Case Studies & Clients

Case Studies


C9 successfully helps their clients to strategize, adopt, implement, monitor, manage, optimize, govern and secure the cloud.

Extreme Cost Reduction

Enterprise: A large online trading company that has over one billion hits daily.

Problem: The company has hundreds of servers in elastic clusters configured for extreme scaling up/down. They were using On-Demand Instances within their clusters. On-Demand instance are the most expensive types of instances clients can purchase from AWS. Even though the client was scaling up and down the cost of doing was very high due to the need for a large amount of On-Demand instances for long periods with the day.

Solution: The clusters were re-designed using Reserved (RI), Spot and On-Demand instances. It was determined that only 12% of the instances needed to be always ON. Those 12% instance very converted into RI’s. About 75% instance were converted in Spot Instances and the rest were 13% were On-Demand Instances.

Note: RI’s can save you over 50% when compared to On-Demand instances and Spot Instances can save over 80%+ when compared to On-Demand instances

Outcome: Extreme cost reduction with 99.999% uptime and it took us less than 1 day to set up.

Instance Right-sizing Resulting In Cost Savings

Enterprise: Large benefits and financial company has over 3000 customized retirement plans from individuals and corporations throughout the United States.

Problem: Applications are not well matched with underlying instances for best price and performance. The Cloud infrastructure was sized based on on-premise sizing and no performance base-lining was done prior to lift and shift to the Cloud

Solution: C9 ran special tools from our partner and generated a baseline application profile. The profile was then used to generate synthetic loads. The synthetic loads were run on various instance types to determine the best instance type and size to run the application. This tool was used to support the “Continuous Optimization” process.

Note: In an Opex environment, “Continuous Optimization” is a must as you are paying by the hour, therefore you have ensure that the environment is continuously optimized as drift will occur due to multiple reasons like traffic pattern changes, application enhancements etc.

Outcome: A proof of concept was conducted for 23 systems. With the right tools and processes, the customer realized 25% reduction in annual cost. This was then implemented across the production environment

High Reliability without Increasing Costs

Enterprise: Large retailer who was using AWS for specialized data-mining services.

Problem: The client was using Spot instances to keep costs down. Spot instance can be taken way at any time with a 2 minutes notice. Reliability suffered due to manual management of Spot instances.

Solution: Used tools to automate the management of Spot instances. The tool fully automated the process of bidding for spot instances, start and stop instance and add spot instance to the cluster as required. The cluster were re-configured to use a small percentages of RI’s to ensure always availability of minimum resources.

Outcome: Achieved 99.99 availability and removed the need for human intervention thus reduce operational costs while improving reliability.

Improve IT Operations – Availability, Reliability, and Scalability

Enterprise: Major car auction provider

Problem:
• Unacceptable Mean Time to Repair (MTTR), Mean Time between Failures (MTBF), and Mean Time to Recovery (MTTR).
• Multiple monitoring and altering tools that were not integrated
• Unhappy customers
• No definition of Tier I,II and III support

Solution: Used Vistara IT operations tools to create a single pane from which the Hybrid Cloud environment could be monitored and managed and out sources Tier I and Tier II IT Operations

Outcome:
• MTTR and MTBF that met defined Service Level Agreements (SLA’s)
• Single tool that can monitor and manage the Hybrid Cloud
• Improved customer services

Implement DevOps to improve speed to market and quality of end products

Enterprise:Social game development company that develops games for Facebook

Problem:
• No formal DevOps team and process
• Minimal collaboration between IT and development team
• Infrastructure management was not automated
• Very rudimentary monitoring for their Hybrid Cloud
• Social games space is highly competitive. All the above factors impacted the speed to market and this hurt the client’s bottom line.

Solution:
Implemented holistic and agile DevOps methodology and capability
• Release Management and Continuos Delivery
• Change Management
• Unified Source Code Management
• Infrastructure-as-Code
Automation platform, Services, and Continual Improvement
• Shared services operations including a dedicated build and release team with expert services architects
• End to end monitoring includes cloud gaming and critical infrastructure components
• Single pane of glass dashboards and reports for infrastructure includes servers and VMs
• Continual improvement via management of performance availability of applications and infrastructure, and capacity mgmt.
• Chef, Puppet, and Jenkins integration
Operating Model support
• Integrated development framework
• Expectations setting and training
• Role clarity between DevOps and TechOps
• Organizational change management

Outcome:
• Speed to market.

Implemented "Continuous Optimization" using CloudQoS

Enterprise: Largest Software company in the world.

Problem: The client was planning to move their SharePoint farms into Azure. They did lift and shift and did a Proof-of-concept and noticed they did not see the cost saving they expected. They were using Azure PaaS platform.

Note: In an Opex environment, “Continuous Optimization” is a must as you are paying by the hour, therefore you have ensure that the environment is continuously optimized as drift will occur due to multiple reasons like traffic pattern changes, application enhancements etc.

What Solution: Ran CloudQoS tools and performed a sizing exercise to ensure that infrastructure was sized correctly based on performance requirements. CloudQoS tools were used to determine that 4-core servers provided sufficient resources to run SharePoint.

Outcome: Provided savings of 37% Net Present Value over five years switching from 8-core servers to 4-core servers. CloudQoS tools were used to implement “Continuous Optimization” process to ensure the environment was always in an optimized state.

Some Key Clients