Backup Central Live! survey yielded interesting results

Tape is still hot, cloud is getting hotter, but companies are still stagnated when it comes to capital purchases, according to a recent survey of 156 IT professionals during the Backup Central Live! seminar series.

I purchased an audience response system from Option Technologies for our Backup Central Live! seminar series.  (We will be in 20 North American cities and several cities abroad this year.) Each attendee is given a keypad that allows them to answer questions displayed on the screen during our seminars.  The system displays what everyone’s answers are and stores them in a database for later retrieval.  This increases audience participation, provides some interesting live verification to the audience of what the speaker is saying, and can provide some interesting marketing information when you compile everything together.

While not every attendee pushed the buttons on their keypads, we got a much larger percentage of response than you would get from any email-type survey. Such surveys yield single-digit percentages at best.  In contrast, the vast majority of our attendees responded to the live survey, giving us a total of 156 respondents.  (3 of them walked off with the keypad!  We know who you are…)

The biggest surprises were the number of people that are still doing things exactly the opposite of what we’ve been telling people to do for years.  (This brings up another advantages of these systems.  The perceived anonymity allows people to be much more candid in their responses.)

Given that I’ve been preaching for years that the one thing that people needed to stop doing was backing up directly to tape across the network, it was surprising for me to learn that 49.1% of them are still doing exactly that — backing up directly to tape with no disk buffer.  By the way, I don’t have a problem using tape at all.  I just believe you should stage backups to disk prior to sending them to tape.  The mismatch between the speed of the backups and the speed of tape devices is the number one cause of backup system failure today.  Every backup system specialist I know has been saying this for nearly a decade, so it’s surprising that half of the respondents are still backing up directly to tape.  A related statistic showed that even though many people have moved to disk staging or deduplicated targets, 88% still use tape as their final destination for backups.  Only 13% of the respondents have gone tapeless.  (61% of the respondents said they are not using dedupe at all.)

Another big surprise came when we talked about long term retention (greater than 1 year) and how it is being accomplished.  The only reason to store backups for multiple years is e-discovery and space reclamation.  Backup is lousy at both of them, so backup software shouldn’t be used to meet this requirement.  This is why it was a surprise to learn that 60% of the respondents were still using backup software that had no e-discovery capabilities to meet their e-discovery requirements!  Couple that with the fact that 24% of the respondents also said their retention on their backups was infinite, or that 79% of the respondents said they retain backups for longer than a year, and you have a “disaster” waiting to happen.  All it will take is one e-discovery request to cost each of these companies far more money than a full archiving system would have cost.  Then they’ll wish they had done something different.

42% of the audience felt they do not have a good understanding of the amount of data they are managing. This is a very common problem as well, leading to difficulty in planning for the capacity of the primary and backup systems.  Storage management software can help, but very few people use it.

My final surprise came when we talked about the cloud.  For those who think that the cloud is all hype and no purchases, consider this.  For something that’s only a few years old, I think it’s impressive that over 10% of the respondents said they were storing reference data in the public cloud, 10% said they were storing primary data in the public cloud, and 10.6% said they were storing backups in the cloud.  78.9% of the respondents are not using the public cloud at all, so there’s certainly room for growth.  One reason for the popularity of the cloud could be that 41% of the respondents said that their company is not making any capital purchases right now. This was cited as the number one barrier to deploying new technology.  Since cloud is all about a monthly bill and not a large capital purchase, it is very compatible with the way many IT departments are now being forced to conduct themselves.

Like I said, tape is still hot, cloud is getting hotter, and many people are not making capital purchases.  Very interesting stuff, if you ask me.


Written by W. Curtis Preston (@wcpreston), four-time O'Reilly author, and host of The Backup Wrap-up podcast. I am now the Technology Evangelist at Sullivan Strickler, which helps companies manage their legacy data

4 comments
  • Curtis,
    Very interesting indeed. I am curious. On your road tour, will you be asking the same questions in every venue? Building a larger sample base?

    Regards,
    Peter

  • A small % response to a vast email campaign is still substantially more feedback than a large % response from a series of small venues, no? I suppose it might be interesting to look at the cost per response instead.

    Having said that, I love the results. We’ve known from end-user conversations that tape use was still significantly higher than disk vendors might have us believe, but we had no hard data. Nice to see numbers that support it.

    Re: the cloud. It would have been more insightful to ask what % of each type of data (primary, backup, etc) they store in the cloud. For example, a company might claim to use the cloud for primary data, but what % of it? 100%? 25%? It reminds me of a typical press release that announces “ACME Inc adopted enterprise Vaporware 8.0” only to discover that ACME uses the app in a very narrow group-level implementation in a satellite office in Siberia. It’s all about understanding the context.

  • Not sure I was trying to say my survey was better than an email survey from a statistical perspective, just that it gets a much higher percentage of the sample size. I’m also saying that it’s a great way to interact with the audience.

  • @Peter

    This was our first time out with five cities. We will probably take many of these questions on the road with our other 20 cities we’re going to do this year, and maybe add a few more. (BTW, the sample size would have been even bigger, but we had technical difficulties in the first city.(