Friday, April 19, 2024

MicroStrategy Reaps Virtualization’s Benefits with Software-Defined Data Center

Datamation content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

By one estimate, only about 5% of Fortune 500 companies are moving toward virtualizing their entire data centers, which makes MicroStrategy’s nearly two-year head start all the more remarkable.

The enterprise software company has been building out its software-defined data center since November 2011, and the result has been some remarkable efficiencies gleaned from running its private cloud in less than 30 square feet of data center floor space.

Alex Freixas, vice president for Worldwide Information Systems at MicroStrategy, discussed the company’s virtualization efforts in an interview with Datamation. The company has achieved a number of efficiencies thanks to its VMware vCloud implementation, including shorter sales cycles, improved customer service, and significant time savings for engineers.

MicroStrategy has been a VMware customer for nearly a decade, so when the company ran into system scalability challenges in 2010, VMware seemed like a logical partner to take its data center to the next level. After a successful beta test and reference architecture for virtualizing network and compute resources, and a successful private cloud rollout for its R&D organization, the company began to look for other use cases for its virtualized data center.

The next stop was technical support. With the new virtual data center’s ability to mimic any operating system and database configuration, MicroStrategy can now replicate customer configurations in an average of 11 minutes, down from an hour previously. For customers who are trying to crunch data quickly, that’s a big time savings, and MicroStrategy customer service reps can help customers make decisions while they’re on the phone.

Next up was the Sales department. By opening its private cloud to customers, MicroStrategy can provision the infrastructure required for a customer proof of concept demonstration in 45 minutes. Left to their own devices, customers might take days or weeks to procure the hardware for a proof of concept demonstration.

MicroStrategy looked at automating classroom setup for training and education programs next. IT personnel had to travel to configure a classroom site, and it took several days to set one up. The company can now set up a classroom in two hours via the cloud, with no IT travel.

MicroStrategy integrates 10 or more branches of its source code every night, including every change made that day, and engineers must install and configure the entire code base the next morning. The setup process used to take engineers 90 minutes at the start of the day; now they can go to a catalogue and deploy the version they need with just five clicks, with an additional 10-15 minutes of machine automated configuration time. The new streamlined process saves the company’s engineers thousands of hours a year.

Finally, MicroStrategy took everything it had learned and rolled it out to customers in the form of a cloud business intelligence offering. That proved to be another benefit for sales folks, because MicroStrategy can get a customer up and running within 48 hours of signing a contract. And it also means new customers who didn’t have the infrastructure or budget to support an ambitious on-site business intelligence implementation.

Superior uptime

MicroStrategy has achieved a number of efficiencies and benefits thanks to its embrace of software-defined data centers, but not all the credit goes to VMware. The company has achieved 100% uptime since the initial November 2011 private cloud launch despite major hardware and software upgrades, which Freixas said “takes a great deal of discipline.”

For high availability, the company designed the system for resilience “from the ground up” and made sure that there were no single points of failure. Failover capabilities on every layer were carefully designed, implemented and tested, and are periodically retested. A very strict change management process also helps, as does training: Staffers are professionally trained and certified on every component of the stack, including networking, storage, VMware, compute and operating systems.

MicroStrategy has also done a lot of implementation work, with dedicated automation groups looking for further efficiencies. “We have put a great deal of engineering effort into automation,” said Freixas.

Freixas said his staff “did not run into major obstacles to implement automation, but what was key was certainly to get our end users involved in every step of the process, so the end result was not what we did to people, but rather what we did for people.”

Coincidentally, at the same time Freixas and his staff were working on automating end-user experience for provisioning machines, the company was also working on a new product offering for commercial customers, System Manager, which is a process automation orchestrator that was enhanced to programmatically support creating VMs and vApps through VMware APIs. “That was key to our end-to-end automation vision,” said Freixas.

Changing IT roles

MicroStrategy’s transformation to a cloud company – both internally and customer-facing – has meant big changes in IT roles too. Everyone is a customer now, and the goal is to meet or exceed service-level agreements (SLAs).

It has also meant letting go of a lot of traditional IT tasks.

“People now get that what a machine can do for you should be done for you,” said Freixas. “Every hour spent running the system is time lost that could have been spent developing solutions.”

And the company has shared that time savings with employees, says Freixas. “They have a life outside of work,” he said.

Freixas expects IT departments’ view of the cloud to change dramatically in the next couple of years. “In 2015, the conversation is going to be different,” he predicts. “It’s going to be, ‘Why not in the cloud?'”

He likens the coming corporate acceptance of the cloud to what has happened with virtualization in recent years. “Six years ago, there was a lot of doubt about virtualizing production apps,” he notes.

The cloud will let companies try new projects – and easily and cheaply pull the plug if they don’t live up to expectations, Freixas says.

Paul Shread is editor in chief of the IT Business Edge network.

Subscribe to Data Insider

Learn the latest news and best practices about data science, big data analytics, artificial intelligence, data security, and more.

Similar articles

Get the Free Newsletter!

Subscribe to Data Insider for top news, trends & analysis

Latest Articles