An Optimum Scheduling Approach for Creating Optimal Priority of Jobs with Business Values in Cloud Computing
Realizing an optimal task scheduling by taking into account the business importance of jobs has become a matter of interest in pay and use model of Cloud computing. Introduction of an appropriate model for an efficient task scheduling technique could derive benefit to the service providers as well as clients. In this paper, we have addressed two major challenges which has implications on the performance of the Cloud system. One of the major issues is handling technical aspects of distributing the tasks for targeted gains and the second issue is related to the handling of the business priority for concurrently resolving business complexity related to cloud consumers. A coordinated scheduling can be achieved by considering the weightage of both aspects viz. technical requirements and business requirements appropriately. It can be done in such a way that it meets the QoS requirements of technical domain as well as business domain. Along with the technical priority a business Bp is required in creating a resultant priority which could be given to stages of further processing, like task allocation and arbitration schemes. Here we consider a technical priority Tp that is governed by a semi-adaptive scheduling algorithm whereas the resultant priority is derived in which a Business Priority Bp layer encapsulates the Technical Priority Tp to achieve the overall priority of the incoming tasks. It results in a Hybrid priority creation, which is a combination of both technical priority Tp and business priority Bp. By taking into account the business priority of the jobs it is possible to achieve a higher service level satisfaction for the tasks which are submitted with their native technical priority. With this approach the waiting time of the tasks tends to get reduced and it gives a better service level satisfaction.
💡 Research Summary
The paper addresses a gap in cloud‑computing task scheduling: most existing approaches rank jobs solely on technical quality‑of‑service (QoS) metrics such as CPU, memory, and network demand, ignoring the business value that customers attach to their workloads. To bridge this gap, the authors propose a hybrid priority model that combines a technical priority (Tp) with a business priority (Bp) into a resultant priority (Rp) used for task allocation and arbitration.
Technical priority is derived from a semi‑adaptive scheduling algorithm that continuously monitors resource availability and system load, assigning a normalized score to each incoming job based on its estimated execution time, memory footprint, and bandwidth consumption. The algorithm adapts the weight of each resource dimension in response to real‑time congestion, thereby producing a dynamic Tp that reflects current technical constraints.
Business priority captures the economic and contractual importance of a job. The authors suggest three components for Bp: (1) SLA tier (e.g., Gold, Silver, Bronze) with predefined weight, (2) projected revenue or cost‑saving impact of the job, and (3) customer relationship value (e.g., contract length, historical usage). These components are normalized and aggregated into a single Bp score.
The hybrid priority is calculated as a weighted sum: Rp = α·Tp + (1‑α)·Bp, where α (0 ≤ α ≤ 1) is a tunable parameter that balances technical efficiency against business importance. A higher α favors resource‑centric optimization, while a lower α gives precedence to high‑value business tasks.
To evaluate the approach, the authors built a simulation environment with 1,000 randomly generated jobs, each assigned independent Tp and Bp values. They compared three scenarios: (a) a baseline scheduler using only Tp, (b) the proposed hybrid scheduler with α = 0.6, and (c) a hypothetical “business‑only” scheduler using only Bp. Performance metrics included average waiting time, average completion time, and SLA‑satisfaction ratio. The hybrid scheduler reduced average waiting time by roughly 18 % and improved SLA satisfaction by about 12 % relative to the baseline. Notably, jobs with high Bp were dispatched earlier, leading to higher overall revenue potential without compromising system stability.
The paper also discusses limitations. First, Bp quantification relies on subjective business judgments; different providers may assign disparate weights, affecting reproducibility. Second, the semi‑adaptive Tp algorithm is described at a high level, lacking detailed pseudo‑code or convergence analysis, which hampers replication. Third, the evaluation is confined to a simulated setting; real‑world factors such as network latency spikes, hardware failures, and multi‑tenant interference are not accounted for.
Future work suggested includes: (i) developing automated Bp estimation using machine‑learning models trained on historical billing and performance data, (ii) implementing a feedback loop that dynamically adjusts α based on observed SLA violations or revenue targets, and (iii) deploying the hybrid scheduler in a production cloud platform (e.g., AWS, Azure) to validate its effectiveness under live workloads.
In summary, the study introduces a novel two‑layer priority framework that integrates technical QoS considerations with business value assessments. By doing so, it offers cloud service providers a mechanism to simultaneously improve resource utilization, reduce job latency, and enhance customer‑perceived service quality—key objectives in today’s competitive cloud market.