Today, we announced an exciting set of new products and capabilities – our customers will now be able to scale their storage infrastructure completely non-disruptively in granular, low-cost increments in a manner that has traditionally never been possible with either traditional scale-up approaches or more modern scale-out architectures.

It also turns out that it has been just over 2 years since we first publicly launched our company, and I wanted to take this opportunity to look back and look ahead.

Looking back on our founding premise

We were founded in 2008 on the premise of three core beliefs.  First, we believed that flash changes everything in storage – more so than any other technology over the last decade.  Further, while most existing companies were working to leverage flash SSDs to target high performance applications, we believed that a much larger opportunity existed in disrupting mainstream modular storage. While it is easier to design a system that leverages flash as a standalone tier or in an all-flash array, designing an intelligent hybrid storage systems that can flexibly combine the complementary characteristics of flash SSDs and HDDs addresses a much broader set of workloads and cost-points.

Our second core belief was that the traditional approach to data protection, which relied on making separate copies of data on to backup systems, was ready for disruption.  While de-duplication made the backup storage system more cost effective, it did not eliminate the transfer time and cost inefficiencies of moving data over a network for every backup and recovery operation.  We believed that a system that could accommodate thousands of space-efficient and performance-efficient snapshots, coupled with WAN-optimized replication, could dramatically lower backup and recovery times and also minimize the amount of data at risk in the event of an outage.

Our third core belief was that core innovation in storage stems from rethinking data layout and the underlying architecture of the storage system.  When we considered how Data Domain designed an entire file-system that was optimized around de-duplication or how Isilon was designed ground-up for scale-out NAS or several similar examples, we came to believe that the ideal hybrid storage system would require rethinking the architecture and data layout from the ground-up.

We translated these core beliefs into an architecture that we brought to bear nearly two years ago.  Our internal mission as an organization is “To deliver the world’s most efficient data storage solutions to enterprises”.  It reflects the fact that flash is a means to the end; that we are targeting a broad range of mainstream Enterprise applications; that we are simultaneously optimizing for a combination of attributes – performance efficiency, capacity efficiency, data protection and dramatic simplicity; and that we want to be the best when a customer tries to optimize more than one of these criteria.

Two years of living the promise

In the two years that we have been deploying our products, our promise to customers is that we accelerate their applications by delivering 4-5 times better performance and 2-5 times more useable capacity than similarly priced systems; that we help them protect their data better through a combination of thousands of integrated backups and WAN-optimized replication; and that we empower their IT teams by being dramatically simpler to deploy and manage and most importantly to support on a day-to-day basis.

The success with which we have lived this promise is reflected in the rapid and accelerating growth that we have seen in the last two years.  We now have over a 1000 deployments, and our pace of customer adoption has been accelerating as we added over 175 new customers in our most recent quarter alone.  Besides our product value proposition, one of the keys to our customer success has been the investment we made in remote support automation that allows us to predict faults and proactively ensure that our customer systems are running well.  We are proud of the customer community that we have built and are acutely aware of the responsibility we have to nurture this community.

Our partner eco-system has strengthened every quarter with a large base of channel partners, and deep relationships with an increasing number of alliance partners.  We have both deepened our relationship with partners such as VMware, Microsoft, Cisco, Citrix and CommVault as well as broadened our base of partnerships with many more alliance partners that are critical to our success.

Our scope of operations is increasingly becoming global as we have expanded substantially in Europe and have just recently established a presence in Australia.  At the same time, our employee base has grown substantially from ~40 people in January 2011 to nearly 240 people today.  More importantly, we have continued to maintain a high bar on talent as well as on the principle of recruiting people that are pleasant to work with!

We have set a scorching pace over the last two years that has been very gratifying – we have a lot to be proud of, and a lot to be grateful for.

Scale to fit: a giant leap forward

I believe that the solution we are launching today is a giant leap forward in how customers are able to scale their storage infrastructure, and a significant new differentiator for Nimble Storage.

Traditional scale-up systems and more modern scale-out systems are rooted in an era where storage capacity and storage performance have had to be tethered together.  Scale-up, the mainstream approach to scaling in the 1990s, forced you into silos when you maxed out the capabilities of a storage controller.  Scale-out approaches that were brought to bear in the last decade allowed you to start with a small footprint and scale over time but each incremental scaling step came in the form of an additional node that was relatively more expensive and delivered performance and capacity and bandwidth even if you only needed to scale one of those dimensions.

What we are bringing to market today is the ability to scale any dimension, and only that dimension at the lowest possible cost, without ever disrupting applications.  You can start with a Nimble array – if you want more performance, you can pull out the old controller and put in a new controller and gain higher performance from the same storage shelf.  If you want more of your data to be on SSDs, you can replace a lower capacity SSD with a higher capacity SSD.  If you need more storage capacity, you can add disk shelves.  If you need to scale all aspects, you can add another Nimble array and scale-out across multiple arrays.  You can do any or all of these steps – without ever disrupting any application.  We have combined the ability to scale-up performance, scale capacity and scale-out across multiple nodes in our scale-to-fit architecture, in a manner that no other system is able to do.

With the scale-to-fit architecture, we are also introducing a new line of platforms – our CS400 series that delivers even higher performance for extremely demanding workloads, supporting hundreds of virtual machines or thousands of VDI users on a single 3U array.

Explore what we can do for you

We truly believe that storage infrastructure is going through the most significant transformation in over a decade. A large number of customers and partners have increasingly embraced us as we work to deliver on our promise of bringing dramatic efficiencies to our customers’ storage infrastructure.

Please explore the Nimble Storage web site or call us – we would love to engage further and jointly explore how we can make your storage infrastructure more efficient and future-proof your investment.

Share →
Buffer

One Response to Two Years Since We Launched: Looking Back and Looking Ahead

  1. […] at the lowest possible cost, without ever disrupting applications,” Vasudevan said in an Aug. 5 post on the company […]

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>