Manag­ing back­ups in vir­tual en­vi­ron­ments

With vir­tu­al­iza­tion emerg­ing as a top pri­or­ity for IT ad­min­is­tra­tors, the backup and re­cov­ery strate­gies need to change

InformationWeek - - Opinion - Vi­jay Mhaskar

Vir­tu­al­iza­tion has “vir­tu­ally” changed the IT world in which we all work and play. Why is vir­tu­al­iza­tion so at­trac­tive to IT ad­min­is­tra­tors? The an­swer is easy — there are many uses and ben­e­fits that we gain through vir­tu­al­iza­tion. For starters, hav­ing a sin­gle server’s phys­i­cal foot­print rep­re­sent many servers on the net­work is a boon for ad­min­is­tra­tors look­ing at con­sol­i­dat­ing space and re­duc­ing op­er­a­tion costs. Hav­ing the abil­ity to quickly stand up a VM copy of a ma­jor ap­pli­ca­tion or work server for patch test­ing is sim­ply a game changer as it al­lows ad­min­is­tra­tors to test dur­ing busi­ness hours. But what about backup of data in vir­tu­al­ized en­vi­ron­ments?

Let’s take a step back to un­der­stand where we stand in the area of backup and re­cov­ery to­day. Sy­man­tec re­cently sur­veyed more than 1,400 IT pro­fes­sion­als on their backup prac­tices and abil­ity to re­cover in­for­ma­tion in the event of a dis­as­ter. The find­ings strongly sug­gest that tra­di­tional ap­proaches to backup are bro­ken and a new ap­proach is nec­es­sary. Here’s why:

Con­fi­dence in backup is lack­ing —

es­pe­cially vir­tual back­ups. The pur­pose of a backup is to have worry-free data re­cov­ery in case of dis­as­ter. Yet, 36 per­cent of re­spon­dents are not con­fi­dent that their backed up data can be quickly re­cov­ered when needed. And when it comes to vir­tu­al­ized backup, 42 per­cent of re­spon­dents re­ported that their vir­tu­al­iza­tion back­ups don’t work ad­e­quately.

Backup SLAs aren’t be­ing met.

One-third of re­spon­dents in­di­cated that they’re ei­ther not meet­ing backup and re­cov­ery SLAs or are un­sure if they are. Of those not meet­ing SLAs, 49 per­cent said they can’t meet them be­cause they have too much data — due to the size of the backup, lack of band­width, and the sheer vol­ume of data.

The cur­rent backup and re­cov­ery ap­proach is com­plex.

More than onethird of the re­spon­dents noted that backup is ex­tremely time con­sum­ing, and nearly as many said the same about re­cov­ery.

In this sce­nario, it is ev­i­dent that backup and re­cov­ery need to change. To ad­dress grow­ing backup needs, and stream­line the com­plex pro­cesses, en­ter­prises an­tic­i­pate mak­ing sig­nif­i­cant changes in the near fu­ture. Within the next 12 months, tape-based back­ups will de­crease by one-third, and more and more or­ga­ni­za­tions will be in­ves­ti­gat­ing ap­pli­ance and cloud­based backup so­lu­tions. As many as 72 per­cent of the or­ga­ni­za­tions sur­veyed said that they would change ven­dors if they could dou­ble the speed of their back­ups.

Bring vir­tu­al­iza­tion into the pic­ture, and the paradigm changes sig­nif­i­cantly. I of­ten speak with ad­min­is­tra­tors who look for ways to sim­ply pro­tect their vir­tu­al­ized as­sets for the pur­pose of full re­cov­ery in the event of dis­as­ter — i.e. their backup so­lu­tion is only work­ing to back up, but does not truly em­brace their vir­tual so­lu­tion. What if we start tak­ing the ap­proach of hav­ing backup soft­ware ac­tu­ally use vir­tu­al­iza­tion as a true ex­ten­sion of the re­cov­ery plan? Can we take vir­tu­al­iza­tion to be a re­source that can be lever­aged as the plat­form for re­cov­ery for both phys­i­cal and vir­tual servers alike?

To an­swer this, we need to first un­der­stand the cur­rent en­vi­ron­ment. Sure, the world is go­ing vir­tual in a strong way but this is not some­thing that is go­ing to hap­pen overnight. Al­though many early adopters have moved for­ward to be­come nearly 100 per­cent vir­tu­al­ized, most ad­min­is­tra­tors are still gov­ern­ing en­vi­ron­ments that are heav­ily

com­prised of both phys­i­cal and vir­tual server as­sets. As such, ad­min­is­tra­tors need a so­lu­tion that is not only pur­pose-built to work for their en­tire en­vi­ron­ment, but one that takes ad­van­tage of the vir­tual in­fra­struc­ture, specif­i­cally to al­low them to fur­ther lever­age their IT in­vest­ments.

Fur­ther­more, busi­nesses need to be able to not only lever­age their ex­ist­ing vir­tual in­fra­struc­ture for items in­clud­ing in­stant re­cov­ery of any phys­i­cal or vir­tual server that is pro­tected, but also lever­age the cloud to re­cover, test or mi­grate any vir­tual ma­chine that is in the en­vi­ron­ment. And for good mea­sure, imag­ine that when it is time to mi­grate that phys­i­cal server to a new vir­tual body, you sim­ply power on the vir­tual copy that was cre­ated and main­tained as part of the stan­dard backup of a phys­i­cal server! This greatly sim­pli­fies the com­plex­ity that ex­ists around backup to­day, and it means that in the near fu­ture, we can backup and re­cover our most im­por­tant in­for­ma­tion lit­er­ally at the touch of a but­ton.

To suc­cess­fully em­brace new tech­nolo­gies and in­crease con­fi­dence in backing up mis­sion- crit­i­cal data, or­ga­ni­za­tions are rec­om­mended to fol­low th­ese steps:

1 . Break the backup win­dow.

Elim­i­nate out-of-control backup win­dows by us­ing so­lu­tions that ac­cel­er­ate full back­ups mul­ti­fold, without sac­ri­fic­ing re­cov­ery time ei­ther. 2. Unite phys­i­cal and vir­tual back­ups.

Us­ing a sin­gle so­lu­tion for both en­vi­ron­ments will drive down op­er­at­ing costs, re­duce stor­age, and ac­cel­er­ate re­cov­er­ies.

3. Con­sol­i­date backup and re­cov­ery tools in a sin­gle ap­pli­ance.

In­te­grat­ing backup, dedu­pli­ca­tion, and stor­age in a sin­gle so­lu­tion will drive down op­er­at­ing costs and cap­i­tal ex­pense while sim­pli­fy­ing dayto- day op­er­a­tions in the data cen­ter and re­mote of­fices.

4. Fight in­fi­nite re­ten­tion.

Many busi­nesses have tremen­dous le­gal risk and cost ex­po­sure re­sult­ing from over­re­ten­tion of backup tapes. Iden­tify what in­for­ma­tion to archive and what to delete, based on rel­e­vance to le­gal dis­cov­ery or com­pli­ance cases. This re­duces the time and cost IT spends on eDis­cov­ery and elim­i­nates the need to keep back­ups for­ever.

5. Stop put­ting tapes on trucks.

Com­bine dedu­pli­ca­tion with dis­as­ter re­cov­ery to trans­mit data over the net­work from the pro­duc­tion site to the DR site in­stead of load­ing tapes onto trucks.

Ad­min­is­tra­tors need a so­lu­tion that is not only pur­pose­built to work for their en­tire en­vi­ron­ment, but one that takes ad­van­tage of the vir­tual in­fra­struc­ture specif­i­cally to al­low them to fur­ther lever­age their IT in­vest­ments

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.