In a recent customer profile “All-Flash in Real Life: CMA – XtremIO Database Consolidation Case Study” Storage Switzerland did an extensive interview with CMA, a leading healthcare application service provider for Medicaid payments, systems processing, and data analytics. During that call we discussed how their integration of EMC XtremIO led to a transformation of their workloads. One of the interesting side benefits of their move was how all-flash tore down the silos within IT.
Types of Storage Silos
In data centers we see two types of silos that relate to storage. The first is the storage system itself. It’s not uncommon for companies to have separate storage systems for each database environment, for the virtual server environment, for the virtual desktop environment and for the file sharing environment. Each storage system is tuned for the specific task at hand in order to extract maximum performance at minimal cost.
The other type of storage silo is the “knowledge silo” that gets created due to the differences in skill sets and expertise between the storage team and other teams within IT. For example, in CMA’s case application and database administrators lacked the specialized knowledge to configure traditional storage arrays. As a result they could not fully collaborate in the storage configuration process. Due to this knowledge barrier, CMA storage engineers, who were not application experts, often had to make critical storage design decisions without direct input from the people that developed the application.
All-Flash Eliminates Storage Silos
CMA is an excellent example of one of the side effects of going all-flash, both storage silos are eliminated. The first type, the task-specific storage system, is eliminated by the performance capabilities of an all-flash system. For most environments all-flash provides more performance then they will ever need. In fact, it provides more performance than all the environments within the data center combined will need. Most all-flash systems also provide deduplication and compression so that flash capacity can be optimized, driving down the price per GB.
The second silo, the skill set division between storage teams and other IT teams, is also eliminated by all-flash. Thanks again to an all-flash system’s performance, special “storage master tricks” are no longer needed. Standard configuration of the system is typically more than appropriate for most if not all workloads. This is not to say that a storage specialist isn’t needed, they are, but their focus can now be on infrastructure design instead of application optimization.
Most all-flash arrays are purchased to solve a very specific performance problem. In most cases they accomplish that mission, but can also create many side benefits that the customer never expected; the elimination of storage silos being just one. Another is a massive simplification of the storage architecture. You can learn about these and other side-benefits of all-flash arrays in our CMA case study. Fill out the simple registration form below for an exclusive copy of the complete report.