On Saturday 07 April 2007 22:56, Ilya Volvovski wrote: > Hi, > > My name is Ilya Volvovski and I am a software architect for a company > called Cleversafe (www.cleversafe.org <http://www.cleversafe.org/> and > www.cleversafe.com <http://www.cleversafe.com/> ) > > I address this discussion group to gauge interest in our proposition > from people who make living by making sure that data is not lost. > Cleversafe mission is to offer a product that would provide unlimited, > secure, scalable, reliable data storage facility. We want to move > storage from home tape devices, CD/DVD drives to Internet.
This is something that the Bacula project had planned to do in 2000. However there are still problems of Internet reliability for high volumes (dropped lines), bandwidth, and security concerns that poses major problems. > > If you look at our website, you would learn that are patented approach > allows to achieve very high level of reliability with reasonably small > data blowup (2.0 now but our next release would make it configurable: > 30% data overhead still would still guarantee high reliability). Data > could be optionally encrypted, compressed and then dispersed into > several locations. Cleversafe Storage Grid is also self-healing: if a > few slices are lost due to hardware problems, they could be recreated by > the grid. Well, a problem here is that you mentioned the word "patent", and patents don't generally go very well with GPL software, so I would like to hear more about how you are using the patent. If it is a question of making Bacula interface to a patented device -- I don't see a big problem -- after all, Bacula interfaces to computers, disk drives, tape drives, ... all of which are patented devices (or have patented components). > > Our software would give means to store data, but we are not backup > experts. We want the existing backup solutions be used as before but > enabled to store data on Cleversafe Storage Grid. Being an open source > project we first looked at open source solutions and this lead us to > Bacula. As mentioned above, I would like to understand your patent vs you Open Source position a bit more. Also, I would be interested in your overall plan -- e.g. are you proposing it to all backup solutions? or are you just targeting certain ones? > > Cleversafe has no expertise in using Bacula. However we were able to > install and we performed a very trivial integration with NO changes to > the existing Bacula codebase. We confirmed that if we configure Bacula > with a file type device, which happen to be our grid, we could use > Bacula to store data on it. Nice. It sounds like you have made your grid compatible with a filesystem. > > I have several questions to the group: > - how interesting this idea sounds to system administrators and people > who routinely perform backups It would depend on what advantages it offers. I haven't yet looked at your site, and it wasn't totally clear from your email. Most often, it is a question of price and reliability. Most big corporations who need the ultimate reliability want to maintain full control of their data on "internal" secured networks. In that sense, a solution that is integrated with a backup program such as Bacula that can be configured to use "their" servers would be very interesting. > - how interesting this idea sounds to software developers who work in > the backup/storage world. Is there potential to change/enhance this > software in order to target grid market I would like Bacula to be able to address all big markets. I am totally ignorant of the grid market, so I cannot really say. We certainly are very open to discuss this -- keeping in mind that we are an Open Source project and patents could be a problem. > - are there any caveats of using Bacula with file device? None that I know of. The only major feature users have requested is a way for Bacula to automatically delete or truncate a file Volume when it is "purged" so that the physical space is freed up. Currently the Volume is truncated only when it is going to be reused. This keeps the data there much longer (as long as it exists in the Volume, it can be recovered), but it also uses space. > As I > understand that it is not mainstream usage; Bacula is primarily geared > towards tape devices. I am not sure what percentage of users use disk vs tape, but IMO, the current implementation is equally functional for both -- neither is favored over the other. I wouldn't say that Bacula is primarily geared towards tape devices. They are harder to implement, and they do have more operational problems, so much of the email pertains to them. It is clear that there is a huge battle going on between disk and tape manufacturers to capture the backup market. Disks are evolving in size and reliabilty quickly but it seems to me that tapes are evolving even faster. I don't think I would place all my bets on either one of them. Both offer good solutions for particular markets, which are currently growing at a very fast pace. Regards, Kern ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
