Accelerate Applications by Offloading File Metadata Operations
Posted in tech
Applications used in data-driven industries can actively access millions of files during normal operations. In Media and Entertainment, (M&E), High Performance Computing (HPC), Oil and Gas, and Financial Services companies, optimizing performance for these applications is critical to business operations.
Not surprisingly, IT teams in these industries spend lots of time and budget to ensure fast performance. Once bandwidth constraints are resolved, a common performance problem is the overhead introduced by the metadata operations associated with accessing and updating all these files. Storage array CPUs can spend significant processing cycles keeping up with all the metadata operations such as managing file permissions and tracking open and close events that might otherwise be used for processing data I/O. In some cases, this can bring both storage and client application performance to a crawl.
The Primary Data DataSphere platform natively accelerates the performance of file-intensive workloads. DataSphere virtualizes data, abstracting logical files from their physical location. This enables the orchestration of data across all storage types, and also offers a secondary benefit for file-intensive applications. By offloading file metadata operations from storage arrays onto DataSphere , the backing storage resources are freed from the job of processing metadata operations. This can significantly speed up data I/O and therefore the processing of applications to boost business productivity. DataSphere also orchestrates parallel access to available storage resources to resolve a common source of bandwidth constraints.
Figure 1. The diagram above illustrates how DataSphere moves the metadata workload off storage arrays.
In addition to accelerating file-intensive applications, DataSphere also ensures data has the performance it needs by automatically placing data on the ideal storage for its requirements, as defined by objectives. This ensures companies in data-driven industries can get the performance they need to accelerate business, stay competitive, and save big. Media and Entertainment can quickly access the files needed to keep post production on task, HPC can scale across all workloads, Oil and Gas companies can quickly crunch through vast amounts of exploratory data, and financial services companies can respond nearly instantly to real-time data, across any number of sources.