Infrastructure consultant interview questions shared by candidates
The case study questions were most difficult as it was open ended and constrained by only 45 mins. Case Summary: A banking company that holds offices worldwide (Chicago, New York, Toronto and London) has 70% of its employees acting as future traders operating on the Chicago trading floor. Other members of the company are in areas of sales/marketing and engineering, and support trading. As the client has aging storage and application servers, WMP has been given the task to develop a new server infrastructure that minimizes operating risk, while allotting for planned growth. The client is expecting a solution design to be delivered in three weeks that addresses expected costs and time requirements.
Assumptions: The client is currently using 15 servers for local storage in the Chicago office space totaling 10 terabytes. These servers host “images” that traders create and access. There will be a resource demand for basic office use and file sharing across all departments. Traders will require a low latency backbone to remain optimally operable (minimizing risk of missing a trade opportunity). The client is utilizing a smaller shared server space on the Chicago trading floor. There are currently 500 employees with an expected growth rate of 15% per year. Any hardware purchased should have at least a 5 year life cycle with sufficient scalability options, and is at industry standard performance. New servers cost $2k a piece. The market price for Cloud services at the required volume is $1k per year, and allows for enough bandwidth and storage for non trade based requests with the anticipated growth. The cloud provider should be researched and have appropriate layered security. The client is currently using an Active Directory to host objects, but is poorly organized. Storage redundancy and sufficient operation rollback options should be part of the proposed solution. Solution: Connect physical low latency network to purchased/leased Cloud database that is leveraged to dynamically virtualize remaining necessary resources. Old servers could be repurposed as storage if physical office space outside the trading floor allows. In order for the Chicago trading floor to operate at the acceptable latency level, 2 new application servers and one file server with 4 1TB drives in RAID5 should be purchased and installed. The redundancy of the application servers will ensure that the floor always remains operational. The RAID configuration will allow traders to store and access all recent and relevant images. Only images from the last month should be stored locally. Comprehensive records are sent and stored in a location within the cloud database. The additional storage and application resources (not trade based) can be hosted on virtual servers created as a cloud service. These servers should be created and delegated such that proportional resources are given to departments that experience growth. This infrastructure would give a $6k upfront cost, with $1k in annual expenses, and a life cycle of 5 years. While procuring and installing the new servers in the shared space on the Chicago floor, user information can be gathered from the old AD. A new, more organized forest can be created to govern the entities and attributes of the new network. Measures will be taken to match credentials between the physical and virtual environments. Allowing 2 weeks for the forest creation and server initialization, and an additional 2 week testing period for batched beta users of all departments to attempt a soft launch, a target project length of 4 weeks upon receiving of the servers is reasonable for a complete transition. If we add an additional 1 week buffer for possible delays, 5 weeks should be allowed for completion of the infrastructure refresh.
See Interview Questions for Similar Jobs
- Technology Consultant
- Software Engineer
- Senior Consultant
- Software Engineering
- Software Engineer Intern
- Experienced Consultant
- Integration Consultant
- Software Development Engineer