Startup News Flash part 10

There we are, part 10 of the Startup News Flash. Someone asked me on Twitter last week why Company XYZ was never included in the news flash. Let it be clear that I am not leaving anyone out (unless I feel they aren’t relevant to this newsletter or my audience), I have limited time so typically do not do briefings… Which means that if the marketing team doesn’t sent me the details via email and I haven’t somehow stumbled across the announcement it will not appear on here. If you want your company to be listed, make sure they sent their press releases over.

Some new models announced by Nutanix. Funny to see how they’ve been pushing hard from a marketing perspective to remove the “pure VDI play” label they had and now launch a VDI focused model called the 7000 series. (Do not get me wrong, I think this is a brilliant move!) The 7000 series offers you the option to include NVIDIA K1 or K2 Grid cards. Primarily intended to accelerate graphics, so if you are for instance doing a lot of 3D rendering or just are a heavy graphical VDI user these could really provide a benefit over their (and other vendors) normal offerings. On top of that the 3000 and 6000 series has been overhauled. The NX-3061 and NX-3061 with 10 Core (2.8GHz) Ivy Bridge have been introduced and the NX6060 and NX6080 10 Core (2.8 and 3.0GHz respectively) have been introduced. Haven’t seen anything around pricing, so can’t comment on that.

No clue what it is exactly these guys do to be honest. I find their teaser video very intriguing. Not much detail to be found around what they are doing other than “re-imagine enteprise computing”. Hoping to hear more from these guys in the future as their teaser did make me curious.

I don’t care much about benchmarks, but it is always nice to see a smaller (or the underdog) company beat the big players. Kaminario managed to outperform Oracle, IBM and Fujitsu with their SPC-2 Performance Benchmark using their scale-out all flash array K2 v4. Just a couple of weeks after breaking the SPC-1 Benchmark World Record again. Like I said, I don’t care much about benchmarks  as it doesn’t typically say much about the operational efficiency etc. Still it is a nice indication of what can be achieved, though your results may vary depending on your IO pattern of course.

Be Sociable, Share!


    1. says

      Dear Duncan,
      Thanks for your comments about Kaminario and the SPC-1 and SPC-2 benchmarks. We agree that these synthetic benchmarks may or may not resemble an organizations actual data workflow. The SPC-1 and SPC-2 benchmarks are a grueling 24 hour sustained 60% write and 40% read benchmark that surpasses most real world production workloads. What we believe it does show is Kaminario’s ability to handle any type of workload due to our scale-out architecture, wide striping all data and metadata across all system nodes, and our variable block size capability between 4k to 128k. As an example of XtremIO speed EMC says that 1,000 linked VDI clones can start up in 15 minutes, while The Taneja Group did a lab validation on the Kaminario K2 v4 demonstrating it took only 3 minutes and 32 seconds to start up 1,000 VDI seats. The real challenge comes in when suspending and then resuming those same 1,000 VDI seats, because as you’ll see in the Kaminario testing the bandwidth required to do these functions are now out of the range of many flash vendors capabilities at up to 12.37GB/s!

    Leave a Reply