NoSQL benchmarking

OData support
Supervisor:
Dr. Gajdos Sándor
Department of Telecommunications and Media Informatics

The needs of the market were fulfilled by relational database management systems, but nowadays the database systems have to face new challenges. The consistent data and high availability are not anymore untouchable, because sometimes the right decision proves to be the prioritization of small latency and high throughput. These and similar functions are provided by NoSQL systems, which widely-spread at the end of the last decade. For a long time the decision-makers in business were able to base on reliable performance tests at the purchase of a new hardware or software, but this is nowadays not true. The systems that fulfill the distorted requirements provide false, not relevant measurement results in some cases, which can cause the companies considerable costs or even show a plan unreachable, even though it can be achievable.

In my thesis I study the methodology of measuring distributed database systems and I also present this in several measurements, products and hardware. I hope that my results will support the work of both the database systems’ creators and both the business decision-makers.

Downloads

Please sign in to download the files of this thesis.