Video Screencast Help
Netting Out NetBackup

The New Sheriff in Town

Created: 10 Jan 2008 • Updated: 22 Jan 2013 • 1 comment
TimBurlowski's picture
0 0 Votes
Login to vote
 
 

Hi, my name is Tim Burlowski. I am a technical product manager and I work in the NetBackup product management group. If you are wondering, I am certainly not the "new sheriff."

I've worked at Symantec for almost 10 years. The last ten years have been a long strange trip to say the least.

When I first started working on NetBackup the initial configuration was not for the faint of heart. One of the first tasks I was given was performing some release testing for NetBackup 3.11. I was given a sheet of paper with directions for remotely mounting a tape and installing client software from that tape. I had to find an AIX server, mount the 8mm tape and then install a client across the network from a Solaris system. I recall that it all worked by the way, but it would be generous and kind to describe the process as anything but cryptic. At that time you could actually order NetBackup on tape. It made sense since most everyone was doing backup to tape, everyone had a tape drive. My previous job was related to software and hardware built for the desktop publishing company where everyone had CD-ROM drives so I was surprised that people actually installed software from tape — I had a lot to learn.

At that time SAN attached storage was rare and NetBackup didn't share drives on a SAN even if you asked nice. Configuring a tape drive often meant modifying system files and sometimes even creating operating system devices from scratch. We were still working on porting NetBackup server to an upstart operating system from Microsoft.

Fast forward a few years and we really targeted usability. We added "wizards" for setting up tape drives, policies, storage units and other configuration items. At the time there were people convinced that wizards were a very bad idea. They argued " … if anyone could easily set up a device for NetBackup to use, maybe they would do it wrong." Logic prevailed; we added the wizards and we made it harder to configure the product wrong.

Since then we have added more features to NetBackup than I could possibly list — VMWare VCB integration, hot catalog backup, inline copy, snapshots, disk staging, SSO, San Client, Vault, Storage Lifecycle Policies — just to name a few. We've added usability enhancements incrementally from time to time to support new features, but our basic user interfaces have been the same for quite a long time.

This brings me back to the "new sheriff in town". A few months ago my VP, Matt Kixmoeller, called me into his office and pointed at a new org chart displayed on his monitor. Once I verified that I was still on the org chart, I noticed I was in a new group. The new group was entitled "The Experience". The charter was simple, make the customer experience better. It is clear to a lot of experts, including industry analysts, that we have the vision and the features that everyone wants — now we need to make NetBackup easier to deploy, use and manage. My team’s charter is to make "the experience" better. The new sheriff in town is not a person — it is the idea that making our products simpler to use and more manageable, even for very large scale deployments, is one of the most important things we can do to delight our customers. It’s a big effort and to be successful it’s going to have to be a collaborative effort involving our customers and partners.

We’ll be looking for your input soon.

Message Edited by Turlas on 04-15-2008 10:50 AM

Comments 1 CommentJump to latest comment

IanF 2's picture

Tim,

Having just come from the customer side of things, there are a few "improvements" that would be very useful, which should not be that had to achieve.

Firstly, having the storage unit list in alphabetical order when selecting them in new policies, storage unit groups, in-line taoe coipy etc woudl be very useful. If you have a large number of different storage units (we had about 80 of them), then having to slowly scroll through the full list to find the one you needed was very time consuming.

The second thing would be the ability to fully cluster NetBackup. Currently, you are unable to cluster things such as bp.conf and nblog.conf. This means that if changes are made using the Admin console, then these are not reflected to the other nodes within the cluster, and hence a fail-over means that the NBU config could now be different. There does not appear to be any Symantec supplied ability to perform this synchronisation between these files (and some of the touch files that may need to be created). This caused several issues in the environment that I have just come from (particularly when going into DR)

0
Login to vote