Berlin SPARQL Benchmark, evaluates the performance of query engines in e-commerce use case: searching products and navigating through related information. Randomized "query mixes" (of 25 queries each) are evaluated continuously through client application which communicates with the repository through SPARQL end-point. The benchmark allows evaluation with respect to changing sizes of the dataset and different number of simultaneous clients. Recent results from evaluation of few of the most popular engines are published here.

The table below provides information regarding query evaluation with growing number of simultaneous clients (1, 4, 8,16) against 25M and 100M versions of the BSBM dataset, which contain respectively 25M and 100M explicit statements.

Run	OWLIM version	Scale	1 client	4 clients	8 clients	16 clients	test environment
b31bs25os	BigOWLIM 3.1	25M	2,584	7,602	8,235	7,531	ontosol
b31bs100os	BigOWLIM 3.1	100M	1,181	2,998	2,341	2,367	ontosol

Performance highlights:

• BigOWLIM scales-up well given multiple cores. In the 25M run, 4 clients deliver 2.5 higher throughput. In the 100M run, 8 clients process 3.2 more query mixes, which is close to the theoretical gains, given the quad-core CPU of the desktop machine used in runs b31bs25os and b31bs100os.
• The performance of OWLIM matches the best results of native triple stores published at the BSBM site..