Cisco Blogs
Share

Three Ways to Future-Proof Your Storage Networks for the Next Decade through an All-Flash Data Center


September 7, 2018 - 0 Comments

When Cisco and Hitachi first began working together in the storage networking industry, the total known size of the digital universe was roughly 1.5 exabytes. Today due to the hyper-growth of users, devices, and data, the world generates roughly 2.5 exabytes a day.

It is no wonder why enterprises are struggling to keep pace as more and more people, devices, and applications are coming online. Further, with the advent of artificial intelligence and machine learning, there will be billions if not hundreds of billions of autonomous devices operating everything generating massive amounts of data that have to be collected, transported, and analyzed on a real-time basis.

As a result, the storage and storage networking infrastructure that host and power/transport/ protect the key applications also need to evolve to keep up with the increasing demand. Modern applications are data-driven and require more complex processing than past applications and generate a large amount of data.

Thankfully, storage technologies such as flash and NVMe have revolutionized key performance factors and have a huge impact on how future storage infrastructure is designed and operated. Let us examine some of the key challenges that customers face and how Cisco solutions help to alleviate some of these challenges of building and future proofing your storage infrastructure.

Three Key Challenges and Solutions

The Present – Migrating from non-flash to all Flash arrays: In an IDC user demand Survey 2017, Customers were asked what their top three criteria for choosing all-flash arrays with two of the three criteria were about performance and performance consistency. The fact is that the current infrastructure is not adequate enough to support the high-performance requirements of Flash.

If the infrastructure is not carefully planned and designed for new apps in mind, you might end up with a speed mismatch, severe congestion, latency or degrading performance. Modernizing the SAN infrastructure with Cisco MDS 32G solution will give you the required performance, deep visibility and a hardware-based solution for congestion detection and avoidance.

Near-Term Future – All Flash Arrays Enterprises: As the SAN infrastructure grows and as more hosts and targets are added to support new app requirements, troubleshooting and maintenance will become more complex.  In order to provide continuous fabric availability, deep visibility and enhance telemetry is required. Modernizing the SAN infrastructure with MDS 32G will provide you with integrated analytics for deep visibility. This includes the ability to monitor all flows on all ports to empower storage admins with the ability to make informed and proactive operational decisions.

The new MDS Diagnostic Suite provides users with enhanced visibility at the edges as well as the core of their SAN fabric. This is critical to guarantee continuous fabric availability for a diverse set of apps and workloads in the modern enterprise. Fabric-wide visibility is achieved through new features such as Multipath Trace, Link Cable Beacon and Read Diagnostic Parameter designed to be used for troubleshooting at the fabric edge.

Long-Term Future – NVMe based high-performance Arrays and NVMe over Fabric: NVM Express (NVMe) is an optimized, high-performance, scalable interface designed to work with current and the next-generation NVM technologies. The NVMe interface is defined to enable host software to communicate with nonvolatile memory over PCI Express (PCIe).  Earlier technologies (such as SAS and SATA) that were architected for hard disk drives are unable to take full advantage of SSDs’ potential. This warranted the need for a new architecture: Non-Volatile Memory Express or NVMe for short

The Flash with NVMe can sustain full line rate and hence can meet the high performance that is expected by the application. Benefits of NVMe can be extended to centralized storage infrastructure over a SAN built using Cisco MDS switches. Fibre Channel is the preferred protocol for connecting all-flash arrays in today’s data centers due to its performance, availability, scalability, and plug-and-play architecture. FC-NVMe offers the best of Fibre Channel and NVMe. You get the improved performance of NVMe along with the flexibility and the scalability of the shared storage architecture. Check out Cisco MDS solution for All Flash and NVMe over Fabric.

For More Information

These topics and other ways to modernize your SAN infrastructure will be discussed during a session “Realizing the full potential of your Digital Transformation” that Paresh Gupta (Sr. Technical Marketing Engineer) will be giving during the Hitachi Next 2018 in San Diego, California .

Tony Antony
Sr.Marketing Manager

 



In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.