Numerous applications, e.g., in the industrial sector, produce large amounts of time-series data, which must be stored and made available for distributed processing. While outsourcing data storage and processing to third-party service providers offers many benefits, it raises data privacy issues. In light of this problem, techniques have been proposed to share only encrypted data with the remote service provider, yet the capability to run meaningful queries over the data is preserved. However, timeseries data is typically compressed at the server to save space, which is not easily possible when dealing with encrypted data. Moreover, data must be compressed in such a way that queries can still be executed efficiently. As a first step in this direction, we present an approach that preserves data privacy, enables compression at the server, and supports querying of the stored data. Our evaluation using realworld time-series data shows that our compression mechanism can reduce the required space drastically. Moreover, the median running time of all considered queries increases marginally, implying that compression can be introduced without sacrificing performance of query execution.