• Panchashree
• Athmaranjan K
• Manisha

Cloud computing has become an essential platform for running scientific workflows because of its flexibility and cost-efficiency. Scientific Cloud Service Providers (SCSPs) serve as intermediaries by renting virtual machines (VMs) from Infrastructure-as-a-Service (IaaS) providers to fulfil workflow execution requirements. These providers generate profit when workflows are completed within their specified deadlines. However, two major challenges affect this profitability: the cold start problem and the complexity of managing various VM pricing models such as reserved, on-demand, and spot instances. This paper introduces a hybrid scheduling framework that combines initial resource planning using historical workload data with real-time adjustments based on current workload changes. In the planning phase, the system allocates VMs using reserved and spot instances according to predicted demands. During execution, it dynamically scales by adding more VMs through on-demand or spot instances when there are sudden increases in tasks. Additionally, the framework uses a dependency-aware scheduling method that considers both cold start delays and the fluctuation in spot instance prices. Experimental results using real-world benchmark datasets show that this approach performs better than existing methods, offering up to 20% improvement over cold-start-based techniques and 15% better efficiency than methods focusing only on pricing model optimization.

scientific workflow, cold start, pricing model, spot instances, VM renting

"SCIENTIFIC WORKFLOW SCHEDULING IN CLOUD CONSIDERING COLD START AND VARIABLE PRICING MODEL", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.123-126, July-2025, Available :http://www.jetir.org/papers/JETIRGX06023.pdf

"SCIENTIFIC WORKFLOW SCHEDULING IN CLOUD CONSIDERING COLD START AND VARIABLE PRICING MODEL", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. pp123-126, July-2025, Available at : http://www.jetir.org/papers/JETIRGX06023.pdf