The introduction of the non-volatile memory express (NVMe) storage protocol brings with it a new storage performance tier. Whereas older Small Computer System Interface (SCSI) and Advanced Technology Attachment (ATA) standards were designed for slower-performing hard disk drive (HDDs) and tape media, NVMe taps more into the raw performance potential of faster solid-state disk (SSD) media through reduced latency and higher input/output operations per second (IOPS).
Storage planners are now faced with challenging decisions regarding which workloads to migrate to NVMe and when, to ensure that they are maximizing their return on investment (ROI) in more expensive NVMe media. NVMe has more obvious advantages for workloads such as high-velocity analytics and artificial intelligence (AI) that require maximum levels of throughput and processing power. However, it also stands to benefit more traditional applications such as Microsoft SQL Server and Oracle databases, that remain mission critical and that in fact still comprise the majority of the workload ecosystem for most businesses today. Storage Switzerland recently provided more in-depth discussion of this topic in the recent blog, “Does Your Application Really Need NVMe Performance?”.
The need for workload testing is clear, but creating a test environment of premium-priced NVMe systems is prohibitively expensive for most enterprises. Additionally, obtaining a complete picture of ROI is complex and includes a range of granular error conditions and triggers. With workload simulation, IT can dramatically reduce the cost and the complexity of determining how and when to introduce NVMe into the data center.
With workload simulation, storage planners create profiles of production workloads and detailed emulations of specific storage infrastructure. They can then apply models of real-world traffic loads and other scenarios to test for performance, scalability and other limits under a multitude of complex scenarios. This helps to determine the optimal fit from a storage technology and configuration perspective for specific workloads.
When it comes to simulating workloads for NVMe systems, IT planners should look for a solution that is capable of handling demanding IOPS loads. To maximize simplicity of the process and of collaboration on technology and workload needs, they should also look for an output and a user interface that balance ease of use, with the flexibility to customize for the data and reporting that are most relevant.
For more on how to build an effective storage performance validation practice that is built on workload simulation, join experts from Storage Switzerland, Virtual Instruments and SANBlaze during our on demand webinar.
Senior Analyst, Krista Macomber produces analyst commentary and contributes to a range of client deliverables including white papers, webinars and videos for Storage Switzerland. She has a decade of experience covering all things storage, data center and cloud infrastructure, including: technology and vendor portfolio developments; customer buying behavior trends; and vendor ecosystems, go-to-market positioning, and business models. Her previous experience includes leading the IT infrastructure practice of analyst firm Technology Business Research, and leading market intelligence initiatives for media company TechTarget.
StorageSwiss.com - The Home of Storage Switzerland 