As the great adage says, if you fail to plan, you’re planning to fail – and when you’re making strides in uncharted territory, it can be difficult to develop a winning strategy off the bat. Recently, Lyft learned that lesson in a very expensive way. They recently released documents for their IPO filing showing that they were contractually bound to spend $300 million with Amazon’s Web Services for their cloud storage over the next few years.
While it’s very common for growing companies to rely on cloud and supercomputing services from larger data centers, it can become expensive and cumbersome to move away from third-party servers once your data has reached a certain capacity.
As ADAS and autonomous vehicle research grows, it’s only a matter of time before data generation begins to surge, which is why it’s imperative that a data ingest, storage, and archive strategy be in-place from the beginning. Autonomous vehicle research isn’t slowing down, either. One vehicle driving and collecting data collects between 2 and 20 terabytes of data per hours. When you’re using a fleet of vehicles, that data grows to petabytes rather quickly, and if the cloud is your only method of storage it’s going to get expensive and even harder to control.
Dealing with and planning for data surges is key to a scalable operation. Quantum’s StorNext File System Meta Data Controller provides the flexibility for any organization, at any point in their research. If you’re just starting your R&D and your data is still manageable, StorNext acts as the gateway to any storage tier that makes the most sense to support your strategy and strategies can change as data scales and performance requirements grow. Plus, with the Quantum R-Series, your fleets can collect data with a ruggedized chassis and removable magazines, and go through a fully automated ingest with StorNext, and set policy to automatically bring static data to an archive tier.
Moral of the story? Don’t get caught without a strategy, and always have a plan and remain flexible. With Quantum, options are limitless. Use us for end-to-end storage, or just for the file system to automate the data so you can focus on the research that makes a difference.