Flash has been making its way into data centers primarily as an accelerant to existing workloads. Most of that adoption has been in the form of a hybrid use case, the combination of flash and hard disk drives. As we discuss in the below video with Kaminario’s Ritu Jyoti, all-flash arrays are ready to step up to the full challenge of being the data center’s exclusive storage for primary data.
As we discuss in the video, one of the appeals of all-flash is that these systems provide very consistent performance. There is no concern of a cache or tier miss and data is always delivered very quickly. But not every all-flash array can deliver this new capability of consistent high performance without sacrificing capabilities that the data center already expects from their legacy high performance storage systems.
For all-flash arrays to replace current primary storage offerings they need to provide their performance increase while delivering complete resiliency from failure. After all, if a key component of the array fails, high performance quickly becomes NO performance.
After high availability, organizations are going to demand that their storage system can dynamically scale to meet the growing needs of the organization. Flash storage systems have the potential to deliver enough performance that a single system should be able to meet the demands of the entire enterprise, and doing so could greatly simplify storage management. But in order to achieve this goal the all-flash array needs to scale seamlessly so that additional performance and capacity can be added the same system.
There are other tier 1 storage needs that we discuss in the video like self-management and self-optimization. Most critically, however, all of these capabilities have to be delivered in a fashion where the total cost of ownership of an all-flash array is less expensive than an equivalently capable hard disk based system.
This Video is Sponsored by Kaminario