This is the story of our very first customer-deployed array – shipped out on April 19, 2010, well before Nimble Storage was even formally launched! The original system shipped was a CS220, our initial entry-level system. In fact, the system’s controller had been hand-assembled by our engineers, rather than a contract manufacturer. In the words of Huy Duong, one of our most senior support engineers who’s now taken the role of a field engineer, “it even flexed a little bit – like an old vintage Porsche.”
Huy learned recently that the customer wanted to increase the performance of the system – three years after the initial system had been deployed.
First, the customer replaced all of the old flash solid state drives (SSDs) in the system to higher capacity SSDs during the day, watching the amount of cache increase, even as the system continued operations with no disruption to any applications. That evening, Huy arrived and pulled out the old controller and popped in a new, higher performance CS420 controller. He then failed back to the upgraded controller, and replaced the second older controller with a newer CS420 controller. No downtime or disruption to any application, and the performance was now three times higher than with the CS220! The system was now brand new, ready to chug along for several more years.
Having been in the storage industry for a long time, I know that non-disruptive software upgrades are marketed far more often than they are truly realized in customer data centers. Non-disruptive controller upgrades and cache upgrades are non-existent within mid-range enterprise storage systems. This is one of the many capabilities of our architecture that is allowing us to earn customer appreciation and build a strong customer community!
Interesting. I have the first shipped out of state arrays (a pair of CS240s that arrived in May 2010). I wonder if I should bug the guys about upgrading my controllers too
[...] Anyway, the chassis has two controllers that are hot-swappable in the back in an Active / Passive configuration. Most SANs are Active / Active, and for those of you who are uncomfortable with this, in a nutshell, here is how the controller works. As data is written, it is first written to NVRAM, which is mirrored across controllers – these means that if Controller A drops out in the middle of a write, Controller B already has the exact same data in NVRAM and therefore nothing is lost. The failovers are absolutely seamless. We have done several test failovers and there is absolutely 0 interruption. The controllers are housed in the back of the unit and slide out. These can be upgraded at any time should your infrastructure need it (IE, you can go from a CS240 to a CS440 just by swapping controllers). This can be done hot. [...]