消息队列 Kafka 流处理 Spark Streaming Flink Storm 实时计算(原流计算) EMR(开源流计算组件） 分析型数据存储 数据仓库： GreenPlum/Impala/Presto/Hive NoSQL：Hbase 数据仓库：MaxCompute/ Hologres/分析 型数据库 NoSQL:云数据库 Hbase 版/表格存储 MapReduce MaxCompute MR Apache Spark MaxCompute Spark 交互式分析 Impala Presto Hawk GreenPlum 等交互式分析 MaxCompute Lightning，提供只读的交互式查 询服务 图计算 Spark GraphX MaxCompute Spark GraphX MaxCompute

"<$apple-pay>" as ap rectangle "<$dinersclub>" as dc rectangle "<$discover>" as d rectangle "<$google-pay>" as gp rectangle "<$jcb>" as j rectangle "<$maestro>" as ma rectangle "<$mastercard>" as m rectangle "<$paypal>" "<$visaelectron>" as ve rectangle "..." as etc customer --> ap customer ---> dc customer --> d customer ---> gp customer --> j customer ---> ma customer --> m customer ---> p customer --> up customer ---> v customer

"<$apple-pay>" as ap rectangle "<$dinersclub>" as dc rectangle "<$discover>" as d rectangle "<$google-pay>" as gp rectangle "<$jcb>" as j rectangle "<$maestro>" as ma rectangle "<$mastercard>" as m rectangle "<$paypal>" "<$visaelectron>" as ve rectangle "..." as etc customer --> ap customer ---> dc customer --> d customer ---> gp customer --> j customer ---> ma customer --> m customer ---> p customer --> up customer ---> v customer

"<$apple-pay>" as ap rectangle "<$dinersclub>" as dc rectangle "<$discover>" as d rectangle "<$google-pay>" as gp rectangle "<$jcb>" as j rectangle "<$maestro>" as ma rectangle "<$mastercard>" as m rectangle "<$paypal>" "<$visaelectron>" as ve rectangle "..." as etc customer --> ap customer ---> dc customer --> d customer ---> gp customer --> j customer ---> ma customer --> m customer ---> p customer --> up customer ---> v customer

"<$apple-pay>" as ap rectangle "<$dinersclub>" as dc rectangle "<$discover>" as d rectangle "<$google-pay>" as gp rectangle "<$jcb>" as j rectangle "<$maestro>" as ma rectangle "<$mastercard>" as m rectangle "<$paypal>" "<$visaelectron>" as ve rectangle "..." as etc customer --> ap customer ---> dc customer --> d customer ---> gp customer --> j customer ---> ma customer --> m customer ---> p customer --> up customer ---> v customer

{S|D} rd,rs1,rs2 x2 sp Callee MULtiply R VMUL rd,rs1,rs2 DIVide R FDIV.{S|D} rd,rs1,rs2 x3 gp --- DIVide R VDIV rd,rs1,rs2 SQuare RooT R FSQRT.{S|D} rd,rs1 x4 tp --- SQuare RooT R VSQRT FRFLAGS rd sp Stack pointer MERGE R VMERGE rd,rs1,rs2 Swap Status Reg R FSCSR rd,rs1 gp Global pointer SELECT R VSELECT rd,rs1,rs2 Swap Rounding Mode R FSRM rd,rs1 tp Thread Capítulo 3 explica la convención de llamadas de RISC-V, la idea detrás de los varios punteros (sp, gp, tp, fp), registros guardados (en inglés, saved) (s0-s11), y Temporales (t0-t6) (La Figura 2.1 y Tabla