AWR.ai：基于LLM的数据库性能新调查[LRN2899].pdf

《AWR.ai：基于LLM的数据库性能新调查[LRN2899].pdf》由会员分享，可在线阅读，更多相关《AWR.ai：基于LLM的数据库性能新调查[LRN2899].pdf（29页珍藏版）》请在三个皮匠报告上搜索。

1、 AWR.ai:New LLM-Powered Investigations into Database Performance LRN2899The following is intended to outline our general product direction.It is intended for information purposes only,and may not be incorporated into any contract.It is not a commitment to deliver any material,code,or functionality,a

2、nd should not be relied upon in making purchasing decisions.The development,release,timing,and pricing of any features or functionality described for Oracles products may change and remains at the sole discretion of Oracle Corporation.Safe harbor statement2Copyright 2025,Oracle and/or its affiliates

3、SpeakersShantanu JoshiSenior Director,Development,DB Observability&ManagementTrung TranArchitect,Development,DB Observability&ManagementSai PenumuruOracle ACE Director,President of AIOUG,Principal Director,Accenture(UK)Copyright 2025,Oracle and/or its affiliates3Agenda4Copyright 2025,Oracle and/or i

4、ts affiliates1.2.3.4.5.6.AWR and LLM:Background and challengesAWR.ai:GenAI powered solutionSQLPerf.ai:SQL performance AI analysisExtensible ADDMAsk EMConclusion5Copyright 2025,Oracle and/or its affiliatesRich Performance Statistics Available in AWRDetailed performance statistics collected at regular

5、 snapshot intervalsSystem performanceSystem and session-level activity(wait events)SQL workloadObject usageStored and managed in the database AWR viewsOver 150 views Sampled dataSpan Oracle architectureRAC,Exadata,Multi-tenant,In-MemoryHost and OS statistics5Copyright 2025,Oracle and/or its affiliat

6、es|Confidential:Internal/Restricted/Highly RestrictedIn-depth,comprehensive analysis of database activitySnapshotsADDMADDM ResultsAlertsAWR in-memorystatisticsWorkload Repository(AWR)SGAMMONEnterprise ManagerDatabaseEvolution of AWR for Performance Diagnosis 6Copyright 2025,Oracle and/or its affilia