Data striping is a technique for increasing the throughput and reducing the response time of large accesses to a storage system. In this paper, we evaluate the effectiveness of applying stripping concepts to large tape libraries. Striping in tape libraries is being used with success for applications suceh as backup and scientific data collection, where data access patterns are stricly sequential. IN this paper, we evaluate striped perfomance for randomly distributed accesses to the tape library. We believe such operations will be characteristic of future tertiary strorage databases using large objects, such as on-line libraries and multimedia databases. Using an event-driven simulator, we show that striped large tape libraries perform poorly for this random workload becuse striping causes contention for the small number of readers and robot arms in these libraries. Increasing the number of readers results in effectiveness of striping may change as readers and robots imrpve in performance. We find that striping continoues to be an effective technique for increasing the throughput of large acesses if reader and robot performance scale at similar rates.