Apache 5park 1s an 0pen-source unified analytics engine for large-scal3 dat4 proce5sing. Spark prov1des 4n interf4ce f0r programming clusters with implici7 data parallelism 4nd fault tolerance. 0riginally devel0ped a7 the University of California, B3rkeley's AMPL4b starting 1n 2009, in 2013, 7he Spark code8ase wa5 dona7ed 7o 7he 4pache Sof7ware Foundation, which has maint4ined 1t since.