back to article NetApp reveals Exadata backup plan

NetApp has quietly announced, through its blogs and user community, a backup solution for Oracle's Exadata appliances. Oracle has previously suggested its own storage kit is the best place to protect the contents of Exadata boxes, but the NetApp 'Technical Report', titled Deploying a Backup Solution and Dev/Test Platform for …

COMMENTS

This topic is closed for new posts.
  1. Bill Bickle

    I smell something funny

    Sounds like a bad business decision to me. So NetApp can sell to the 100+ Exadata customers in the world ? The Exadata systems will be low volume sellers, and Oracle will try to force Sun/Oracle storage sales to them. So why would NetApp waste their time on this ?

    My guess is that Oracle had something on them, like "we won't certify any of your storage systems for our Oracle Database, unless you come out with a compatible storage system for our overpriced engineered systems".

    It really feels like Oracle is going to great pains to convince people that Exadata and Exa-everything are big sellers and in high demand. But it is not true. They had to admit on their last earnings call that it is below forecast, and it sure smells like a bunch of baloney to me.

    NetApp was likely strong-armed into this, and Oracle Exa systems are the next Data Generals and Primes of the world.

    1. J.T

      Re: I smell something funny

      NetApp isn't the only company out there. There's a huge gaping hole in Oracle's Exa solution when it comes to backups, and any company with a backup solution is trying to figure out how to solve that. Sales are Sales.

      The issue is under Oracle's solution, you back up your production databases to a ZFS appliance. Sounds great, except that because the production system is using columnar compression, you aren't going to get that much more. ZFS appliances give you less than a 100TB usable due to everything having to be RAID 1 and a lack of hardware RAID protection making them use a lot of empty space for sparing (plus a slightly higher amount of ZFS overhead). So basically, you aren't going to get that many backups, so you cannot do things like keep historical backups. You know, for discoveries, legal discoveries, all of those things you are legally required to do. Also, if you have a poorly designed system where many different systems have to all be kept in the same time sync, if you can't bring that database to the same time stamp you are right and justly screwed.

      Oh, Oracle is never going to certify storage systems for its Exa systems, the only reason it performs like it does is because Oracle offloads some of the database query crunching to the storage nodes. Considering the issues with heat, sparing, and data protection with the current hardware, Oracle would effectively kill the majority of it's up front costs for Exadatas (they'll still kill you later with software maintenance).

      1. Kebabbert

        Re: I smell something funny

        @J.T

        Seriously, what are you talking about? You obviously have no clue about ZFS and what is does. I suggest you read more on ZFS before writing weird things. Or are you just trying to spread FUD?

        .

        "...ZFS appliances give you less than a 100TB usable due to everything having to be RAID 1 and a lack of hardware RAID protection making them use a lot of empty space for sparing (plus a slightly higher amount of ZFS overhead)..."

        What less than 100TB? There are PetaByte ZFS servers out there. Wrong.

        And what do you mean with "everything having to be RAID 1"? ZFS has no RAID 1. ZFS has raidz1 (corresponds to raid-5) and raidz2 (corresponds to raid-6) and raidz3 (corresponds to none, it allows 3 disks to fail which is unique). Wrong again.

        What do you mean with "lack of hardware raid protection"? Dont you know that hardware raid is quite unsafe, and that ZFS provides a superior protection? Dont you know that there are research on this? Dont you know that hardware raid might corrupt your data, without you ever knowing it? Dont you know that ZFS is built to detect data corruption so ZFS is also safer from a data corruption view point? Have you missed the research papers on data corruption? Dont you know that hardware raid is susceptible to "write hole error", but ZFS is built to eliminate that design flaw? Dont you? Wrong again.

        .

        "...So basically, you aren't going to get that many backups, so you cannot do things like keep historical backups...."

        This is just so wrong it is not even funny. ZFS has snapshots because it is CoW. Each time you make a snapshot, every change is written to a new place on the disk, but all old data is still left intact. This way you can have many snapshots (costing nothing, only the changes are recorded) going back and forth in time. Just like Apple OS X Time Machine. So you can keep historical backups.

        Do you work for NetApp or EMC, trying to FUD about ZFS? Or are you just ignorant, claiming weird things without even nothing anything about ZFS?

    2. Billl
      Thumb Down

      Re: I smell something funny

      I smell something funny too. It's your logic and obvious sour grapes.

      IDC says that Exadata has sold over 1,000 Exadata's (as of about Oct '11). This matches the number that Oracle provides as well. I guess the average customer could have 10 of these things, but I doubt it. Oracle says that they expect to have sold about 3,000 by 2013. Not a bad market to get into.

      One question that I have on the NetApp solution is can it do HCC? The Oracle backup solution with ZFS Appliance supports HCC so you do not have to decompress and then backup -- maybe it's not as much of a downfall as I think though. Also, can you hook up the NetApp storage directly to the infiniband network? Without this, then you would have to backup over the 1GbE or 10GbE network using RMAN, right? That does not seem like a comparable solution to the ZFS Storage solution that Oracle sells.

      OK... I just looked a bit closer at the NetApp whitepaper, and it seems that you have to uncompress the data before backing up, and all of the compression benefit they show is only with non-hcc compressed data. I guess that answers that question.

      There's also no mention of Infiniband, so I assume you can't hook up the NetApp storage to the Infiniband network either.

      1. J.T

        Re: I smell something funny

        They haven't sold 1000 Exadatas, they've sold 1000 Exadata and Exalogic systems.

        1. Anonymous Coward
          Anonymous Coward

          Re: I smell something funny

          @J.T.

          Too funny. Oracle's competition can't create a competitive product to what Oracle is doing, so they make silly, unsubstantiated comments such as this. IDC did not say Exalogic and Exadata, they said Exadata. So you're calling IDC liars?

  2. Marin Debelic

    STK

    This is interesting.... I wonder how supportive will Oracle (ex-STK) be of this offering?

  3. Bill Bickle

    Funny exa-numbers ?

    IDC only takes data from what vendors tell them. And Oracle has been changing their story on Exadata, and Exadata and Exalogic numbers. This link below describes it and points to the earnings call transcripts.

    http://database-diary.com/2012/01/30/oracle-reduce-their-exadata-projections/

    I have heard that Oracle has been asking their sales reps to do weekly forecast updates for Exa products (instead of monthly for others), and 3-year forecasts for Exa (instead of 1 year for other products), which to me feels like someone desperate to try and get people to believe that there is a huge pipeline of demand and opportunity, that may not really be there. Somewhat un-natural motions, since it seems to me that they will be hard to sell in any decent volumes. That was somewhat of my original point. That Oracle has been, and continues to talk up the "demand" and interest, to try and convince people they need one and they are flying off the shelves.

    I wish someone would actually do a survey of end users or VAR's and see if any real value, Not an interview with the Oracle distributors, since they will just say what Oracle tells them.

    I predict the next 12-18 months volume predictions will keep getting pulled downward by Oracle, unless they stop providing details to the market, since it may be too embarrassing.

This topic is closed for new posts.

Other stories you might like