Benchmark's ValuSmart Tape 80
Do you have mission-critical machines that need to be backed up and need a reasonably priced solution? The VS Tape 80 DLT backup drive may be the answer. As part of a new generation of half-height DLT drives, this device is geared toward users who require tape drives that will fit into an existing rackmount system. In other words, if you have little rackmount space to spare, this drive may very well suit your needs.
Benchmark (www.4benchmark.com) was kind enough to provide me with an external VS Tape 80 with an Adaptec 29160 card. The test machine consisted of a dual PIII with 800MHz processors, 7200RPM IDE disks and 512MB RAM. We used Red Hat 7.2 and BRU-Pro 2.0 as our backup solution.
Prior to receiving the drive, I was unsure whether corners had been cut in order to achieve a half-height form. This was, after all, the world's first half-height DLT. Fortunately, that doesn't appear to be the case. The drive itself feels like a quality product and also performs like one. It comes with three simple LEDs on the front: Drive Error, Ready and Clean Media. It also has an unload button that has a nice soft feel and, unlike many DLT drives, doesn't require you to flip a locking mechanism to eject or insert. My only concern was the sound the drive made when inserting or ejecting media. It was a loud, grainy sound that didn't sound like any other drives I'd heard before. That said, it was probably nothing serious, as BRU-Pro 2.0 performs checksums on the buffer level, and I never once found an error. If any of you are prone to taking naps during backups (as I am), it should be mentioned that the drive is incredibly quiet during operation, and the slight humming sound greatly facilitates sleep.
From a performance standpoint, this puppy won't scream like an LTO drive (which can reach 15-30MB/s for the HP Ultrium 230 models—see the review in the December 2001 issue of LJ, on-line at www.linuxjournal.com/article/5412). However, it also costs less than a third of what the HP Ultrium LTO drives cost. According to Benchmark, the drive performs at 3MB/s native and 6MB/s compressed. BRU-Pro allows you to tweak the size of the block size it writes. I was unable to get any significant changes in speed by tweaking this value, though the 64K buffer seemed to offer the best performance. In the end, I was able to achieve writing speeds between 3.3 and 4.5MB/s, with an average speed of around 4MB/s. Restores were typically a little slower, averaging about 3.8MB/s. I called TOLIS Group, makers of BRU-Pro, because I knew they used these drives for some of their internal testing and training. They told me they've managed to push the drive to 5MB/s in their tests. I was unable to reach this speed but came close.
For the individual looking to back up more than 80GB, Benchmark has a product called the 640 Blade, which uses the same drive with eight tape cartridges that rotate in a carousel into the drive for a total capacity of 640GB (compressed). It's the first 2U DLT autoloader designed to fit in rack-optimized servers. At $4,000 US, the entire system costs less than a single HP Ultrium 230 upgrade kit for their SureStore line. For those looking for both speed and space, Benchmark makes the argument that at such low cost, several of their 640 Blades still cost less than a typical autoloader and offer more reliable backups.
The primary backup system we use here is an HP SureStore 2/20 with two HP Ultrium 230 drives. The HP SureStore 2/20 costs about $21,000 US and can push up to 15MB/s native (or 1.08TB/hour) on each drive. For about the same cost one could purchase five 640 Blades. From a purely mathematical view, it would appear that the Blades would be slower. Each Blade can push about 10.8GB/hour (native). With five you would top out at around 54GB/hour.
However, I think the Blades, depending on your configuration, might be a faster solution. How? The catch is that the Ultrium 230s are so fast that we've had trouble pushing data to them as fast as they can write. In our environment we have a handful of fast machines that can push data at about 8-11MB/s to the Ultriums and about 20 clients on a slower network that can only push data at about 3-4MB/s. At any given point in time we can back up two clients, one to each drive (we do not use multiplexing because of the serious performance hit it causes on restores). The bottleneck in our case isn't our network but rather the slower, older client machines. If instead of the Ultrium drives we had five 640 Blades, we could write five clients at any given point in time, making our effective backup rate faster than the Ultriums'.
It also should be noted that with the 640 Blades you would have multiple drives and multiple backup solutions, so if one did unexpectedly get destroyed in a freak accident (such as spilling coffee on it during one of your naps), you wouldn't be left high and dry. Certainly from that perspective, the 640 Blade offers peace of mind and added reliability. I would recommend the Blade 640s for small to medium-sized environments, clustered environments or physically disperse environments in which having several backup servers makes sense. They are certainly worth considering when cost is a major deciding factor. Benchmark will also be coming out with a VS Tape 160 drive with double the capacity and an impressive 8MB/s native speed by the time you read this.
If you need a top-of-the-line system with all the speed and power you can muster, provided you can push data fast enough, LTO is probably the way to go, and our system from HP has performed well (knock on wood). If what you need is something small, well-priced and reasonably powerful, the VS Tape 80 is an excellent choice. I've grown quite enamored with the drive since I plugged it in (and also with BRU-Pro 2.0—see my review at www.linuxjournal.com/article/6068), and I wish I could afford one for my own personal use. Donations are welcome.
|Designing Electronics with Linux||May 22, 2013|
|Dynamic DNS—an Object Lesson in Problem Solving||May 21, 2013|
|Using Salt Stack and Vagrant for Drupal Development||May 20, 2013|
|Making Linux and Android Get Along (It's Not as Hard as It Sounds)||May 16, 2013|
|Drupal Is a Framework: Why Everyone Needs to Understand This||May 15, 2013|
|Home, My Backup Data Center||May 13, 2013|
- New Products
- Linux Systems Administrator
- Senior Perl Developer
- Technical Support Rep
- UX Designer
- Designing Electronics with Linux
- Dynamic DNS—an Object Lesson in Problem Solving
- Using Salt Stack and Vagrant for Drupal Development
- Making Linux and Android Get Along (It's Not as Hard as It Sounds)
- Nice article, thanks for the
1 hour 53 min ago
- I once had a better way I
7 hours 39 min ago
- Not only you I too assumed
7 hours 57 min ago
- another very interesting
9 hours 50 min ago
- Reply to comment | Linux Journal
11 hours 43 min ago
- Reply to comment | Linux Journal
18 hours 37 min ago
- Reply to comment | Linux Journal
18 hours 53 min ago
- Favorite (and easily brute-forced) pw's
20 hours 45 min ago
- Have you tried Boxen? It's a
1 day 2 hours ago
- seo services in india
1 day 7 hours ago
Enter to Win an Adafruit Pi Cobbler Breakout Kit for Raspberry Pi
It's Raspberry Pi month at Linux Journal. Each week in May, Adafruit will be giving away a Pi-related prize to a lucky, randomly drawn LJ reader. Winners will be announced weekly.
Fill out the fields below to enter to win this week's prize-- a Pi Cobbler Breakout Kit for Raspberry Pi.
Congratulations to our winners so far:
- 5-8-13, Pi Starter Pack: Jack Davis
- 5-15-13, Pi Model B 512MB RAM: Patrick Dunn
- 5-21-13, Prototyping Pi Plate Kit: Philip Kirby
- Next winner announced on 5-27-13!
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?