NonStop SQL is designed to run effectively on parallel computers, adding functionality for distributed data, distributed execution, and distributed transactions.
First released in 1987, a second version in 19892 added the ability to run queries in parallel, and the product became fairly famous for being one of the few systems that scales almost linearly with the number of processors in the machine: adding a second CPU to an existing NonStop SQL server almost exactly doubled its performance.
The second version added /MP to its name, for Massively Parallel. A third version, NonStop SQL/MX, created a product that was more ANSI SQL compliant than its predecessor. NonStop SQL/MX has shipped on the NonStop platform since 2002, and can access tables created by NonStop SQL/MP, although only "Native SQL/MX tables" offer ANSI compliance and many "Oracle-like" enhancements. The HP Neoview business intelligence platform was built using NonStop SQL as its origins. NonStop SQL/MX is HP's only OLTP database product.
Parts of the Neoview code base were open sourced in 2014 under the name Trafodion, which is now a top-level Apache project.
Tandem Performance Group (1988). "A benchmark of non-stop SQL on the debit credit transaction". Proceedings of the 1988 ACM SIGMOD international conference on Management of data - SIGMOD '88. Vol. 17. Chicago, Illinois, United States: ACM Press. pp. 337–341. doi:10.1145/50202.50243. ISBN 9780897912686. S2CID 15257584. 9780897912686 ↩
Tandem Database Group (1989). "NonStop SQL: A Distributed, High-Performance, HighAvailability Implementation of SQL". In D. Gawlick (ed.). Proceedings of 2nd High Performance Transaction Processing Workshop. Springer Verlag. pp. 60–104. ISBN 3-540-51085-0. 3-540-51085-0 ↩