BandL Weblog

Debunking the Myths about Tape Storage

Tweet Share

Google the keywords “tape storage myths,” and you’ll turn up no less than 1,780,000 results. While certainly not all of these results are relevant, the sheer volume of listings speaks to the fact that there’s no shortage of myths out there about storing data on tape.

Myths—a word that implies unproven or false information, of course—abound about how tape compares to disk: “tape is obsolete,” “tape is more expensive to operate,” “tape is unreliable,” “tape is less secure,” “tape is slower,” “tape is less resilient,” the list goes on. If you dig into these search results, you’ll also find that there’s a lot of good information out there dispelling, debunking and generally disproving these false statements about tape. Some especially strong statements from Oracle and Storage Switzerland earlier this year do a great job of presenting a factual case for tape, and rather than rehash their excellent reports I’ll simply suggest you check these out:

• “Debunking Five Myths About Tape Storage” from Oracle
  
  
• “Breaking the Top Four Myths of Tape vs. Disk Backup” from Storage Switzerland

The ease with which the myths surrounding tape can be debunked makes me wonder: Where did these myths come from in the first place? Could it have been vendors of disk-only solutions making a hard sell for their products? I’m not saying anyone intentionally flooded the marketplace with false information about tape. But I think it’s possible—likely, even—that the purveyors of the myths around tape were vendors offering a very single-minded, “either/or,” down-with-tape-and-up-with-disk view of where the enterprise storage model was going. (Remember the “Tape Sucks, Move On” slogan?) That way, vendors of disk-only solutions didn’t have to “worry about integration” of tape and disk storage, as George Crump of Storage Switzerland puts it. It’s that kind of thinking that has focused many in our industry on the tape vs. disk debate, rather than on intelligently incorporating multiple storage media into a tiered enterprise storage strategy. 

With this in mind, perhaps the most untrue myth about tape is the last one George Crump cites in his post on the subject: “tape and disk must be separate.” George is right: Today, there’s no good reason why companies can’t use tape and disk, along with cloud solutions, together as part of a tiered storage solution. Tape and disk can indeed coexist, allowing enterprises to take advantage of the strengths of both media in different storage tiers—and that’s no myth. 

 

The Future of Tape: The Short and the Long of It

Tweet Share

Two recent storage media technology news items underscore the fact that tape will live on in the data storage world in both the short and long term:

 

• LTO-6: In the relatively near term (although the availability date hasn’t been announced yet), Linear Tape-Open Ultrium Generation 6 tape format licenses are available, as I mentioned over the summer. A recent post on IT Knowledge Exchange states that vendors already are “starting to make noises about LTO-6.” LTO-6’s 8 TB capacity (triple that of its LTO-5 predecessor) and increase in data transfer speed to up to 525 MB per second demonstrate that tape is continuing to evolve to meet the storage needs of the future. The LTO Consortium also has erased any doubt that tape will continue evolving over the long term with its roadmap showing the next two generations beyond LTO-6, with further enhancements in capacity and data transfer speed.
  
  
•  “Petabyte in a Rack Unit”: IBM is developing a “futuristic” storage medium, which it expects to launch in about three years, that could store data for as long as 50 years. This technology advancement would eliminate the need to move data from one storage device to a newer generation of hardware, also doing away with the cost and risk such a transfer incurs. Described only as an “unspecified magnetic technology that could store a petabyte of data in a standard 1U rack unit,” this technology would be used in conjunction with tape, maybe to access archived data via the cloud. 

Both IBM’s thinking around this development and the LTO Consortium’s plans for LTO-6 send a strong statement: Rather than disappearing, tape will continue to play a role in data storage, and even technologies described as “futuristic” will continue to work alongside the old storage gold standard. As storage technologies evolve, so too will tape—a medium that, clearly, will remain alive and well in both the short and long term. 

 

Why Tape Storage Helps—Not Hurts—Competitiveness

Tweet Share

A recent article paraphrased an EMC exec as saying that “businesses that still store data on tape are less competitive and more vulnerable to security breaches and data loss.”

Really? So companies like Intel, AT&T and Anheuser-Busch (just to name a few of our tape management clients) are made less competitive by their decision to incorporate a low-cost, reliable, secure, green, durable medium into their data storage strategies? In a word: no. The fact of the matter is that incorporating tape into an enterprise storage strategy can help keep the organization running efficiently and effectively—enhancing, not reducing, its competitiveness.

The EMC spokesperson goes on to say that “Businesses and public sector organizations can dramatically cut costs and become leaner by getting rid of tape, which is cumbersome and vulnerable to security breaches.” In fact, tape beats disk on cost, with the average tape-based solution having been found to cost 1/15 of the average disk-based solution. Tape storage is extremely straightforward to manage, and is simplified further when you add in a tape management solution. As far as security vulnerability, data stored offline on tape is out of hackers’ reach and safe from system corruption, unlike data stored online or in the cloud. And, in this day and age, we know how to keep track of tape so there’s no excuse for the security breach of having one go missing. If properly managed, tape storage is an extremely secure way to go.

The EMC source cites a “pressing” need for companies to switch from tapes to electronic disk libraries. The two key reasons he gives:

1. “relentless data growth” driving the need for cost-effective storage
   
   
2. the need for organizations to retain data for extended timeframes for legal reasons

Truth be told, these are actually reasons to keep or even expand tape as part of your storage strategy.

Tape’s relative cost advantage makes it very well suited to play a role in storing all of those bits and bytes resulting from the data explosion. Tape especially can add value as part of a tiered storage strategy or active archive, providing a low-cost way to store the large percentage of organizational data that is inactive. Tape’s cost-efficiency, environmental friendliness, reliability and durability also make it perfectly suited for long-term data storage for legal, compliance or business reasons.

These are the facts of the case for tape and, although they are still disputed, they really do speak for themselves. I beg to differ, Mr. Ward of EMC: Tape does not lower a business’ competitiveness—exactly the opposite. Let the healthy debate continue, as tape continues to prove itself as an efficient, effective, even competitive way for businesses to store their ever-growing volumes of data.

What do you think? Does tape help or hinder organizations?

Tape Management: A Necessary Nuisance

Tweet Share

A survey of 125 small businesses turned up this not-so-surprising result: 31 percent of small-business owners said that backing up data is a nuisance that takes time away from running their business. With this outlook in mind, it’s also not surprising that the same study revealed that 48 percent of small businesses with between two and 20 employees had experienced data loss. 

Of course, the study does talk about the data backup methods these small businesses are using—so, clearly, they seem to be doing some level of backup. But one might venture a guess that many organizations for which backup is a nonessential hassle—versus a critical part of running the business—are managing the backup process haphazardly or not at all, resulting in data loss.

Backup management is one area where the old saying really applies: “An ounce of prevention is worth a pound of cure.” No matter how much of a hassle backing up your data is, it doesn’t compare to the world of hurt a preventable data loss can bring to your organization.

For companies backing up to tape, the ounce of prevention is not just about making sure that you do run backups, but also about managing your tapes effectively. Investing the time and budget up front in developing a strategy and/or a system for managing your tapes can save you from needing pounds and pounds of “cure” down the road.

Managing backup tapes, like other routine technology tasks, is not why most of us got into IT in the first place—or why most entrepreneurs started their small businesses. But while there’s nothing glamorous about tape management, not managing your tapes could have a decidedly unglamorous result. Not being able to find a tape needed to perform a data restore, respond to an e-discovery request or meet compliance requirements is a much bigger hassle than just managing backups properly in the first place.

Tape offers a low-cost, reliable backup method that is straightforward to manage—and with today’s tape management systems, there’s no excuse for the type of data loss the surveyed organizations reported. Nuisance or not, tape management is a critical part of protecting your business—one that you’ll ignore at your own peril. 

4 Best Practices for Automating Your Data Center

Tweet Share

Thinking about automating some of your data center processes? These days, IT shops everywhere are automating processes like provisioning, compliance, routine maintenance, configuration management and tape and vault management. The end goal: to improve efficiency, raise productivity, reduce costs and improve quality—in other words, enable IT to do more with less.

Data center managers can take their pick from among the vast numbers of data center automation tools on the market today, and they’re less expensive and more powerful than ever before. But, buyer beware: As with most things IT, the tool alone is not the answer. Automating a convoluted process that isn’t working well to begin with is going to create more work for IT, not less; lower efficiency, not increase; more expense, not less.

If you’re considering automating processes in your data center, here are some proven practices that will serve you well:

1. Use a structured project methodology.

This should be a no-brainer and is a hallmark of any well-managed IT project, but we all know that reality doesn’t always match up with the best-laid plans. According to the data center automation experts at HP, the right approach would include:

• Assessing operational maturity and developing a vision for the end result—one that reflects input from all business and IT stakeholders.
  
  
• Defining comprehensive “value chains”—that is, end-to-end sets of steps that IT executes to deliver value to the business—in the areas you want to automate. Whether you’re looking to automate application code promotion, OS provisioning, patch management, software management, monitoring and alert, troubleshooting and repair, audit and compliance, or even disaster recovery, process must always come first.
  
  
• Developing standardized, “gold standard” processes for the areas you want to automate.
  
  
• Implementing the new and automated processes, with clear agreement up front among all stakeholders about what success means and what ROI the project will deliver.

2. Let go of homegrown solutions.

Many IT organizations have gone to the laudable effort to create their own scripting tools and workflow engines. But these tools don’t scale as the organization’s needs and its data centers grow.

3. Go with an open environment.

When choosing an IT automation technology, make sure the environment is open and extensible, so that you’ll be able to integrate any application in the future. The wise investment is in tools “that offer the most extensible approach possible because, after all, no one really knows what might be coming through the data center door next,” according to Mike Vizard of CTOEdge.com.

4. Keep it simple.

Choose data center automation tools that will have a short learning curve for your staff, for example, those with a drag-and-drop interface.

Reaping the value of automation isn’t automatic. There’s more to it than just implementing new software. Let your business needs and well-defined processes drive your technology solutions, and you could be well on your way to managing a more efficient, effective data center that better meets the demands of your business.

And if you do decide to automate your datacenter and want to include a tape management solution, contact B&L. We'll be happy to talk to you about your challenges and offer some time- and money-saving solutions.

What do you think? Is there anything else you would add to the list?

How 9/11 Has Changed the Data Center

Tweet Share

Over the past 10 years, the 9/11 terrorist attacks have changed the face of America in countless ways. Among them, the shock of that devastating event has influenced the way many companies manage their data.

Here are some of the key trends in the way businesses manage and protect their data as a result:

• Disaster recovery quickly rose to the top of the priority list following 9/11. And the DR industry continues to grow, even in an era of shrinking IT budgets. ABI Research predicted that spending on business continuity and data disaster recovery products would grow from $14.3 billion in 2009 to more than $39 billion in 2015.
  
  
• Data center backup and redundancy has taken on a whole new meaning. No longer is a backup server in the closet across the hall considered adequate. Today, redundancy means a mirrored system with geographic separation.
  
  
• 9/11 is credited with contributing to the modern cloud storage industry, enabling companies to store data at multiple locations to reduce vulnerability.
  
  
• Data center design today takes into account the threat of terrorist attacks, which many companies didn’t see as a real possibility until 9/11. Today, data centers are less likely to be built in cities and more likely to have more open space and multiple levels of heightened security.
  
  
• Companies have more data to capture and manage in order to comply with new regulations like the Patriot Act.
  
  
• Firms have a heightened awareness of cybersecurity and are making investments to protect their data.

Clearly, a lot has changed in the data center in the last 10 years with much of it driven—at least in part—by the fact that business data is vunerable. Still, nearly half of all small to medium-sized businesses don’t have a disaster recovery plan. And 30 to 40 percent of all organizations either have no DR system in place or don’t know how to use it properly.

If there’s one thing we should have learned over the last ten years is that no business, large or small, is exempt from the potential to be affected by a disaster of some kind. If you haven't rethought your data center disaster recovery strategy lately, maybe now is a good opportunity to revisit this important topic.

 

Cloud Backup: We’re Not There Yet

Tweet Share

A few weeks back, short-lived cloud outages at Amazon Web Services and Microsoft Office 365 reminded us of one of the uncertainties that linger about cloud computing: its reliability.

While the recent outages impacted cloud hosting services, for those of us in the data storage end of the business, they called to mind the implications for cloud backup.

So far, the uncertainty about cloud reliability, among other factors, is proving too much for many organizations to consider shifting some or all of their backup processes to the cloud. As a result, most large enterprises are sticking with traditional backup methods, although cloud backup is starting to catch on with smaller firms.

Aside from reliability, organizations are feeling wary about other aspects of cloud backup services, including: 

• Security
  It seems that many cloud service providers just haven’t yet made the investment in IT resources dedicated to security. Encryption is essential but is just one piece of the puzzle.
  
  
• Recovery timeframes: Depending on bandwidth availability, recovering data from the cloud may prove too time consuming. In the event of a data loss, firms might find themselves having data shipped from their cloud service provider on tape—and wishing they had relied on traditional backup methods in the first place—rather than waiting for data to transfer across a slow connection.
  
  
• Size limitations: Every organization will have a limit on how much data can be transferred to the cloud, which might not prove sufficient for some backup needs.
  
  
• Vendor relationships: Complex cost structures, complicated termination requirements and unclear service-level agreements are also leaving organizations feeling uncertain about turning backup over to a cloud service provider.

Many organizations are using the cloud for various applications and reaping the benefits. But, so far, with these factors in mind, backup is one area you’ll want to think twice about leveraging the cloud. 

What do you think of cloud backup? Can it be trusted?

 

With Tape Storage, Everything Old is New Again

Tweet Share

There was a time when tape was “it” when it came to data storage. Disk was too expensive for most storage needs. The cloud didn’t yet exist. Most businesses were backing up to tape, which was inexpensive, reliable and straightforward to manage. Enter increasingly affordable hard disk arrays, and the world of data storage changed. Today, tape, disk and the cloud all play a role in storing the rapidly expanding data universe, with businesses using different mixes of media depending on their needs.

Somewhere in between the old days of tape’s dominance and the new world of peaceful storage media coexistence, tape-bashing came into vogue. Predictions of tape’s pending demise abounded. And while I still do come across the occasional declaration that “tape is dead,” just as often these days I find evidence of tape’s strengths—like in this recent press release from the LTO (Linear Tape-Open Program Technology Provider Companies. OK, so the studies referenced in the press release aren’t exactly new themselves. But they do offer more proof of how the attributes that made tape a great storage medium in the “old” data storage world continue to make it a viable part of the storage media mix in the “new” world.

Specifically, the Enterprise Strategy Group and The Clipper Group studies both found “that tape had significant total cost of ownership benefits over disk, including lower cost per GB, lower operating expenses and lower energy costs.” Sure, tape has always been cost effective—that’s old news. What’s newer news is how tape’s relative cost advantage makes it a better-suited medium for certain applications—primarily long-term storage and archiving—than the alternative.

I’ve said it before and, with the release of these studies, I’ll say it again: Despite many predictions to the contrary, it turns out, tape doesn’t suck. The evidence shared recently by the LTO group, not to mention the experience of millions of businesses that incorporate tape into their enterprise storage strategies, yet again prove otherwise.

What do you think? Does tape have a role in the modern data center?

A Hurricane Disaster Recovery Plan Checklist

Tweet Share

The destructiveness of Hurricane Irene is still fresh in mind here on the East Coast, and the storm’s disruption is still ongoing, bringing with it a reminder of how important disaster recovery planning is. And while hurricanes are typically limited to certain geographies, planning for them probably offers some lessons that prove valuable in data center planning for other disasters as well.

Unless your business is still feeling the effects of Hurricane Irene, now could be a good time to review a checklist for recovering your data center from a hurricane or other disaster. Industry web site Data Center Knowledge.com offers a concise yet thorough guide for hurricane disaster recovery planning. The author’s advice boils down to this:

• Take inventory of IT equipment, complete with serial numbers and photographs, to show your insurance provider following a disaster.
  
  
• Assess your risk. Understand what the impact of a hurricane or other disaster would be for your business in terms of revenue, the client experience and your brand reputation. Help the organization articulate how much data it could afford to lose and what your recovery time objectives (RTO) should be for each system or application.
  
  
• Ensure mobile device security. Most disaster recovery plans involve at least some staff working remotely when your site is unavailable. Take steps to protect data on laptops and other devices, like flash drives, and make sure employees know how to do the same.
  
  
• Test your disaster recovery plan regularly. Quarterly, test your plan. Don’t wait until a storm is brewing to find out if you’ll be able to recover your data or whether you and your vendors see eye-to-eye about RTO. Thoroughly test and drill your staff on your plan.

No one likes to think about the innumerable hypothetical disasters that could befall their businesses. But Hurricane Irene reminds us that this fear of the unknown is exactly why we need to conduct disaster recovery planning in the first place.

What is your DR plan? What advice do you have about developing one? Let us know in the comments section.