The Nimble Scale

Last week I was in San Jose for Storage Field Day 2. On the first day of this event we visited a Nimble Storage to get a great presentation about their products. With their Cache Accelerated Sequential Layout (CASL) architecture the Nimble Storage arrays are able to Scale-up as well as Scale-Out, or as Nimble likes to call it Scale-to-fit.

This video from Storage Field Day 2 gives you all the information about the company and their products:

The CASL design

The CASL architecture is designed from the ground up with the advanced technologies (CPU, Storage and Networking) that came available in the last years. This makes that the Nimble Storage Arrays provide great performance and flexibility to the customer. The CASL design is build by technicians that gained their knowledge at companies like NetApp and DataDomain. This is also why many of the best of the technologies from different vendors come together in the Nimble Storage Arrays. Nimble Storage says their product is storage, Disaster Recovery (DR) and backup in on solution.

In the CASL design writes comes into the array in the and lands on NVRAM. There the data will acknowledged and send to DRAM. Within DRAM the data goes through CRC checksum and is Lempel-Ziv (LZ) compressed (2-4X). After this the array writes sequential stripes to the High-capacity disk and the flash layer based on data characteristics. The flash layer is intended for a 100% read cache (effectively this means that 96% is read from Flash). CASL tracks all data hot data blocks inline, providing accelerated reads and sub-millisecond latency. Because all the writes are committed the flash dish act as a JBOD.

More information about CASL and the Nimble Storage architecture can be found in this second video:

The perfect fit

Nimble Storage makes it possible to Scale storage performance as well as capacity. This makes the Nimble products extremely flexible and economical. By providing Scale-Up as well as Scale-Out capabilities Nimble Storage can easily address all their customers needs. Combining different Nimble Storage Arrays (CS200 and CS400 series) is possible as well as replacing your storage controllers without disruption to add more compute power. Last but not least to you can replace flash and disk drives to upgrade one of these two layers.

Dynamic Storage Pools

Nimble Storage provides the possibility to create storage clusters. A cluster can be logically divided into storage pools. A storage pool consists of one or more arrays over which data is striped. The opportunity to migrate volumes between different pools in a cluster non-disruptively eliminates downtime in case of maintenance or other cases in which an array should be taken offline. By automatically rebalancing data across pools Nimble gives customers the chance to easily expand, shrink or merge pools.

Replication efficiency

Nimble is able through scheduled replicating compressed changed data to a secondary (off-site) system to reduce bandwidth costs for WAN replication and to deploy a DR solution that is low-cost and easy to manage.

An easy way to upgrade

As already told there are many ways to upgrade a Nimble Storage Array. You can do this by either upgrading the storage processors or the (flash) drives. And the great thing about it is that it can be done non-disruptive. This is a great way to start low and aim high. If your company grows, or the demand for more IO or data grows you have an easy way to grow your storage with that.

Here is the last video of storage field day 2 in which you’ll get more information on the way to upgrade a Nimble Storage Array:

For more information please visit the Nimble Storage site and the Tech Field Day Site.


My travel, accommodations, and meals were paid for by the Storage Field Day sponsors. Often there is swag involved as well. Like me, all TFD delegates are independents. We tweet, write, and say whatever we wish at all times, including during the sessions.

The first day of Storage Field Day 2

Tomorrow we’ll have a busy schedule on the first day of Storage Field Day 2. I’m really looking forward on hearing about the cool stuff these companies will be presenting. Here is our schedule for tomorrow:

08:00-10:00 Nimble Storage
10:30-12:30 Nimbus Data
13:30-15:30 Zerto
16:00-18:00 Tintri
18:45-19:30 Asigra

Let’s take a closer look at the companies we’ll be visiting tomorrow (click on the pictures for more information):

The mission of Nimble Storage is simple: Give our customers the most efficient, high performing, and scalable storage platform on the market. Nimble arrays can easily handle all your applications, user data, and server and desktop virtualization projects. Based on the CASL™ architecture, Nimble Storage accelerates applications, protects more data, and empowers IT to take on new projects and drive growth.

Nimble Storage was founded and is run by industry veterans responsible for ground-breaking technologies and architectures around primary storage, backup storage, and data protection. They’ve built a tight-knit culture of innovation focused on changing the data storage game, taking a unique, bottoms-up approach to reinventing how the data for next-generation datacenter applications is stored, served, and protected.

Nimbus Data Systems, Inc. develops award-winning Sustainable Storage® systems, the most intelligent, efficient and fault-tolerant solid state storage platform engineered for server and desktop virtualization, databases, HPC, and next-generation cloud infrastructure. Combining low-latency flash memory hardware, comprehensive data management and protection software, and highly-scalable multiprotocol storage features, Nimbus systems deliver dramatically greater performance at a significantly lower operating cost than conventional disk-based primary storage arrays, all at a comparable acquisition cost.

In fact, up to 50x greater performance. Nimbus’ all-flash enterprise storage systems leverage the incredible IO performance of solid state technology to deliver superior application performance to Server and Desktop Virtualization, Databases and OLTP, and compute-intensive HPC and Rich Media.

Cut 80% of your storage power budget while reclaiming racks of datacenter space. Delivering the IOps equivalent of 14 racks of 15K RPM disks in just one rack of flash memory, and with power consumption as low as 5 watts per terabyte, Nimbus delivers the industry’s most efficient primary storage.

Dramatically simplify storage provisioning and high availability with unified storage. Complete software and enterprise-grade fault tolerance proven over 5 years at customers like eBay and the US Department of Defense simplify storage administration so your IT staff can deliver better uptime in less time.

Zerto provides enterprise-class disaster recovery and business continuity solutions for virtualized data centers and cloud environments.

Zerto provides large enterprises with data replication solutions designed specifically for virtualized infrastructure and the cloud. Zerto Virtual Replication is the industry’s first hypervisor-based replication solution for tier-one applications, replacing traditional array-based BC/DR solutions that were not built to deal with the virtual paradigm.

Today, enterprises of all sizes are deploying applications on virtualized IT infrastructures and clouds. The values of virtualization including flexibility, scalability and mobility can complicate business continuity efforts in times of disaster. The advantages of virtualization can be in conflict with your ability to recover the business in the event of disaster.

Tintri has an exceptionally strong team. With a demonstrated track record of recent success — including key leaders from VMware, NetApp & Data Domain — Tintri’s talent pool is unusually deep and accomplished. The fusion of pioneers from both virtualization and storage domains gives Tintri a unique perspective, unburdened by traditional orthodoxies and preconceptions.

From the beginning, Tintri has focused exclusively on storage for VMs. Tintri’s founders identified the clear mismatch between traditional networked storage approaches and an elegant shared storage model for VMs. This sharp focus enables both pervasive simplification, as well as the creative application of key new technologies in service of furthering virtualization.

Tintri’s vision is simple: everything in the data center should be virtualized. Our mission is to drive dramatically increased adoption of virtualization by providing a clearly superior alternative to traditional networked storage adapted to VMs. We believe that this is only possible with storage specifically designed for VMs.

Asigra brings a Cloud Model to backup and recovery – including the backup and recovery of mobile devices such as tablets and smart phones – through our Cloud Backup, Recovery and Restore software solution: Asigra Cloud Backup™ v11.

The business model is one of building and fostering partnerships. Asigra does not sell direct, but through a global ecosystem of Managed Service Providers (MSPs) and Resellers that are Powered by Asigra. We are not a MSP and there is no channel conflict between Asigra and our partners.

It’s great to see all these companies on the Storage Field Day. A few of them I’ve met on Storage Field Day 1 in April,, and a few of them I know from meeting them on VMworld or other conferences.

Make sure to tune in at Or the live stream