Real Time Analytics and Data intensive applications have driven the adoption of high performance, low latency, highly parallel NVMe solutions and now we are in the cusp of wide adoption of NVMe over Fabrics (NVMe-oF) Storage in both On-Prem and Cloud Data Centers. Since the introduction of NVMe-oF, a lot of investment and effort have gone in to improve the overall storage performance and features resulting in substantial improvement of IOPs and CPU utilization and lowering the TCO for the customer. NVMe over Fabrics lends itself well to cater to the demands of modern applications on Kubernetes Platforms and operating systems are now able to scale up and saturate the Storage Bandwidth. We will go over various architectures and their benefits/pitfalls. How modern applications are driving requirements for low latency IO. How latest 64G Fibre Channel provides 10+M IOPs with round 10 microsecs latency. What does this mean to the end customer? We will go over the performance benefits from the recent changes with respect to NVMe over Fabrics. Also, will focus on how commonly used applications running in VMs/Containers are driving increase in IO Density and how to get optimum performance