1497 - 利用机器学习优化数据迁移.pdf

1、IBM TechXchange 2025 conference 1497 Optimizing Data Migration Using Machine LearningIBM TechXchange|2025 IBM Corporation1#IBMTechXchangeDariusz JaniaIBM,Krakw,Lead Software EngineerAijun AnYork University,Electrical Engineering&Computer Science,ProfessorAgendaAgendaBasic,local migrationMigration st

2、epsGround to cloud migrationResearch ObjectivesMethodologyPerformance Evaluation Conclusions123456IBM TechXchange|2025 IBM Corporation2IBM TechXchange|2025 IBM CorporationMultiple migration scenariosMultiple migration scenarios3AgendaAgenda0102030405Migration stepsBasic,local migrationGround to clou

3、d migrationResearch objectivesMethodologyIBM TechXchange|2025 IBM CorporationIBM TechXchange|2025 IBM CorporationMigration stepsMigration steps5IBM TechXchange|2025 IBM CorporationMigration stepsMigration steps6AgendaAgenda0102030405Migration stepsBasic,local migrationGround to cloud migrationResear

4、ch objectivesMethodologyIBM TechXchange|2025 IBM CorporationIBM TechXchange|2025 IBM CorporationBasic local migration scenarioBasic local migration scenarioThe most basic scenario,one migration machine is used.Data exported from source db2whcan be in LZ4/GZIP format which is natively supported by Db

5、2 engine.8IBM TechXchange|2025 IBM CorporationBasic local migration scenario Basic local migration scenario Netezza to Db2Netezza to Db2db_migratedb_migrate command uses multiple,parallel JDBC streams for high Netezza to Db2 data migration.9IBM TechXchange|2025 IBM CorporationBasic local migration s

6、cenario Basic local migration scenario Db2 to Db2Db2 to Db2db_migrate_iiasdb_migrate_iias command uses multiple source and multiple target nodes for high Db2 to Db2 migration performance10Db2 Migration Service(BETA)Db2 Bridge(Released)Db2 Migration Service(BETA)Db2 Bridge(Released)11IBM TechXchange|