Tech Blog and Case Studies

How we helped a multi-billion dollar financial company save money by implementing its own in-house data centers.

Business Challenge

A multi-billion dollar financial company was looking to save millions of dollars by moving out of a hosted datacenter environment and into its own in-house data centers. A huge challenge awaited as the entire operations of the company depended on the success of this project.

 

The major goal was to ensure that the design, implementation and support of this new data center was robust, scalable, secure and flexible enough to handle the entire current and future operations of the company.

 

A quick highlight of other important requirements:

  • First perform an assessment of its entire current infrastructure to understand where they were and how to get to where they needed to be.
  • Come up with a plan and design to implement the new datacenter parallel to the old datacenter.
  • Implement multiple data centers across different geographical regions of the country for load-balancing and redundancy purposes.
  • Come up with a project plan of how to migrate out of the old data center to the new datacenter without greatly impacting business operations.
  • Migrate all the remote locations from their previous WAN connections to a new MPLS connection to the new data center.
  • Everything had to be done by a certain date so as to be able to cancel their previous hosting contract on time to start realizing the cost savings.

Solution:

VeeMost was engaged to perform the overall infrastructure assessment to determine the current state of the technology. Afterwards, working together with the internal IT staff of the company a plan was developed to build out the new data center. Although the client purchased and installed the preliminary equipment through a different vendor, VeeMost was re-engaged to complete the rest of the implementation, integration and migration process.

 

Using Cisco’s line of products, specifically the UCS platform, Nexus, catalyst and ISR products, we built out different modules of the network.

 

Technologies implemented include, but were not limited to the following:

  • Cisco UCS platform used as the unified computing platform. It also converges Ethernet and storage in one place.
  • Nexus 2K/5K/7K , Cisco 4500 used as access, distribution and core layer switches.
  • Cisco 3850 switches used as access switches because of its wireless controller capabilities.
  • OTV to connect the different geographically dispersed datacenters together to extend the server LAN for technologies such as vMotion and data replication, etc.
  • LISP used to control and ensure efficienct inbound routing. (With multiple datacenters having the same IP subnet a way was needed to ensure that the right path to the servers were taken all the time.)
  • MPLS used to connect remote office locations to the datacenters
  • DMVPN used as the secondary connection from the remote locations to the datacenters via the internet.
  • Multi-homing of the internet connections from the different datacenters so that loss of one datacenter’s internet connection does not cause a service outage for the company. Rather, another datacenter will take over and route everything through and from the internet for the organization.

Result:

The company was able to meet their deadline of migrating out of their hosted datacenter and into their own data centers. The design implemented was scalable and robust enough to handle business operations and ensure fast and secure access to services and information.

 

Thanks to the company’s great management and technical team and their relentless belief in technology as a tool that can impact the company’s bottom line, the company was able to save over $45million dollars in fees that were being paid to host their datacenter with another provider. Now they have brought everything in-house.

 

With the success of the project, the company has also retained the services of VeeMost for high-end design and implementation projects while their highly capable in-house team of engineers handle the herculean task of keeping the datacenter running. Our model is to help plan, design, implement and then train the engineers on how to manage the new solution.