3355 - 基于 IBM Z 的 Red Hat OpenShift AI 实现可扩展且安全的 AI 推理.pdf

《3355 - 基于 IBM Z 的 Red Hat OpenShift AI 实现可扩展且安全的 AI 推理.pdf》由会员分享，可在线阅读，更多相关《3355 - 基于 IBM Z 的 Red Hat OpenShift AI 实现可扩展且安全的 AI 推理.pdf（31页珍藏版）》请在三个皮匠报告上搜索。

1、IBM TechXchange2025 conference 3355 Scalable and secure AI inferencing with RedHat OpenShift AI on IBM ZKhadija SouissiPrincipal solution architectDistinguished technical specialist AI on IBM Z and LinuxONE#IBMTechXchangeIBM Z TechXchange|2025 IBM Corporation1Scalable and Secure AI Inferencing with

2、Red Hat OpenShift AI on IBM ZScalable and Secure AI Inferencing with Red Hat OpenShift AI on IBM ZTable of ContentsThe AI landscape between theory and reality IntroductionSecure,scalable and sustainable AI on IBM Z and LinuxONE Red Hat OpenShift AI a new supported family memberHow can you use it?How

3、 to get startedIBM Z TechXchange|2025 IBM Corporation2Session objectivesLearn about how AI on IBM Z and LinuxOne can make the difference in your businessesExplore OpenShift AI on IBM Z and LinuxONE will open up new opportunities to innovate and embrace advanced AIIBM Z TechXchange|2025 IBM Corporati

4、on3AI capabilities are growing rapidlyAI that predictsPredictive AIAI that createsGenerative AIAI that chatsAI assistantsAI that does workAI agentsIBM Z TechXchange|2025 IBM Corporation4Operationalizing AI is still a challenging processSource:Gartner Peer Insights,Open Source AI for Enterprise surve

5、y,2023What is the average AI/ML timeline from idea to operationalizing the model?Half of respondents(50%)say their average AI/ML timeline from idea to operationalizing the model is 7-12 months.50%50%7-12 months15%15%3-6 months4%4%Unsure26%26%1 year or more5%5%Havent done this yet/Still in experiment

6、 phaseIBM Z TechXchange|2025 IBM Corporation50101Low-latency requirements for high volume workloads0202Integration of AI into existing environments 0303Model accuracy0404Compute requirements for GenAI0505Energy consumption,related costs,and environmental footprintBuilding AI is difficult;building AI