Kaminario K2 [EOL] Reviews

The solution has a very simple configuration. It has a good configuration, it's very simple. 

Access to technical support should be improved for our region. Technical support is good, but they're very hard to access.

When we have a problem with this storage, it's usually related to a very bad configuration. 

All-Flash Storage is very, very expensive. They are important solutions so I don't really have any ideas about how to rectify this. 

The device could use better monitoring tools.

production real time environments in 3 data centers 

Miami, London and Frankfurt

About 4500 virtual machines

We used to be work with EMC solutions VNX and VMAX, which were very costly, hard to maintain, and no matter how many upgrades we did every year, it was always a bottleneck and the first place to look during an operational issue.

After we upgraded, I don't remember anyone in the company coming to check if the problem was with the storage and all the DB transactions that caused downtime in the past are now a daily event without anyone noticing.

• Scale out: Scale out is a differentiator for them, especially in the enterprise market. It's key for a lot of customers.
• There's no forced forklift upgrades. The customer that has a first generation of the product or later of the product knows that they will all work together harmoniously.
• The data round is not a big deal. There's no hard down points to this stuff, which can be a big differentiator.
• Adaptive block size is another differentiator. A lot of customers nowadays commoditize their storage. They want it as easy and simple as possible. A lot of storage rates out there, they do different block sizes very well. Where they can change block sizes, or set certain block sizes. You don't worry about any of that with Kaminario. The adaptive block size just moves it around, gets it all figured out, and does it. Where I see this working well with a lot of customers, are those who want something extremely simple, performant, and scalable.

It's very hard to tell what else is super important to customers. Besides, maybe it's obvious stuff like the maintenance program. They have a great maintenance program and as the systems go up, year over year, it stays the same. It can go down, but won't go up, year over year. So coupled with the simplicity, performance, and just absolute ease of use of the management of it, I don't know why anyone wouldn't buy it. I'm kind of confused why people don't buy it, actually. I know there's a lot of political talk where people will buy, an EMC, HP, or IBM. These guys have got some of the best venture capitalists behind them and they've got some of the best numbers quarter over quarter. Otherwise, I don't know why people wouldn't want to even look at this product.

They might not like me saying this, but I would love to see capacity on its DRAM. I know it's not cost effective for them to do it, but I think that it could be a big differentiator and was a big differentiator from the beginning. It might make the reseller's jobs easier. Not having to wait for something; just give them the quote for the amount of time, they can have as much as they want. That message resonated very well with customers, as you can imagine, so I wish that came back. I understand why it's not there, financially it doesn't make sense, but I wish it were still around.

I wish they had more ISP or certified partnerships with specific use cases. For instance, epic and health care would be a huge use case that I know they're working on. But there's a few of those, like epic and health care where it might not matter if it's the best technology, the best people, or the best support system. If it's not on a specific list or a conference level, I think if they worked hard at what they're doing at getting these ISPs and specific application awarenesses to these companies that it would be gold mine.

Image backups of our production servers that used to take hours now take minutes. This has greatly increased performance of our database and business systems, and eliminated periods of sluggish performance.

The primary benefit is they actually guarantee performance and capacity. I don’t know of any other firm that backs this up with a commitment to add hardware until the performance guarantee is satisfied.

The increased performance is many times above our previous array performance in all metrics. Integration with vSphere features is also a definite plus.

The speed is the most valuable feature; the speed and, for us in particular in what we're doing, the data de-duplication. We're using this for our continuous-build system. We just have a lot of the same systems that have the same information on them; while they get built, de-dup's huge.

We trialed POCs starting in November of last year (2015). We purchased in January of this year. We've been going now for nine months in production, and three months prior to that as a trial of the hardware.

• Ability to deploy large numbers of virtual desktops
• Enables quick updates to virtual desktops

We thought we could only afford a hybrid solutions.

Kaminario allowed us a solid state solution with marginal incremental cost.

More important was the software/logic of how Kaminario de-dups and manages the data sets.

• Solid-state storage drives
• Price per gigabyte
• Logic/software management

We no longer have to consider the performance penalties associated with different application profiles in our cloud stack. Having a high-performance SAN core gives us instant resources, with capacity on demand.

Standard-based monitoring means we aren't locked in to using platform-specific or proprietary tooling to track and monitor our device performance.

Built-in snapshot support gives us SAN-side functionality most other platforms would have had us license separately.

Unexpectedly high compression ratios give us more capacity for our investment; again without a nickel-and-dime licensing model.

Rocketing IOPS results in performance we don't have to fixate on. We can instead focus efforts on other layers in the stack, knowing that our SAN layer is lightning fast, with capacity to spare.

Using snapshots allows us to automate production database refreshes to development environments in minutes, rather than in hours.

• Data reduction and snapshot abilities: Smaller footprint in the datacenter (lower cost for power, cooling, etc.).