News Stay informed about the latest enterprise technology news and product updates.

Backups and backstops

Like the diving catch in baseball, mastering data backup and disaster recovery can be nothing short of an acrobatic feat.

CHICAGO -- In front of roughly 800 storage managers at Storage Management 2002 conference, author and commentator Jon William Toigo spoke about how mastering data backup and disaster recovery can be as tricky as baseball's diving catch -- one of the sport's most acrobatic feats.

Toigo stepped back about 20-30 years to look at the mainframe and how disaster recovery was treated in years past.

"Back in those days disaster recovery (DR) was looked upon as a 1-to-1 replacement," said Toigo. "Now it's way too expensive to do 1-to-1 replacement. Think about buying a replacement for all of the systems on your floor at a different location."

Toigo also touched upon how time-to-data, or the amount of time to recovery all of the data lost in your systems has been slashed. Previously companies had about a 24-72 hour window to retrieve all lost data -- now companies have to have everything up and running and restored generally within four hours.

"Back only a few years when we had 100 people working in exporation and production we may have gone down for an afternoon and worn it," said Bill Johnson, IT Architect at Woodside Energy Ltd., based in Perth, Western Australia. "Now we have 300 of them and systems need to be up 24/7."

Johnson's company, Australia's largest independent oil & gas exploration company by market capitilization -- is also under the watchful eye of government officials who have new regulations that account for not only data recovery, but what Toigo calls the number one asset in a company -- people.

"For example, we have applications in Perth that are monitoring systems that can directly affect people's safety. The government makes us account for every single one of our employees on each rig at all times," said Johnson. "So availability of our data is a top priority."

Toigo also notes two musts of DR planning: disaster avoidance and data protection. While Toigo admits that you cannot avoid all disasters, it's most important to look at what you can do to minimize the impact of a disaster by looking into high availability systems, for example.

Under the umbrella of data protection is the notion that all data needs to be replicated data not replaced data. For many smaller to mid-sized companies Toigo recommends tape as a preferred method of backup.

Toigo also did not advocate the implementation of a SAN as a replacement for tape.

"The last SAN implementation I witnessed, you better have had a tape backup system because it was so unstable," said Toigo.

Toigo has also been noticing a trend towards a re-birth in another concept borrowed from the mainframe, hierarchical storage management (HSM). Under this concept users are employing a 'cheaper disk' method where they are putting static and less accessed data on lower cost disks.

One of the challenges of using HSM Togio points out is making sure the correct data goes on the cheaper solution.

"If you are saving streaming porn, then a few dropped frames might not matter," said Toigo. "But if you are in the medical profession and you have x-rays of a lung cancer patient's lungs, and there is a shadow over one of the frames, then you have a problem."

For more information:

Special event coverage from Storage Management 2002 in Chicago

  • Listen to and archive of John Toigo's 'Building disaster-tolerant storage infrastructures'

  • Read John Toigo's monthly column 'Toigo's take on storage'

  • Take a look at our weekly backup tips

  • Scrool through our best web links on backup and disaster recovery

  • Dig Deeper on Data storage strategy

    Start the conversation

    Send me notifications when other members comment.

    Please create a username to comment.