My 2 cents on this topic: SDB will support more than a few million triples, but you may have to hand tune the database tables using partitioning and such.
It will support more than a few million triples without tuning but insert and search times will increase dramatically. The problem is, as with any large scale relational installation, the trade off between indexes and partitioning to make table scans/indexes faster. If you can use TDB that is probably your best solution. I understand there are a couple of implementations that sit on top of Hadoop or No SQL storage platforms, perhaps one of those might work for you -- but you would need to spend some time evaluating them. In my mind the key questions are: 1) Will your data be updated on a regular basis? If so how frequently. 2) Why do you need to use a relational DB? I can understand if this is an external requirement such as we have to deploy on infrastructure X and it only supplies relational DB Y as a data store. 3) How much time do you have to learn how to best tune the DB to handle the SDB architecture? 4) How much time do you have to explore other implementations (e.g. TDB, Hadoop) and understand their limitations and impacts on your infrastructure/architecture? Claude On Wed, Oct 8, 2014 at 1:29 PM, Donald Patrichi < donald.patri...@gastechnology.org> wrote: > Hi Andy, > > Tell me, please, what would be the most recommended SDB or implementation > that could support more than a few million triplets? > > Thank you, > > Don Patrichi > > -----Original Message----- > From: Andy Seaborne [mailto:a...@apache.org] > Sent: Wednesday, October 08, 2014 5:51 AM > To: users@jena.apache.org > Subject: Re: SDB Databases Supported - SQL Server ver > > On 06/10/14 21:23, Donald Patrichi wrote: > > > > Hi, > > > > Tell me, please, does Jenna support > > > > > > 1) MS SQL 2008R2 > > > > 2) MS SQL 2012 > > > > 3) MS SQL 2014 > > > > > > as a SDB? > > > > Thank you, > > > > Don Patrichi > > Don, > > We don't recommend SDB for general use in new applications. It has only > limited useful for relatively small amounts of data (less than a few > million triples). There isn't anyone volunteering to test it at each > release. > > It was tested (some time ago) as documented on the website so if those > versions are compatible with what was tested, it should work. > > SDB only makes fairly basic use of SQL features. It is mainly the > detailed syntax that varies across systems. > > Hopefully, some users or reshippers will be able to supply up-to-date > information (you know who you are!). > > Andy > > ________________________________ > > This communication is for the use of the intended recipient only. It may > contain information that is privileged and confidential. If you are not the > intended recipient of this communication, the disclosure, copying, > distribution or use hereof is prohibited. If you have received this > communication in error, please advise me by return e-mail or by telephone > and then delete it immediately. > -- I like: Like Like - The likeliest place on the web <http://like-like.xenei.com> LinkedIn: http://www.linkedin.com/in/claudewarren
