Big data management is a key enabling factor for enterprises that want to compete in the global market. Data coming from enterprise production processes, if properly analyzed, can support a boost in the enterprise management and optimization, guaranteeing faster processes, better customer management, and lower overheads/costs. Guaranteeing a proper Big Data analytics is the Holy Grail of Big Data, often opposed by the difficulty of evaluating the precision of the Big Data analytics results. This problem is even worse when Big Data analytics are provided as a service in the cloud, and must comply with both users' requirements and laws. Recently, the Big Data community has started noticing that there is the need to complete Big Data with assurance techniques proving the correct behavior of Big Data analytics and management. This paper provides an assurance solution based on Service-Level Agreements (Slas), where stochastic formal models are used to define, negotiate, and monitor Slas targeting the behavior of Big Data platforms and analytics delivered as a service.