Pursuing Masters in cloud computing domain has given us a lot of insights in various areas of research. Cloud computing involves a lot of virtual infrastructures which needs continuous monitoring and inventory tracking the expenses. In the world of Hybrid and Heterogeneous cloud, on-demand computing always need optimal use of infrastructure which can only be enabled by merging cloud providers and services together onto a single platform and making them interact with each other which gave us motivation to come up with the solution
CloudHop which not only aims cost minimization and also provisioning and management of multi-cloud resources.
What it does
The main purpose of the project is to build a heterogeneous cloud solution for scheduling compute and storage specific operations. This product gives a single intuitive interface to manage and monitor multiple cloud providers through its console. The product is extensible and modular which can be further developed into a more matured cloud monitoring and provisioning system.
How we built it
We used the below tools and technologies to build an intuitive interface over the RESTful services-
- Implemented using python 2.7
- Flask framework for REST APIs
- Apache Lib Cloud for multi-cloud interfacing.
- HTML5 and JQUERY for UI designing.
- Compatible on all browser platforms.
- Can be hosted on Linux, Windows, and MAC platforms.
- And a hell lot of PASSION.
Challenges we ran into
- In order to provide a single interface for multiple cloud providers and services, we had to come up with a common service abstraction layer which made us cautious about the code design.
- Working on variant tools, technologies and services most of the challenges were faced in terms of integrations.
Accomplishments that we're proud of
All the milestones of the project have been met as followed-
- Virtual machine bootup on AWS (Amazon Web Services) EC2 and Google compute Engine.
- Provisioning of storage buckets on both google cloud storage and Amazon S3.
- Unified Inventory Report for Compute and Storage services of AWS and Google Cloud.
- Restful interface for provisioning the instances and scheduling the provision.
- Web-based interface consuming the RESTful API’s.
What we learned
- Team Management- How to divide tasks among the team in terms of product design and development.
- Time Management- 24 hours of rigorous coding gave us a prototype of our future enterprise product.
- Rapid prototyping
- Stress Management
- We can overcome our sleep when we are obsessed with our idea.
What's next for CloudHop
The current prototype gives promising results in the domain of heterogeneous and hybrid cloud solutions. This implementation can be further developed into more scalable enterprise level product having many companies working on multiple cloud providers and their services. Current implementation gives a viable solution for inventory management in distributed cloud infrastructure.