In　recent　years,　deep　learning　has　achieved　great　success　in　the　field　of　artificial　intelligence.　The　rise　of　deep　learning　has　accelerated　the　advent　of　the　intelligent　era.　However,　with　the　deepening　of　the　layers　of　deep　neural　networks　and　the　increasing　complexity　of　algorithm　models,　the　size　of　the　data　set　used　for　training　is　also　getting　larger　and　larger,　which　causes　the　time　required　for　training　to　continue　to　increase,　and　the　parallelization　of　deep　neural　networks　can　effectively　solve　this　problem.　Therefore,　the　research　on　the　parallelization　of　deep　learning　becomes　more　and　more　important.　This　article　summarizes　the　current　research　status　of　deep　learning　and　its　parallelization　implementation,　and　also　introduces　the　related　research　progress　in　spatial　data　processing　especially.　©　2020　IEEE.