Evaluation of NoSQL prototypes for the CMS conditions database

With the restart of the LHC in 2015, the growth of the CMS conditions dataset will continue, therefore the need of a consistent and highly available access to the conditions makes a great cause to revisit different aspects of the current data storage solutions. We present a study of alternative data storage backends for the CMS Conditions Databases, evaluating some of the most popular NoSQL ones to support a key-value representation of the CMS conditions. In addition to the baseline performance comparison between a document store, a column-oriented, and a plain key-value store, the access layer for these databases in the CMS software was developed, in order to provide transparent support for these alternative data stores in the CMS context. The necessary changes in the software infrastructure, and in the modeling approaches are also discussed in this paper. We also discuss the validation phase, which plays a key role in the optimization of the different solutions with fine-tuning critical performance factors. In this paper we present a performance comparison between the NoSQL prototypes and the current Oracle solution, using the real dataset, accessed by the CMS software framework.