_i value ≤ ψ_i) then

Send confirm(Accepted) message to TBA_q

Get notify (Accepted ∥ Rejected) from TBA_q

if (getNotification() is Accepted) then

Goto: N1

end if

else

Modify Proposal by Concession Algorithm

Send the submit(Proposal) request to TBA_q

Receive reply(∥ ∥) message of TBA_q

if (getReply() is Accepted) then

Goto: R1

end if

else if (getReply() is Rejected) then

Goto: R2

end else if

else (getReply() is Counter Proposal) then

Goto: R3

end else

until (getReply() is Accepted ∥ Rejected)

end else

until (getNotification(Accepted ∥ Rejected))

N1: if (getNotification() is Accepted) then

Sign the Service Level Agreement Document

Send slaReplay(Document) message to TBA_q

Receive slaCommit(Document) from TBA_q

end if

else (getNotification() is Rejected)

Do not sign the Service Level Agreement Document

Stop the negotiation process

End else

In the sense of notification(Rejected) message, the user will not sign the agreement and stop the negotiation process. In case of notification(Accepted) message, the user will send the signed agreement by slaReplay(Document) message and wait for the broker to receive the slaCommit(Document) message.

A Third party broker is an intermediate who uses the Broker Negotiation Algorithm as shown in Algorithm 2 for making communication with service consumers and service providers. As an initial process, the broker will forward the initiate(Negotiation) message to the set of service providers SPA_q and receive the Negotiation ID as a response. Next process, forward the submit(Proposal) message and receives the reply(Accepted ∥ Rejected ∥ Counter Proposal) message from SPA_q. Then forward reply(Accepted) message to SCA for the single Accepted reply, or in case of more than one Accepted reply forward the reply(Accepted) message of best SPA_q who is having less distance. In the case of Counter Proposal reply, the utility function of all the SPA_q is computed and then forward the reply(Counter Proposal) message of the service provider who is having maximum utility value. Otherwise forwards the message as the same in case of the Rejected reply. If all the received replies are Rejected one, then TBA_q decides to send a modify(Proposal) message to SC. On behalf of SC, it just forward the confirm(Accepted || Rejected) message. After receiving the notification message from SPA_q, it forwards the notify(Accepted) message and stops the negotiation for the notify(Rejected) message. As a final process, it forwards the slaReplay (Document) and slaCommit(Document) message from SC and SPA_q respectively.

Algorithm 2 Broker Negotiation Algorithm

Receive the negotiation request from the Service Consumer SC

Get the upper thresholds ψ_i for the negotiable parameters x_i

Get the set of Service Provider Agent say SPA_q

repeat

for q = 1 to n do

Forward the initiate(Negotiation) message to SPA_q

Receive the Negotiation ID as a response message

Forward the submit(Proposal) message to SPA_q

Receive reply(Accepted ∥ Rejected ∥ Counter Proposal) of SPA_q

end for

define f[], q = 0;

for q = 1 to n do

if (SPA_q.getReply() is Accepted) then

R1: if (negotiable parameters x_i value ≤ ψ_i) then

compute distance function of SPA_q

add to the D_f [q +1]

end if

else if (SPA_q.getReply() is Rejected) then

Goto: S1

end else if

end for

Compute Min(D_f [])

R2: for g = 1 to f.length() do

if (MIN(D_f []) = = D_g)

Forward the Accepted response to SC

Get confirm(Accepted ∥ Rejected) from SC

Choose the SPA_q which is having D_g

R3: if (getConfirm() is Accepted) then

Forward confirm(Accepted) message to SPA_q

Get notify(Accepted) message from the SPA_q

Forward notify(Accepted) message to SC

end if

else (getConfirm() is Rejected)

Forward confirm(Rejected) message to SPA_q

Goto: S1

end else

end if

end for

define v[],r = 0;

for l = 1 to n do

if (SPA_l.getReply() is Counter Proposal) then

if (f.length() ! = 0) then

Send confirm(Rejected) message to SPA_l

Goto: S1

end if

else (f.length() = = 0)

Goto: R1

if (negotiable parameters x_i value > ψ_i) then

compute utility of SPA_l as

U_{\cdot}^{t} t ext SCA

(ρ^CPl)

add to the U_v [r +1]

end if

end else

end if

end for

if (f.length()! = 0) then

Compute Min(D_f [])

Goto: R2

end if

else if (f.length() = = 0)

Send modify(Proposal) message to SC

end else if

if (v.length()! = 1) then

Compute Max(U_v [])

for h = 1 to v.length() do

if (Max(U_v []) = = U_h)

Choose the SPA_q which is having U_h

Forward reply (Counter Proposal) to SC

Get confirm(Accepted ∥ Rejected) from SC

Goto: R3

endif

end for

endif

else if (v.length() = = 1)

Forward reply (Counter Proposal) message to SC

end else if

until (getNotification(Accepted ∥ Rejected))

if(getNotification is (Accepted))

Receive slaReplay(Document) message from SC

Forward slaReplay(Document) message to the SPA_q

Receive slaCommit(Document) message from the SPA_q

Forward slaCommit(Document) message to the SC

endif

S1: Stop the Negotiation

The service provider exploits the Service Provider Negotiation Algorithm as shown in Algorithm 3 for making communication with the TBA_q. As the first process, SPA_q receives the request and sends back the Negotiation ID as a response. Then receives the Proposal message and check whether the negotiable parameters p_k value ≥ φ_i, if yes send the reply(Accepted) message otherwise check whether the negotiable parameters p_k value < φ_i, if yes send the Rejected reply else it reply with the Counter Proposal message. Next process, SPA_q receives the confirmation from TBA_q, then it responds with notify(Accepted) message in case of confirm(Accepted) message and responds with notify(Rejected) message in case of confirm(Rejected) message. Finally, concerning Accepted notification, it waits for the slaReplay(Document) message to send back the slaCommit(Document).

Algorithm 3 Service Provider Negotiation Algorithm

Receive the negotiation request from the TBA_j

Send Negotiation ID to the TBA_j

Receive the submit(PROPOSAL) message from the TBA_q

Set the lower thresholds φ_texti for the negotiable parameters x_k

repeat

R1: if (negotiable parameters x_k value ≥ φ_i) then

Send reply(Accepted) message to the TBA_q

Get confirm(Accepted || Rejected) from the TBA_q

if (getConfirm() is Accepted) then

Send notify(Accepted) message to the TBA_q

Goto: N1

endif

else

Send notify(Rejected) message to the TBA_q

end else

end if

else if (negotiable parameters x_k value < φ_i) then

Send reply(Rejected) message to the TBA_q

end else if

else (parameters x_k value is out of range to φ_texti)

Generate Counter Proposal byTradeoff Algorithm

Send reply(Counter Proposal) message to TBA_q

Receive submit(PROPOSAL) message from the TBA_q

Goto: R1

end else

until (getConfirm(Accepted ∥ Rejected))

N1: if (sendNotification() is Accepted) then

Wait for slaReplay(Document) from the TBA_q

Send the slaCommit(Document) to the TBA_q

endif

The process of generating proposal and counterproposal using the concession and trade-off making algorithms may lead to an increase in the TNT between the SCA and SPA. This situation arises due to the incompatible generation of P such that it leads to generating multiple CPs without letting to the acceptance of either party. In order to avoid the increase in TNT, the number of iterations involved during the negotiation process is fixed randomly (say 5 or 10). During these negotiation rounds, there is the possibility that the negotiation parties may or may not lead to the acceptance of a proposal or counter-proposal. Therefore, to provide an acceptable and feasible solution a broker uses a decision-making model before reaching the final round of the negotiation process. Moreover, it is tough to select the optimal solution (P with high utility) from the list of CP given by the service providers. A novel decision-making model is represented in the following section using the game theory approach.

3.4 Decision-making model

Central part of the negotiation framework is the decision-making model which helps to find the feasible and acceptable solution for the number of counter proposals generated by each SP. Finally, the TB will choose the optimal solution among the feasible solutions obtained from different SP and suggest that solution as the best offer for signing the agreement. The SC can accept the best offer or select any other offer from the remaining feasible solutions of remaining SP. The negotiation and optimal selection process can be modeled as the bargaining game with one SCA, ‘n’ SPA and a TB with ‘n’ number of agents such as TBA. Since there is no direct negotiation between the parties (SCA and SPA), the proposal and counter-proposal are sent through an intermediate TBA. In this scenario, a bargaining game is considered as a pair of ‘n’ two-player non-zero sum game, i.e., ‘n’ number of one-to-one negotiation between the players. Here the many one-to-one negotiations represent the communication between TBA with SPA. Since TBA is negotiating on behalf of SCA, the mathematical model is derived accordingly.

Let ‘u’ be the utility value of the SCA generated by the TBA_i (i = 1, 2,.., n), ‘v’ be the utility value of SPA_i (i = 1, 2,.., n) and c(S) be the solution point in the set S which is convex, compact and includes the origin. Corollary 1 is developed using the three Utility Theories (1, 2, and 3) [23].

Utility Theory 1: If x is the point in S, i.e., there exists another point y in S with the property u(y) >u(x) and v(y) >v(x), then x ¬ = c(S).

Utility Theory 2: If the set T contains the set S and c(T) is in S, then c(T) = c(S).

Utility Theory 3: If S is symmetric, u and v display this, then c(S) is a point of the form (a, a), that is, a point on the line u = v.

Corollary 1: If the utility pair (u_i, v_i) is a feasible solution to the game G which is symmetrical in the line u = v. Then the optimal solution to the G is derived from the Nash equilibrium solution. This means that the optimal solution is a point p_i with the utility value (u_i, v_i) = (1,1) for the players which gives the maximum and equal utility to both players.

Proof: Let us plot the utility of the CPs generated by multiple SPAs by taking u ( ${\underset{\cdot}{U}}_{SCA}^{t}$ ) along X-axis and v ( ${\underset{\cdot}{U}}_{SPA}^{t}$ ) along Y-axis. Now draw the tangent in the first quadrant with the angle of 45⁰ (say it is the line u = v). To select the optimal solution, randomly choose some points (say p₁, p₂, p₃, p₄, p₅, p₆...p_n) in the line u = v as shown in Fig. 2.

Fig. 2

Optimal solution in the utility pair (u,v).

First, plot the point p₁ as (u₁,v₁) which is the utility pair of first two players i.e. the utility value of SC negotiated by TBA₁ and the utility value of SP₁ negotiated by the SPA₁. For example, give the minimum utility value to both the players, which satisfies the Nash equilibrium solution. Since the utility value ranges from 0 to 1, here 0 is assigned to be the minimum utility value for both the players.

$p_{1} = (u_{1}, v_{1}) = (0, 0)$ (4)

Similarly, plot the point p₂, p₃, p₄, p₅, p₆... p_n for the other set of players with their corresponding next higher utility value such as (u₂,v₂), (u₃,v₃), (u₄,v₄), (u₅,v₅), (u₆,v₆)... (u_n,v_n) where (u_n,v_n) is the maximum utility value of the players. $p_{2} = (u_{2}, v_{2}) = (0.1, 0.1)$ (5) $p_{3} = (u_{3}, v_{3}) = (0.3, 0.3)$ (6) $p_{4} = (u_{4}, v_{4}) = (0.5, 0.5)$ (7) $p_{5} = (u_{5}, v_{5}) = (0.7, 0.7)$ (8) $p_{6} = (u_{6}, v_{6}) = (0.9, 0.9)$ (9) $p_{n} = (u_{n}, v_{n}) = (1, 1)$ (10)

Hence from the above Equations (4), (5), (6), (7), (8), (9) and (10) it is clear that the optimal solution to i^th pair of ‘n’ two player game with the utility range 0 to 1 is determined as shown in Equation (11). $p_{i} = (u_{i}, v_{i}) = (1, 1)$ (11)

4 Experimental evaluation

To demonstrate the performance of proposed ADSLANF, an experimental setup is made by integrating the JADE and CloudSim toolkit. In this setup, JADE toolkit enables the SC, TB and SP to act as agents and CloudSim enables the SPA to provision the negotiated cloud service. Assume the negotiations are fixed to 10 rounds where the SCA and SPA generate the ρ^P and ρ^CP using concession and trade-off making algorithms and there is no rejection of ρ before the 10 round. The negotiation round of TBAs (TBA₁, TBA₂, TBA₃, TBA₄ and TBA₅) with corresponding SPAs (SPA₁, SPA₂, SPA₃, SPA₄ and SPA₅) are simulated to 10 rounds. For simplicity, only the negotiation rounds taken between TBA₁ and SPA₁ is shown in Table 3 with attribute weight. Similarly, the negotiation round between other agents was also observed respectively. In order to select the best ρ^CP from each SP, the decision-making model is used which will suggest the SC with optimal solution (ρ^CP with maximum utility for SC and SP) for accepting and signing the SLA.

Table 3
Negotiation Rounds of TBA₁ (on behalf of SCA) with SPA₁

ρ x₁ x₂ x₃ x₄ x₅ x₆ ${\underset{\cdot}{U}}_{SCA}^{t}$ ${\underset{\cdot}{U}}_{SPA 1}^{t}$

(40%) (20%) (10%) (10%) (10%) (10%)

ρ ^P1 a₁ a₂ a₃ a₄ a₅ a₆ 1 0

ρ ^CP1 c₁ c₂ c₃ c₄ c₅ c₆ 0 1

ρ ^P2 a₁ a₂ b₃ a₄ a₅ a₆ 1 0.1

ρ ^CP2 c₁ c₂ b₃ c₄ c₅ c₆ 0.1 1

ρ ^P3 a₁ b₂ b₃ a₄ a₅ a₆ 0.9 0.2

ρ ^CP3 c₁ b₂ c₃ b₄ c₅ c₆ 0.2 0.9

ρ ^P4 a₁ c₂ b₃ a₄ a₅ a₆ 0.8 0.3

ρ ^CP4 a₁ c₂ c₃ c₄ c₅ c₆ 0.4 0.6

ρ ^P5 a₁ c₂ b₃ a₄ a₅ b₆ 0.7 0.3

ρ ^CP5 a₁ c₂ c₃ b₄ c₅ c₆ 0.5 0.6

ρ ^P6 a₁ b₂ a₃ c₄ c₅ b₆ 0.7 0.4

ρ ^CP6 b₁ c₂ b₃ c₄ c₅ b₆ 0.3 0.7

ρ ^P7 b₁ b₂ a₃ b₄ b₅ a₆ 0.6 0.4

ρ ^CP7 b₁ b₂ c₃ a₄ b₅ c₆ 0.5 0.6

ρ ^P8 b₁ b₂ a₃ c₄ b₅ a₆ 0.6 0.5

ρ ^CP8 c₁ b₂ b₃ a₄ b₅ a₆ 0.4 0.6

ρ ^P9 b₁ b₂ b₃ a₄ b₅ b₆ 0.55 0.45

ρ ^CP9 a₁ b₂ c₃ c₄ c₅ c₆ 0.5 0.5

ρ ^P10 b₁ b₂ b₃ b₄ b₅ b₆ 0.5 0.5

ρ ^CP10 a₁ b₂ c₃ c₄ c₅ c₆ 0.5 0.5

ρ	x₁	x₂	x₃	x₄	x₅	x₆	${\underset{\cdot}{U}}_{SCA}^{t}$	${\underset{\cdot}{U}}_{SPA 1}^{t}$
ρ ^P1	a₁	a₂	a₃	a₄	a₅	a₆	1	0
ρ ^CP1	c₁	c₂	c₃	c₄	c₅	c₆	0	1
ρ ^P2	a₁	a₂	b₃	a₄	a₅	a₆	1	0.1
ρ ^CP2	c₁	c₂	b₃	c₄	c₅	c₆	0.1	1
ρ ^P3	a₁	b₂	b₃	a₄	a₅	a₆	0.9	0.2
ρ ^CP3	c₁	b₂	c₃	b₄	c₅	c₆	0.2	0.9
ρ ^P4	a₁	c₂	b₃	a₄	a₅	a₆	0.8	0.3
ρ ^CP4	a₁	c₂	c₃	c₄	c₅	c₆	0.4	0.6
ρ ^P5	a₁	c₂	b₃	a₄	a₅	b₆	0.7	0.3
ρ ^CP5	a₁	c₂	c₃	b₄	c₅	c₆	0.5	0.6
ρ ^P6	a₁	b₂	a₃	c₄	c₅	b₆	0.7	0.4
ρ ^CP6	b₁	c₂	b₃	c₄	c₅	b₆	0.3	0.7
ρ ^P7	b₁	b₂	a₃	b₄	b₅	a₆	0.6	0.4
ρ ^CP7	b₁	b₂	c₃	a₄	b₅	c₆	0.5	0.6
ρ ^P8	b₁	b₂	a₃	c₄	b₅	a₆	0.6	0.5
ρ ^CP8	c₁	b₂	b₃	a₄	b₅	a₆	0.4	0.6
ρ ^P9	b₁	b₂	b₃	a₄	b₅	b₆	0.55	0.45
ρ ^CP9	a₁	b₂	c₃	c₄	c₅	c₆	0.5	0.5
ρ ^P10	b₁	b₂	b₃	b₄	b₅	b₆	0.5	0.5
ρ ^CP10	a₁	b₂	c₃	c₄	c₅	c₆	0.5	0.5

In order to reach the agreement during the negotiation process, the TB exploits the decision-making model before the final round (say nin^th round) to suggest the optimal solution of CPs received from multiple SPs. This process will make sense for the SC to reach the agreement in the final round (say 10^thround), by accepting any of the optimal solutions obtained from the SPs. First, the optimal solution for the number of CPs presents in Table 3 can be derived using the decision-making model defined in section 4.4. Consider u and v is the total utility value in the perspective of TBA_q (on behalf of SCA) and SPA_q. Let us plot the point p₁, p₂, p₃, p₄, p₅, p₆, p₇, p₈ and p₉ with their corresponding utility value (u₁,v₂) = (0,1), (u₂,v₂) = (0.1,1), (u₃,v₃) = (0.2,0.9), (u₄,v₄) = (0.4,0.6), (u₅,v₅) = (0.5,0.6), (u₆,v₆) = (0.3,0.7), (u₇,v₇) = (0.5,0.6), (u₈,v₈) = (0.4,0.6) and (u₉,v₉) = (0.5,0.5) respectively by placing u in X-axis and v in Y-axis. By applying corollary 1, the optimal solution is obtained from the graph shown in Fig. 3, where the point p₉ with utility pair (0.5,0.5) which gives the maximum and equal utility to both the players (TBA₁ and SPA₁).

Fig. 3

Utility value of TBA₁ (on behalf of SCA) with SPA₁.

Similarly, the 10 utility pairs (points) of other TBAs and SPAs can be plotted as graphs like Fig. 3. After applying corollary 1, the optimal solution point for the corresponding figures with respect to players (TBA₂ and SPA₂) is observed as p₇ = p₈ = (0.5,0.5), (TBA₃ and SPA₃) is observed as null (no optimal solution), (TBA₄ and SPA₄) is observed as null (no optimal solution), and (TBA₅ and SPA₅) is observed as p₇ = p₉ = (0.5,0.5). Hence, during the 10^thround, TBAs will suggest the SCA with all the optimal solutions obtained for the respective SPAs. This suggestion will make sense of the SCA in the 10^th round of negotiation, either to accept the counter-proposal for signing SLA or to modify the proposal according to the suggested optimal solution, by changing the negotiation parameters without affecting the total utility of the negotiation party. Hence the proposed decision-making model for the novel ADSLANF will help to result in the signing of SLA by the negotiation parties. Finally, in the 10th round of negotiation between the (TBA₃ and SPA₃) and (TBA₄ and SPA₄) result in the optimal solution.

The proposed ADSLANF is compared with existing Price and Time-slot Negotiation mechanism concerning the number of SPAs committed to signing SLA during the negotiation process. From Table 3 simulation result shows that the negotiation without decision-making model may leads 3 SPAs to commit in SLA. However, after applying the decision-making model during the negotiation process leads all the SPAs to commit in SLA. Because of the suggestiongiven by the TBA (using a decision-making model) helps the SPA to generate the proposal within the acceptable range, which probably results in a commitment by all the SPAs. Hence, the suggestion of generating proposal within the acceptable range increases the number of SLA commitment from 2 SPAs to 5 SPAs.

To evaluate the proposed ADSLANF (indirect agent-based negotiation framework), the negotiation process is simulated with and without an agent for the direct and indirect negotiation information given in Tables 4 and Table 5. From the literature survey, negotiation frameworks are classified into four types such as DNWOA [29], DNWA [10 , 18], INWOA [30] and INWA (proposed ADSLANF) for comparing the performance.

Table 4

Direct Negotiation Information with and without Agent

Negotiation		PGD (sec)	CPGD (sec)	ECT between SC and SP (sec)
				SP₁	SP₂	SP₃	SP₄	SP₅
Direct	Without Agent	5	5	10	12	11	12	10
	With Agent	2	2	10	12	11	12	10

Table 5

Indirect Negotiation Information with and without Agent

Negotiation		PGD (sec)	CPGD (sec)	TBFD (sec)	ECT between SC and TB (sec)	ECT between TB and SP (sec)
						SP₁	SP₂	SP₃	SP₄	SP₅
Indirect	Without Agent	5	5	5	5	5	7	6	7	5
	With Agent	2	2	2	5	5	7	6	7	5

To test the performance of the proposed negotiation model ADSLNF (INWA) with the existing DNWOA, DNWA and INWOA model are compared using the simulated experiment concerning10 negotiation rounds. The TNT values of the corresponding negotiation model are computed using Equations (10) and (11), and the results are compared as shown in Table 6.

Table 6

Results of Negotiation Models

Negotiation Models	Result of TNT values (Seconds)
	1 Round	10 Rounds	50 Rounds	100 Rounds
DNWOA	160	1600	8000	16000
DNWA	130	1300	6500	13000
INWOA	120	1200	6000	12000
INWA (Proposed ADSLANF)	84	840	4200	8400

The performance of DNWOA, DNWA, INWOA and proposed ADSLANF is compared by varying the number of negotiation rounds (1, 10, 50 and 100) concerning TNT as shown in Fig. 4. Since the proposed ADSLANF follows the indirect way of negotiation, the resultant graph shows that the proposed model takes very less TNT when compared to other negotiation models.

Fig. 4

Performance Evaluation of DNWOA, DNWA, INWOA and ADSLANF.

5 Conclusion and future work

In this research work, proposed ASLANF was evaluated by comparing with the existing direct and indirect negotiation mechanisms (DNWOA, DNWA and INWOA). Empirical evaluation is carried out for the scenario of multiple one-to-many (1 service consumer and n service provider) model which shows that the proposed ASLANF with three negotiation algorithms takes very less TNT when compared to the existing negotiation mechanism. In order to show the performance of the agent-based model, DNWA and ADSLANF (INWA) were compared concerning TNT. This comparison shows that the proposed ADSLANF gets 46 seconds less TNT than the existing DNWA. Therefore, on an average the proposed ADSLANF improves the performance by reducing 30%, 18% and 14% of TNT compared to existing DNWOA, DNWA and INWOA respectively. In future, the negotiation decision support system will be designed to support the hierarchical computing capability by integration edge computing and cloud infrastructure. Due to the edge computing capability, quick negotiation decisions can be carried out by the negotiation participant without any computational processing delay from the cloud servers.

References

Buyya

, Garg

S.K.

and Calheiros

R.N.

, SLA-Oriented Resource Provisioning for Cloud Computing: Challenges, Architecture, and Solutions, International Conference on Cloud and Service Computing, Hong Kong, China, (2011), 1–10.

Rajavel

and Thangarathanam

, Adaptive Probabilistic Behavioural Learning System for the Effective Behavioural Decision in Cloud Trading Negotiation Market, Future Generation Computer Systems58 (2016), 29–41.

Rajavel

and Thangarathanam

, Optimizing Negotiation Conflict in the Cloud Service Negotiation Framework using Probabilistic Decision Making Model, The Scientific World Journal2015 (2015), 1–16.

Rajavel

and Thangarathanam

, ADSLANF: A negotiation framework for the cloud management system using Bulk Negotiation Behavioural Learning approach, Turkish Journal of Electrical Engineering & Computer Sciences25 (2017), 563–590.

Rajavel

and Thangarathanam

, A Negotiation Framework for the Cloud Management System using Similarity and Gale Shapely Stable Matching approach, KSII Transactions on Internet and Information Systems9 (2015), 2050–2077.

Rajavel

, et al., Adaptive neuro-fuzzy behavioral learning strategy for effective decision making in the fuzzy-based cloud service negotiation framework, Journal of Intelligent & Fuzzy Systems36 (2019), 2311–2322.

Rajavel

and Thangarathanam

, SLAOCMS: A Layered Architecture of SLA Oriented Cloud Management System for Achieving Agreement during Resource Failure, Advances in Intelligent Systems and Computing236 (2014), 801–809.

Zheng

, Martin

, Powley

and Brohman

, Applying Bargaining Game Theory toWeb Service Negotiation, IEEE International Conference on Services Computing, Miami, Florida, USA, (2010), 218–225.

Richter

, Chhetri

M.B.

, Kowalczyk

and Vo

Q.B.

, Establishing composite SLAs through concurrent QoS negotiation with surplus redistribution, Concurrency and Computation: Practice and Experience24 (2012), 938–955.

10.

Son

and Sim

K.M.

, A Price- and-Time-Slot-Negotiation Mechanism for Cloud Service Reservations,—, Part B: Cybernetics42 (2012), 713–728.

11.

Son

and Sim

K.M.

, An Adaptive Tradeoff Algorithm for Multi-issue SLA Negotiation, Korea, CCIS 121, International Conferences on GDC and CA121 (2010), 32–41.

12.

Akhani

, Chuadhary

and Somani

, Negotiation for Resource Allocation in IaaS Cloud, COMPUTE - 4th Annual ACM Bangalore Conference, India, (2011), Article 15.

13.

Yoo

and Sim

K.M.

, A Multilateral Negotiation Model for Cloud Service Market, Korea, CCIS 121, International Conferences on GDC and CA121 (2010), 54–63.

14.

Chaw

L.T.

, Keong

P.K.

, Fong

A.T.

and Pei

L.G.

, Multi-area QoS Provisioning using Hierarchical Bandwidth Brokers Architecture, Malaysian Journal of Computer Science21 (2008), 88–100.

15.

Ragone

, Noia

T.D.

, Sciascio

E.D.

and Donini

F.M.

, Alternating-offers protocol for multi-issue bilateral negotiation in semantic-enabled marketplaces, ACM, ISWC’07/ASWC’07 Proceedings of the 6th International The semantic web and 2nd Asian conference on Asian semantic web conference, (2007), 395–408.

16.

Resinas

, Fernandez

and Corchuelo

, A bargaining-specific architecture for supporting automated service agreement negotiation systems, Science of Computer Programming77 (2012), 4–28.

17.

Kalai

, Games in Coalitional Form, Forthcoming in the New Palgrave Dictionary of Economics, second edition, (2007).

18.

, Lesser

, Irwin

and Zink

, Automated Negotiation with Decommitment for Dynamic Resource Allocation in Cloud Computing, 9th International Conference on Automated Agents and Multiagent Systems, Toronto, Canada, (2010), 981–988.

19.

Ragone

, Noia

T.D.

, Sciascio

E.D.

and Donini

F.M.

, Propositional-Logic Approach to One-Shot Multi Issue Bilateral Negotiation, ACM SIGecom Exchanges5(5) (2005), 11–21.

20.

Kurbel

, Loutchko

and Teuteberg

, FuzzyMAN: An Agent-Based Electronic Marketplace with a Multilateral Negotiation Protocol, Second German Conference on Multiagent System Technologies, Erfurt, Germany, (2004), LNAI 3187, 126–140.

21.

Juhasz

and Paul

, Scalability Analysis of the Contract Net Protocol, IEEE International Symposium on Cluster Computing and the Grid (2002), 346.

22.

Alfredson

, Cungu

, Negotiation Theory and Practice: A Review of the Literature, FAO Policy Learning Program, Conceptual and Technical Material, EASYPol Module 179.

23.

Nash

, Two-Person Cooperative Games, Econometrica21 (1953), 128–140.

24.

Magoc

and Kreinovich

, A New Simplified Derivation of Nash Bargaining Solution, Applied Mathematical Sciences3 (2009), 1097–1101.

25.

Driesen

, Perea

and Peters

, The Kalai-Smorodinsky bargaining solution with loss aversion, Mathematical Social Sciences61 (2011), 58–64.

26.

Bozbay

, Dietrich

and Peters

, Bargaining with endogenous disagreement: the extended Kalai-Smorodinsky solution, Games and Economic Behavior74 (2011), 407–417.

27.

Zhang

, A Logical Model of Nash Bargaining Solution, 19th International Joint conference on Artificial Intelligence, Edinburgh, Scotland, UK, (2005), 983–988.

28.

Saglam

, A simple axiomatization of the egalitarian solution, Munich Personal RePEc Archive (MPRA), TOBB University of Economics and Technology, Department of Economics, Working Papers No. 1201, (2012).

29.

Koshutanski

, Lazouski

, Martinelli

and Mori

, Enhancing grid security by fine-grained behavioral control and negotiation-based authorization, International Journal of Information Security8 (2009), 291–314.

30.

Brandic

, Music

and Dustdar

, Service Mediation and Negotiation Bootstrapping as First Achievements Towards Self-adaptable Grid and Cloud Services, 6th International Conference Industry Session on Grids Meets Autonomic Computing, Barcelona, Spain, June 15–19, (2009), 1–8.