Background
Our client, a major player in the hospitality industry, utilizes a performance monitoring tool (PMI) to track key operational metrics, manage guest services, and optimize hotel workflows. To support various automated tasks, they previously relied on Azure Virtual Machines (VMs) running Windows Task Scheduler to manage job execution. However, as their business grew and demand increased, this legacy approach became inefficient, particularly in terms of scalability, flexibility, and high availability.
The existing infrastructure faced several challenges:
- Limited ability to scale jobs based on demand.
- Inflexibility in handling increasing workloads.
- Issues with outbound port access, crucial for running SFTP tasks in their environment.
Challenges
The previous solution, involving Azure VMs and Windows Task Scheduler, was not suitable for handling the growing number of jobs and the dynamic nature of the hospitality industry.
Additionally, the client faced the following issues:
- High Operational Costs: The virtual machine infrastructure was expensive and lacked the flexibility needed for scaling.
- Inefficiency in Job Execution: The reliance on Task Scheduler and manual intervention led to delays in processing large job queues.
- Lack of Fault Tolerance: The existing solution did not handle large spikes in job load well, leading to performance degradation.
Solution 1:
Scalable SFTP Server Solution
The client was using an SFTP server hosted on a virtual machine to transfer files. However, Azure App Services and Azure Functions do not provide outbound ports required for running an SFTP environment.
Our Approach:
We suggested moving the SFTP server to a more scalable solution using Azure Cloud Services with extended support. This allowed the client to maintain a fully functional, secure, and high-performance SFTP environment while benefiting from Azure’s cloud capabilities, such as scalability and high availability.
Solution 2:
Flexible Queue Job Processing with Hangfire.io
The client managed a large number of processing queue jobs related to guest services, reservations, and other hospitality operations. The legacy system was not well-suited to handle the increasing job load efficiently.
Our Approach:
We migrated the queue jobs to Hangfire.io running as an App Service on Azure. Hangfire.io provides a scalable, flexible solution for managing background jobs in .NET. By leveraging Azure App Services with auto-scaling capabilities, we were able to ensure that the client’s queue processing could scale dynamically based on load.
Solution 3:
Azure Storage Queue for Job Execution
For another type of processing queue, we implemented a Storage Queue Solution on Azure. This solution leverages Azure’s scalable Storage Queue to pick up jobs and execute them in a fault-tolerant and highly available manner.
Our Approach:
We used Azure Storage Queues for job management, enabling automated execution without manual intervention. Azure Storage Queues are highly reliable, and jobs can be processed asynchronously, ensuring high throughput even during peak loads.
Results
This migration from legacy Windows Task Scheduler to Azure-based cloud solutions not only addressed scalability and flexibility issues but also provided a more cost-effective and reliable infrastructure for our client in the hospitality domain. By leveraging Azure’s PaaS offerings, we were able to create a highly scalable, fault-tolerant solution capable of handling varying workloads, all while reducing operational costs. This case demonstrates how cloud migration, when approached strategically, can transform legacy systems into modern, high-performance environments suited for dynamic industries like hospitality.