Switch from Hyperconverged to Tintri and Cisco UCS Results in Significant Cost Savings, Faster Deployments, and Higher Performance for Enterprise Applications
Baxter Spins off Baxalta in 2015
Baxalta is a leading global biopharmaceutical company based in Bannockburn, Illinois. Baxalta was formed through a spin-off from healthcare giant Baxter International in July of 2015. As soon as the split was finalized, Baxalta needed to vacate Baxter's data centers as quickly as possible to avoid paying millions of dollars for the continued use of Baxter’s IT infrastructure.
Baxalta needed to find an IT solution that could be deployed rapidly in order to avoid excessive costs. After a review of their infrastructure options, Baxalta chose Tintri due to its ease of deployment and management. “Our previous infrastructure consisted of a mix of storage solutions, including IBM storage, EMC, and NetApp systems,” explained Mike Baker, Data Center Architect. “It used to take over a month to get those storage systems off the ground and deployed, even before we started migrating workloads over to a new environment. We were able to fully deploy Tintri, the Cisco UCS servers, and have VMware fully installed with applications running at our new data centers in less than a week, with complete confidence that everything was set up correctly.”
Why did Baxalta Choose Tintri?
In addition to the much faster deployments, Baxalta was also impressed by Tintri’s enterprise cloud features. “Tintri’s autonomous QoS was a big selling point for us,” Baker noted. “Tintri ensures that all of our applications perform predictably as we scale. We were also amazed at the VM-level visibility we get with Tintri Global Center, enabling us to pinpoint exactly what’s going on with any individual application at any point in time.”
Shire Plc Acquires Baxalta in 2016
After being spun off by Baxter in 2015, Baxalta was acquired by Shire Plc. in January of 2016. Shire is a leading global biotechnology company focused on serving people with rare diseases and other highly specialized conditions in more than 100 countries.
Shire had already standardized on hyperconverged for all of its data centers in the United States. In addition to the US deployment, the pharmaceutical company had already made plans to double its hyperconverged footprint, building a second IT environment using the same hyperconverged systems in its European data center.
Standardizing on Either Hyperconverged or Tintri
Baxalta was now faced with an additional infrastructure integration challenge. They had to decide whether to continue expanding their very successful Tintri and Cisco UCS deployment, or switch over to Shire’s then corporate-standard hyperconverged platform. As the IT teams worked together to develop the technology roadmap, discussions centered on the performance issues Shire was experiencing with its hyperconverged infrastructure. “It was always difficult to manage our databases with hyperconverged,” shared Baker. “We had to split our database servers into lots of smaller disks, and then tie them back together just to get the performance we needed out of those databases. It was a very inefficient and time-consuming process.”
Baxalta’s IT team had already proven how quickly they could deploy the Tintri and Cisco UCS pods after the Baxter spin off, and how consistently they performed in their new environment. So instead of going forward with the planned hyperconverged purchase, Baxalta convinced Shire to standardize on Tintri storage systems and Cisco UCS servers. Shire was able to purchase all of the Tintri and Cisco infrastructure they needed for both the US and Europe data centers for the same budget they had set aside for just buying additional hyperconverged systems for the European expansion project.
Reduced IT Management Time
“We also invested in Cisco UCS Director. With Tintri’s plug-in to UCS Director, we are now able to manage compute, network and storage from one central location—and see the performance of every application across our entire infrastructure,” noted Baker. Previously, Shire had to depend on managed services to maintain its infrastructure. And since it took more than one month to stand up a new environment, those services contracts were substantial. “With Tintri and Cisco UCS we were able to terminate our managed services contracts—now it takes just one person a quarter of their time to manage a footprint that spans multiple data centers. Growing our footprint without adding management burden is a requirement for our enterprise cloud.”
Obtaining an Easily Scalable Architecture
The move to Tintri and Cisco UCS also resolved Shire’s concerns about scaling their hyperconverged footprint. “With our hyperconverged infrastructure, we had to scale compute and storage together, even when we only needed more of one resource and not the other—resulting in excess cost. Since Tintri and Cisco UCS can be scaled independently, we no longer have to buy everything up front,” said Baker. “We can grow incrementally as needed and deploy our new Tintri and Cisco UCS architecture in small and medium-sized pods. When we need more compute, we just put in more UCS blades. If we need more storage, we can add more Tintri VMstore systems. Together, Tintri and Cisco are providing us with all of the flexibility and agility of an enterprise cloud in our on-premises infrastructure.”
Illinois, United States
Cisco UCS Servers
Cisco UCS Director
IBM, EMC, NetApp and hyperconverged infrastructure
Existing storage systems were not easy to deploy or manage. Performance was unpredictable. Scaling was non-linear and so hyperconverged proved both complex and expensive
Tintri VMstore systems
Tintri Global Center
Primary Use Case
All enterprise workloads across three global data centers
Reduced storage deployment time from over a month to less than one week
Simplifed ongoing storage and server management
Reduced datacenter footprint and power
Obtained the exibility and agility of an enterprise cloud for on-premise infrastructure