6 ordinary things I really like about IBM N Series

I seem to have been doing a lot of work recently on solutions that involve IBM N Series (Netapp) products. There are a few annoying things about the product (e.g. the price, fractional reserve) but there are some things I really like, and they’re not necessarily exciting things or things that we make a big deal about in sales presentations, but they’re the kind of solid features that make a storage architect’s life that bit easier and the pursuit of elegance that little bit more achievable.

  1. Reliability. There was a comment on a recent Infosmack podcast along the lines that in the storage business all you had to do to have a customer for life was to avoid disappointing them – not let them down. Reliability is not unique, but over the years I have watched various competing technologies from other companies disappoint customers with product reliability. So far I haven’t seen that problem on N series.
  2. PAM Modules. This seems to me to be a really smart way to use SSD technology – as a caching layer. I’m not convinced that some alternative approaches like bunging SSDs in relatively dumb disk systems is a good approach.
  3. Replication over IP. If this was easy everyone would do it. Netapp pioneered this and while I’m not sure that sync replication is its strong suit, most customers use replication for DR, so snapshot-based replication over IP is ideal. Also seems to be more tolerant of low bandwidth than most competing systems.
  4. N series Operations Manager Software (AKA DataFabric Manager). This is very reasonably priced add-on software (heavyweight Windows server app) for managing your N Series systems. Good detailed performance reporting, and optional module for file classification and reporting.
  5. Good SATA performance. Yes I know it sounds mundane, and it’s somewhat anecdotal, but customer feedback is good and I have also modeled the performance using IBM’s Disk Magic on several occasions. Again this is not unique, and I know DS8000 SATA performance is very good and EVA used to do a reasonable job with FATA, but there are other technologies out there that seem to run into higher than expected latency and lower than expected sequential throughput on SATA – maybe it’s something to do with the bridge technology used.
  6. Volume striping (aggregates up to 16TB with ONTAP7). Yes I’m really getting dull now. That’s certainly not new or unique, but ONTAP has been doing this well for a long time and it seems less complex than some other vendors’ approaches and much better than those who don’t do volume striping at all.

IBM has N Series in its portfolio primarily for its NAS capability, but with simple solid features like these I also often consider N Series for SAN requirements.

2 Responses

  1. Thanks for the nice words, I’m surprised about the price thing annoying you, the latest IDC figures showed that for ANZ, the $/GB for NetApp was lower than any other vendor and lower than most internal storage, though things might be different for N-Series.

    SATA performance is probably due to a bunch of factors, including the way block checksums are handled, interconnect capability, and the way WAFL will generally turn random write I/O into sequential blocks. The relatively wide striping done via the volume striping also helps, and with 64bit aggregates in 8.x the effective stripe witdths increase quite nicely.

    If you like Operations Manager/DFM now, I think you’ll be pleasantly surprised by some of the upcoming changes and improvements. The storage efficiency plugin is just the beginning :-)

    Like

    • @John Martin
      I don’t want to get drawn into a lengthy debate about pricing, but the disk pricing issue is a generic Netapp issue and not specific to IBM N series. I have also had a Netapp exec admit this to me. IDC numbers need to be treated with caution e.g. our N Series $/MB would be low for the last 12 months as we have sold mainly SATA drives.

      Like

Leave a comment