INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
A Review of Cost Management Approaches in Cloud Computing  
1Ankit Pandey,1Rounak Bhardwaj,1Keshab Mondal,1Ritesh Choudhry, 2, *Pradip Ghanty  
1
Final Year Students of B. Tech in Computer Science and Engineering (Data Science), Kazi Nazrul  
University, Asansol, West Bengal, India.  
2Assistant Professor & Coordinator, Department of Computer Science, Asansol Girls’ College, Asansol,  
West Bengal, India  
*Corresponding author  
DOI : https://doi.org/10.51583/IJLTEMAS.2025.1411000084  
Received: 28 November 2025; Accepted: 03 December 2025; Published: 18 December 2025  
ABSTRACT  
Cloud computing has revolutionized the IT industry by offering scalable and on-demand resources under a pay-  
as-you-go model. However, the same flexibility that makes cloud computing attractive also introduces  
challenges in managing costs effectively. Cloud Cost Management (CCM) focuses on monitoring, analyzing and  
optimizing expenditure to achieve an ideal balance between performance and affordability. This paper reviews  
existing cost management strategies, techniques, and tools used across different cloud environments. It highlights  
traditional budgeting and monitoring methods, advanced optimization models using machine learning, and  
automation through FinOps practices. There view also identifies current challenges such as complex pricing  
structures, lack of cost visibility in multi-cloud environments, and unpredictable billing patterns. Finally, the  
paper discusses emerging trends and future research directions aimed at intelligent and sustainable cost  
optimization in cloud ecosystems.  
INTRODUCTION  
The rapid adoption of cloud computing [8] has fundamentally transformed how organizations deploy and manage  
IT infrastructure. Instead of maintaining costly on-premises systems, businesses increasingly rely on cloud  
service providers such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP)  
to access computing resources on demand. While the pay-as-you-go pricing model offers significant flexibility  
and scalability advantages, it frequently results in unpredictable and escalating expenses when cloud resources  
are not managed efficiently.  
As cloud environments grow in scale and complexity, Cloud Cost Management (CCM) has emerged as a critical  
focus area for both enterprises and researchers. Effective CCM ensures optimal allocation of cloud resources  
[11], thereby minimizing over-provisioning, reducing idle resource usage, and improving overall financial  
efficiency. CCM encompasses a combination of financial monitoring, operational governance, and technological  
optimization practices that aim to balance performance requirements with cost constraints.  
Although major cloud platforms provide cost-monitoring dashboards and pricing calculators, these solutions  
often prove insufficient in handling complex cost dynamics in multi-cloud environments [3]. Differences in  
pricing models, billing metrics, and service abstractions across providers limit unified cost visibility and  
centralized governance. To address these limitations, researchers and practitioners have proposed a wide range  
of approaches, spanning static budgeting and rule-based monitoring techniques to dynamic resource scheduling  
mechanisms and AI-driven cost prediction models [9].  
Despite the availability of diverse cost management strategies, existing studies are often fragmented, focusing  
on individual techniques without offering a comprehensive comparative perspective. This review paper aims to  
systematically analyze and synthesize the existing literature on cloud cost management, evaluate the  
890  
effectiveness and limitations of current approaches, and highlight emerging trends that have the potential to  
www.ijltemas.in  
Page 890  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
shape the next generation of cloud cost optimization. The ultimate objective is to provide a consolidated  
understanding of how cloud cost management strategies evolve to meet the growing demands of scalable,  
sustainable, and intelligent cloud ecosystems.  
REVIEW METHODOLOGY  
This review follows a structured and systematic literature analysis to examine existing approaches in cloud cost  
management. Relevant research articles were collected from widely recognized digital libraries and academic  
repositories, including IEEE Xplore, SpringerLink, ACM Digital Library, ScienceDirect, and arXiv.  
The selection process focused on peer-reviewed journal articles, conference papers, and high-quality preprints  
published between 2010 and 2025, emphasizing studies related to cloud cost optimization, FinOps practices,  
AI/ML-based cost prediction, dynamic resource provisioning, and multi-cloud cost management. In addition to  
recent studies, a limited number of foundational works were included to establish baseline concepts in cloud  
pricing and resource allocation.  
Papers were selected based on the following inclusion criteria:  
Relevance to cloud cost management or optimization,  
)
ii)  
Discussion of financial, operational, or algorithmic cost-control mechanisms, and  
Conceptual clarity or experimental validation.  
iii)  
Studies were excluded if they focused solely on performance optimization without explicit consideration of cost,  
lacked sufficient technical or analytical depth, or were unrelated to cloud-based environments.  
The selected literature was systematically analyzed and classified into three major categories-static approaches,  
dynamic approaches, and AI/ML-based optimization techniques-based on their adaptability, automation level,  
and decision-making capability. This classification provides a coherent analytical framework to compare  
existing methods, identify limitations, and highlight emerging research directions in cloud cost management.  
Cloud Cost Management Overview  
Cloud Cost Management (CCM) refers to the set of strategies, tools, and practices used to monitor, analyze, and  
optimize the financial expenditure associated with cloud resources. In The pay-as-you-go model adopted by  
most cloud service providers, organizations are charged based on the resources they consume such as computing  
power, storage capacity, network bandwidth, and additional services. Although this model offers scalability and  
flexibility, it also introduces complexity in predicting, controlling, and optimizing costs, especially in multi-  
cloud or hybrid environments.  
Effective CCM involves several key objectives  
Cost Visibility - providing detailed insight into where and how money is being spent across cloud resources and  
services [6].  
Cost Optimization-identifying underutilized or idle resources and adjusting usage to minimize waste.  
Forecasting and Budgeting [14] –using analytics and predictive models to estimate future costs and plan budgets  
accordingly.  
Governance and Accountability-ensuring that each team, projector department is responsible for its cloud  
expenditure.  
891  
www.ijltemas.in  
Page 891  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
The major cost drivers in cloud computing include  
Figure 1: Conceptual Cloud Cost Management framework  
Computer resources: Virtual machines (VMs), containers and serverless functions that run applications.  
Storage: Costs associated with persistent storage, database services, and data backup.  
Data transfer: Network usage costs, particularly dataegress between regions or providers.  
Licensing and third-party services: Software subscriptions, APIs, and managed services integrated into cloud  
operations.  
CCM frameworks often employ automation to control costs in real time through features such as auto-scaling  
[1], instance scheduling, and rightsizing. Additionally, practices like tagging (assigning meta data to resources)  
and FinOps (Financial Operations) help organizations align financial goals with engineering decisions.  
In essence, Cloud Cost Management is not just about reducing expenses; it’s about achieving an optimal balance  
between performance, scalability, and financial efficiency. As organizations increasingly adopt multi-cloud and  
hybrid models, CCM plays a critical role in ensuring operational sustainability and strategic decision-making.  
Conceptual Framework for Cloud Cost Management  
Figure 1 presents a conceptual Cloud Cost Management (CCM) framework that integrates technical cost  
optimization mechanisms with financial governance practices. The framework illustrates the flow from cloud  
resource consumption to monitoring, cost visibility, and optimization. Static, dynamic, and AI/ML-based  
techniques operate at the optimization layer to control expenditure, while FinOps practices enforce  
892  
accountability, budgeting, and policy-driven governance. This layered structure highlights that effective cloud  
www.ijltemas.in  
Page 892  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
cost management requires coordinated interaction between technical automation and organizational governance  
rather than isolated cost reduction strategies.  
Existing Approaches and Techniques  
CloudCostManagementhasevolvedthroughavarietyofapproaches,rangingfrombasiccost tracking methods to  
advanced AI-driven optimization frameworks. These techniques can broadly be categorized into static, dynamic,  
and intelligent (AI/ML-based) approaches. The approaches are described in subsections. In Table 1 Comparison  
of Cloud Cost Management Approaches are depicted.  
Static Approaches  
Static approaches to cloud cost management represent the most traditional layer of cost-control mechanisms.  
These approaches rely on predefined rules, fixed budgets, and manual or semi-automated monitoring to regulate  
cloud expenditures. Due to their simplicity and low implementation complexity, static techniques are widely  
adopted in early-stage cloud deployments and environments with relatively stable and predictable workloads.  
METHODOLOGY:  
Budgeting and Cost Allocation: Where spending limits are predefined for departments, projects, or services,  
and cloud usage is monitored against these limits. Such budgeting mechanisms help organizations gain financial  
visibility and prevent uncontrolled spending.  
Tagging and Resource Grouping: It support accountability by associating metadata with cloud resources,  
allowing costs to be traced back to specific teams or applications. These practices are essential for governance  
and reporting, particularly in enterprise-scale deployments [11].  
Scheduling and Instance Shutdown: Configuring resources to automatically shut down during off-peak hours  
[16].  
Cost Alerts and Reporting: Using built-in cloud dash boards such as AWS Cost Explorer, Azure Cost  
Management, and GCP Billing Reports, which notify users when predefined budget thresholds are exceeded and  
provide periodic expenditure summaries [6].  
Critical Evaluation:  
Although static approaches improve cost visibility and financial discipline, they lack adaptability in dynamic  
cloud environments. Fixed budgets and rule-based alerts do not respond effectively to sudden workload  
variations, leading to delayed corrective actions and potential cost overruns. These approaches are particularly  
ineffective for micro services-based, event-driven, or highly elastic applications, where resource demand  
changes rapidly. Consequently, static techniques are best suited as a foundational governance layer rather than  
a standalone solution and must be complemented by dynamic or intelligent optimization mechanisms for  
effective cloud cost management.  
Dynamic Approaches  
Dynamic approaches to cloud cost management focus on adjusting resource allocation in response to real-time  
workload variations. Unlike static techniques, these methods continuously monitor system performance and  
automatically scale resources to balance cost efficiency and application performance. Dynamic approaches are  
particularly effective in cloud-native environments where demand fluctuates due to varying user activity or  
event-driven workloads.  
Common techniques include:  
Auto-Scaling: It enables cloud resources to scale up or down based on predefined performance metrics such as  
893  
www.ijltemas.in  
Page 893  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
CPU utilization, memory usage, or incoming request rate. Auto-scaling may be implemented either horizontally  
by adding or removing compute instances or vertically by resizing instance capacities. Studies on cloud-native  
applications show that well-designed auto-scaling policies can significantly minimize resource underutilization  
and reduce operational costs while maintaining acceptable performance levels [1].  
Spot and Preemptible Instances: A dynamic strategy which exploit unused cloud capacity at significantly  
reduced prices. Although these instances offer substantial cost savings, their availability is not guaranteed, as  
they may be terminated without notice. Consequently, this approach is most suitable for fault-tolerant, batch, or  
non-critical workloads.  
Serverless Computing: Using pay-per-execution services like AWS Lambda or Google Cloud Functions to  
eliminate idle infrastructure costs.  
Load-Balancing and Scheduling: Dynamic approaches are particularly beneficial in multi-cloud environments,  
where load balancing and runtime scheduling can distribute workloads across providers based on cost and  
performance considerations. Research on runtime microservice re-orchestration demonstrates that dynamically  
migrating workloads can achieve meaningful cost reductions in multi-cloud systems [3]. Similarly, adaptive  
orchestration mechanisms enable applications to optimize performance-cost trade-offs across heterogeneous  
cloud infrastructures [5].  
Critical Evaluation:  
Despite their advantages, dynamic approaches introduce notable operational challenges. Frequent scaling  
operations may cause performance instability, increased application latency, and management overhead.  
Additionally, reliance on spot or preemptible instances increases vulnerability to unexpected resource  
termination, requiring robust fault-tolerance mechanisms. Provider-specific implementations and configuration  
complexity further hinder the portability of dynamic cost optimization solutions across multi-cloud platforms.  
As a result, while dynamic approaches outperform static methods in adaptability, they often benefit from  
intelligent prediction and automation, highlighting the need for AI and ML-based optimization techniques  
discussed in the following section.  
AI and ML-Based Optimization  
Recent advancements in cloud cost management increasingly leverage artificial intelligence (AI) and machine  
learning (ML) techniques to enable proactive and intelligent cost optimization. Unlike static and dynamic  
approaches that rely on predefined rules or reactive scaling, AI/ML-based approaches learn from historical usage  
patterns and adapt resource provisioning strategies to optimize cost–performance trade-offs in evolving cloud  
environments.  
Examples Include  
Predictive Cost Modeling: where time-series and regression-based models are employed to forecast future  
resource demands and associated costs. Statistical approaches such as ARIMA-based forecasting have been  
applied to estimate short-term cloud usage trends [14], while deep learning models, including Long Short-Term  
Memory (LSTM) networks, demonstrate improved prediction accuracy for complex and nonlinear workload  
patterns [15]. These predictive techniques support proactive budgeting, capacity planning, and early detection  
of potential cost overruns.  
Reinforcement Learning for Resource Allocation: RL-based systems continuously interact with the cloud  
environment and learn optimal provisioning policies by balancing reward functions that incorporate both  
performance metrics and cost objectives. Graph-based and optimization-driven learning frameworks further  
enhance cost modeling by capturing interdependencies between cloud resources and services [2].  
Intelligent Orchestration: AI-driven approaches are particularly effective in multi-cloud and hybrid  
894  
environments, where intelligent orchestration mechanisms dynamically distribute workloads across providers  
www.ijltemas.in  
Page 894  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
based on cost efficiency and performance considerations. Advanced orchestration frameworks enable runtime  
decision-making to migrate or reallocate micro-services in response to changing cost conditions [3].  
FinOps Automation Tools: Platforms like ABACUS [4]: A FinOps Service for Cloud Cost Optimization  
employ automation and AI to monitor, analyze, and enforce financial governance policies.  
Critical Evaluation:  
Despite their high adaptability and potential cost savings, AI and ML-based cost optimization techniques face  
several practical challenges. These systems are heavily dependent on the quality and availability of historical  
data, making them vulnerable to model bias and performance degradation due to workload drift. Additionally,  
the limited interpretability of complex learning models reduces transparency in financial decision-making, which  
can hinder trust and adoption among stakeholders. Integration complexity, increased computational overhead,  
and data privacy concerns further restrict the large-scale deployment of AI-driven solutions. As a result, AI/ML-  
based approaches are most effective when combined with robust governance frameworks and carefully designed  
validation mechanisms.  
Table 1: Comparison of Cloud Cost Management Approaches  
Approach  
Static  
Technique  
Adaptability  
Cost Efficiency  
Complexity  
Low  
Key Limitations  
Budgeting, Tagging, Low  
Scheduling  
Moderate  
Poor response to  
dynamic workloads,  
delayed cost correction  
Auto-scaling, Spot  
Instances,  
High  
High  
high  
Performance  
instability,  
Dynamic  
configuration  
complexity  
Serverless  
Predictive Modelling, Very High  
Reinforcement  
Learning  
Very high  
Very high  
High data dependency,  
limited explain ability,  
integration overhead  
AI/ML-  
based  
Challenges and Limitations  
Despite continuous advancements in cloud cost management techniques, organizations still face significant  
challenges in maintaining effective financial control over cloud resources. These challenges arise from the  
inherent complexity of cloud environments, diverse pricing models, and the limitations of existing static,  
dynamic, and AI-driven approaches.  
Limited Cost Visibility in Multi-Cloud Environments:  
One of the most critical challenges in cloud cost management is achieving unified cost visibility across multi-  
cloud deployments. Static and dynamic approaches often rely on provider-specific billing systems, which differ  
in cost metrics and reporting formats. This fragmentation limits centralized governance and hinders real-time  
identification of cost inefficiencies across platforms [6].  
Complex and Unpredictable Pricing Models:  
Cloud service providers continuously update pricing structures, offering a wide range of instance types and  
usage-based billing options. Static budgeting methods are particularly ineffective under such pricing volatility,  
895  
leading to inaccurate forecasting and budget overruns. Unexpected data egress fees further complicate financial  
www.ijltemas.in  
Page 895  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
planning and frequently remain hidden until billing cycles are completed [9].  
Inefficient Resource Utilization:  
Over-provisioning remains a major contributor to inflated cloud costs. While dynamic approaches reduce idle  
resources through scaling mechanisms, they may cause performance instability when not properly tuned. Manual  
configuration alone is insufficient for optimizing cost-performance trade-offs in highly dynamic and  
microservices-based workloads [3].  
Lack of Standardization and Interoperability:  
The absence of standardized billing APIs and cost management frameworks limits interoperability between  
cloud providers. This constraint restricts the effectiveness of centralized cost optimization tools and increases  
reliance on third-party monitoring solutions, which may introduce additional financial and operational overhead  
[6].  
Data Transfer and Hidden Costs  
One of the most overlooked aspects of cloud billing [7] is data movement between regions or providers.  
Transferring  
data  
across  
cloud  
platforms  
incurs  
egress  
fees,  
which  
can  
reports,  
represent  
leading  
a
to  
significantportionoftotalcloudexpenditure.Thesecostsarenotalwaystransparentinbilling  
misinformed financial decisions.  
Limited Adoption of FinOpsCulture  
While FinOps practices aim to bridge the gap between finance and engineering teams, many organizations  
struggle to implement them effectively. The absence of proper financial governance, accountability frameworks,  
and skilled personnel results in fragmented cost management efforts. As observed in “ABACUS: A FinOps  
Service for Cloud Cost Optimization” [4], automation alone cannot replace the organizational mindset and  
collaboration required for sustainable cost governance.  
Integration, Security, and Data Privacy Concerns:  
AI and ML-based cost optimization techniques depend heavily on large volumes of operational and financial  
data. Integrating these systems introduces security and compliance challenges, especially in regulated  
environments. Inaccurate predictive models or mis-configured automation workflows may unintentionally  
increase operational costs rather than reduce them [7].  
Future Trends and Research Directions  
The evolution of cloud cost management is shifting from reactive cost monitoring toward proactive, intelligent,  
and sustainable optimization [13]. As cloud adoption deepens and workloads diversify, emerging technologies  
and financial practices are expected to redefine how organizations manage cloud expenditures.  
AI-Driven Cost Forecasting and Automation  
Artificial intelligence and machine learning are expected to play a central role in next-generation cloud cost  
management systems. Advanced forecasting models based on time-series analysis and deep learning can enable  
more accurate prediction of resource demand and associated costs [14][15]. Future research may explore  
reinforcement learning-based self-optimizing systems capable of autonomously adjusting provisioning strategies  
in real time while balancing cost, performance, and availability constraints [2].  
FinOps Maturity and Cultural Integration  
TheFinOps (Financial Operations) movement is gaining traction as a collaborative framework connecting  
896  
finance, engineering, and operations teams. Next-generation FinOps platforms are expected to evolve beyond  
www.ijltemas.in  
Page 896  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
simple cost dash boards into autonomous decision systems that enforce budgets, predict cost overruns, and  
recommend corrective actions dynamically. Future studies could focus on establishing FinOps maturity models  
that quantify how effectively an organization manages cloud costs, as well as standardized KPIs to measure cost  
efficiency across industries.  
Sustainable and Green Cloud Cost Management  
Sustainability is emerging as a dual priority alongside cost efficiency. Cloud providers are beginning to publish  
carbon footprint metrics for their services, enabling users to make cost energy-aware deployment choices.  
Research opportunities exist in eco-aware resource scheduling and carbon-efficient workload distribution, which  
aim to minimize both financial cost and environmental impact. Integrating sustainability metrics into cost-  
optimization algorithms can drive the development of greener cloud computing frameworks.  
Cross-Cloud Interoperability and Unified Billing Standards  
As multi-cloud adoption [10] increases, interoperability between providers becomes crucial. Future work should  
aim at establishing standardized billing APIs and cross-provider data- exchange protocols to provide unified cost  
visibility. Open-source initiatives and industry consortiums can play a key role in creating these standards,  
reducing dependency on vendor- specific tools.  
Security-Aware Cost Optimization  
Cost  
optimization  
must  
evolve  
alongside  
security  
and  
compliance  
requirements.  
Future  
researchmayexploremodelsthatjointlyoptimizecost,performance,andsecurityposture.For instance, integrating  
policy-driven cost governance with cloud security management could prevent cost leakage caused by redundant  
or mis-configured resources.  
Integration with Edge and Serverless Architectures  
With the rise of edge computing and serverless frameworks, cost management needs to expand beyond  
centralized cloud environments. Research directions include developing adaptive cost-control mechanisms for  
distributed, latency-sensitive edge nodes and optimizing serverless billing models based on real-timeevent loads.  
CONCLUSION  
Cloud Cost Management has emerged as a critical discipline in ensuring the financial sustainability of cloud-  
based infrastructures. As organizations continue to migrate toward multi-cloud and hybrid environments,  
managing and optimizing cloud expenditure has become increasingly complex. This review paper has analyzed  
the evolution of cost management techniques, ranging from static budgeting approaches to dynamic and AI-  
driven optimization strategies.  
While existing tools and frameworks offer partial solutions, major challenges such as cost visibility,  
unpredictable pricing models, and lack of standardization still persist. The findings from recent studies indicate  
that automation, intelligence, and financial collaboration are essential to overcoming these limitations. The  
integration of FinOps principles with AI-based cost prediction and orchestration represents a promising pathway  
toward intelligent, autonomous cost control systems.  
Looking ahead, the future of Cloud Cost Management lies in achieving a balance between economic efficiency,  
operational performance, and environmental sustainability. Research efforts should continue to focus on  
developing interoperable, adaptive, and energy-efficient cost optimization [12] frameworks that align  
technological innovation with organizational goals.  
Ultimately, effective cloud cost management is not only a matter of reducing expenses but also a strategic  
enabler of innovation, agility, and long-term digital resilience.  
897  
www.ijltemas.in  
Page 897  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XI, November 2025  
REFERENCE  
1. N.  
Ramesh,A.Banerjee,andV.K.Singh,“Auto-scalingApproachesforCloud-Native  
Applications:  
ASurvey and Taxonomy,” Sensors, vol. 24, no. 5, pp. 1–18, Mar. 2024.  
2. T.M. Nguyenand Y. Lee, “Cost Modellingand Optimisation for Cloud:AGraph-Based Approach,”,  
Journal of Cloud Computing, vol. 13, no. 2, pp. 115 132, 2024  
3. P.  
GhoshandL.  
Chen,“CostMinimizationinMulti-CloudSystemswithRuntime  
Microservice  
Re-  
orchestration,” arXiv preprint, arXiv:2403.09121, 2024. [Online].  
4. K.Sharma,D.Raj,andJ.K.Verma,“ABACUS:A FinOps Service for Cloud Cost Optimization,”,arXiv  
preprint, arXiv:2501.06732, 2025.  
5. M.Zhaoand,  
R.Gupta,“AdaptiveOrchestrationforPerformance–CostOptimizationin  
Multi-Cloud,”,  
SSRN Electronic Journal, 2020.  
6. K. Bedi et al., “Unified Cost Visibility in Multi-Cloud,” IEEETrans. Cloud Comput., vol. 11, pp. 89–  
101, 2023.  
7. Singh et al., “Blockchain for Cloud Billing Audit,” IEEEAccess, vol. 12, pp. 7823–7836, 2024.  
8. M. Armbrust et al., “A View of Cloud Computing,” Commun.ACM, vol. 53, no. 4, pp. 50–58, 2010.  
9. E. Walker, “The Real Cost of a CPU Hour,” Computer, vol. 42, no.4, pp. 35–41, 2011.  
10. Khajeh-Hosseini et al., “The Cloud Adoption Toolkit,” Softw.Pract. Exp., vol. 42, no. 4, pp. 447–465,  
2012.  
11. R. Buyya et al., “Market-Oriented Cloud Resource Allocation,”Future Gener. Comput. Syst., vol. 26, pp.  
1012–1023, 2013.  
12. Beloglazov and R. Buyya, “Energy-Efficient Resource Allocation,” Future Gener. Comput. Syst., vol.  
28, pp. 755–768, 2012.  
13. S. Jain and R. Gupta, “Policy-Based Cloud Optimization,” IEEE Cloud Comput., vol. 6, pp. 40–48, 2019.  
14. X. Li and Y. Chen, “ARIMA Forecasting in Cloud Environments,”Future Gener. Comput. Syst., vol.  
107, pp. 509–519, 2020.  
15. J. Zhao et al., “LSTM-Based Cost Forecasting,” IEEE Access, vol. 9, pp. 18123–18135, 2021.  
16. M. Yadav, A. Mishra, Energy-efficient workflow scheduling using dynamic task clustering for  
sustainable  
cloud  
computing.  
Discov  
Computing,  
28,  
article  
id  
201,  
2025.  
https://doi.org/10.1007/s10791-025-09712-0.  
898  
www.ijltemas.in  
Page 898