Doing More with Less: How MicroStrategy Cut Cloud Costs by 30%

MicroStrategy uses AWS to cut cloud spend by 30 percent, save $250,000 annually, drive visibility into cloud usage and spend, and deliver 50 percent more training environments. The organization provides a leading analytics and mobility software platform to global enterprises. MicroStrategy relied on AWS Cost & Usage Report and services including Amazon EC2 Reserved Instances to optimize AWS usage.

MicroStrategy, the world’s largest independent, publicly traded business intelligence company, provides a leading analytics and mobility software platform. MicroStrategy offers the Enterprise Semantic Graph, a data analysis tool that delivers personalized insights and powerful search capabilities on top of business information systems or data assets. The tool uses AI to enable smart recommendations on authoring actions for analysts who build dashboards, and it provides smart suggestions on content for business users who are looking for new insights.

For several years, MicroStrategy has run its full analytics and mobility platform on Amazon Web Services (AWS), using services such as Amazon Elastic Compute Cloud (Amazon EC2) for on-demand capacity and Amazon Elastic File System (Amazon EFS) for customer file storage. As the company began investing more in the cloud, it sought to control its growing AWS costs. “Our customers were increasingly interested in our AWS-based analytical solution, and we also had internal R&D programs, such as test workloads to validate and test our products, that were leveraging AWS,” says Clayton Myers, vice president of technology for MicroStrategy. “In response, we needed to deploy a top-down initiative to reduce our AWS spend with specific targets in mind.” To support the initiative, MicroStrategy wanted to reduce its overall cloud spend by 20 percent year-over-year and optimize its Amazon EC2 Reserved Instance utilization to lower unnecessary on-demand compute spend.

MicroStrategy designed and implemented an organizational hub-and-spoke model for Cloud Financial Management, with a hub for creating cost and usage visibility and spokes for leveraging information to enable better cloud consumption. The hub, or cloud cost optimization team, included internal cloud and technology experts. “These experts helped evangelize cost management best practices across a heterogeneous set of spokes,” says Myers.

Two examples of MicroStrategy’s spokes are the Technology and Education business units, which are comprised of senior IT analysts, directors, and financial leaders in charge of budgeting and operations. “It was natural to map spokes to business units that were responsible for their own lines of business and had different ways of consuming AWS resources,” says Myers. Each spoke had its own unique business challenges. The Technology business unit, for example, showed more demand for cloud-based offerings, and it sought to become more agile by using more AWS services. The Education business unit showed an increasing demand for education services internally and externally, such as becoming proficient in the latest versions of MicroStrategy products.

Next, MicroStrategy customized its AWS Cost & Usage Report, delivered multiple times a day to an Amazon Simple Storage Service (Amazon S3) bucket of its choice. The AWS Cost & Usage Report contains comprehensive cost and usage data—by product, resource IDs, and cost allocation tags defined by MicroStrategy—and provides the estimated charges associated with MicroStrategy accounts. The company uploaded its AWS Cost & Usage Report from the Amazon S3 bucket to an Amazon Redshift cluster and used MicroStrategy Analytics to create custom internal dashboards and graphs on top of the cluster. The company then built reports from the graphs that provide detailed AWS cost and usage data that can be filtered by a variety of dimensions (e.g., by business unit and weekly, monthly, or yearly trends). “With a better visual experience and access to more detailed business trends, our employees can make better and faster decisions,” says Myers.

The custom cost and usage reports allowed MicroStrategy to create top-down visibility across the organization. The company also set up report notifications via email and messaging platforms. “Using the AWS Cost & Usage Report and our own analytics, we provided more transparency around cloud spend. Our notifications definitely helped create more cost-aware consumption,” says Myers. “For instance, if an environment was unused for two weeks or more, business owners would be asked to back up or terminate the environment. These notifications drove more engagement.”

The organization also enabled cost insight by combining AWS Cost & Usage Report data with other cloud spend information and providing it in a uniform way to all internal stakeholders. “The cost reports we created were a one-stop shop for everyone to see budgets and actual spend,” says Myers.

After evaluating cloud consumption patterns from spoke teams, MicroStrategy introduced targeted optimizations for each spoke. For the Technology spoke, the company scheduled twice-daily shutdowns for all deployments and set up idle instance detection notifications. Instances were monitored for CPU utilization and a warning message was sent to an application owner if instances were idle for an hour. Instances were stopped if they remained idle for two hours.

For the Education spoke, MicroStrategy also organized twice-daily scheduled deployment shutdowns to ensure application environments only ran during 8-hour windows. In addition, the company set default instance expirations of 14 days, and it used Jupyter Notebook automation for bulk API orchestration. This last action was designed to save instructors time and save the company money by reducing application idle time otherwise caused during manual setup and maintenance operations.

Throughout the process of creating and optimizing the hub-and-spoke model, MicroStrategy received assistance from AWS Enterprise Support. “AWS Enterprise Support was a trusted advisor along our entire Cloud Financial Management journey,” says Myers.

By creating a Cloud Financial Management strategy, and by relying on its own powerful analytics tools that leveraged AWS billing and cost and usage reports, MicroStrategy centrally managed cost-cutting initiatives and applied them to each business unit.

As a result, the company saw a 30 percent reduction in cloud spend year-over-year while increasing its ability to provide functionality and cost and usage transparency. Overall, MicroStrategy saved $250,000 annually by using Amazon EC2 Reserved Instances and implementing optimizations identified through increased cost visibility. These optimizations include centrally managed cost-cutting initiatives, scheduling resources, terminating idle instances, default schedules, bulk orchestration using Jupyter Notebook,
and use of Amazon EC2 Spot Instances.

MicroStrategy has used its Cloud Financial Management strategy to help its Technology and Education organizations accomplish more while using fewer cloud resources. For example, the Technology organization has been able to increase its AWS use to support new benchmarking and testing initiatives, but at a lower overall consumption cost. “Every quarter, budgets have gone down, and the cost management initiatives we implemented made that happen,” says Myers.

MicroStrategy presents a large annual event called MicroStrategy World, which offers training, marketing, and networking with customers. “Because of the way we implemented Cloud Financial Management internally, our total education costs only increased by 10 percent, but our Education organization was able to deliver 50 percent more training environments during the event,” says Myers. “The bottom line is we can do more with less and stay within allocated budgets.”

To learn more, visit the AWS Cost Management page.