BEGIN:VCALENDAR PRODID;X-RICAL-TZSOURCE=TZINFO:-//Calagator//EN CALSCALE:GREGORIAN X-WR-CALNAME:Calagator METHOD:PUBLISH VERSION:2.0 BEGIN:VTIMEZONE TZID;X-RICAL-TZSOURCE=TZINFO:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20260308T020000 RDATE:20260308T020000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 TZNAME:PDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT CREATED;VALUE=DATE-TIME:20260430T222445Z DTEND;TZID=America/Los_Angeles;VALUE=DATE-TIME:20260514T193000 DTSTART;TZID=America/Los_Angeles;VALUE=DATE-TIME:20260514T173000 DTSTAMP;VALUE=DATE-TIME:20260430T222445Z LAST-MODIFIED;VALUE=DATE-TIME:20260430T222445Z UID:http://calagator.org/events/1250482554 DESCRIPTION:Cloud Native PDX May: Change-driven Architecture and Serving Inference at Scale \;\n \;\nThis May is all about scaling\; scalin g your infrastructure management\, or scaling your LLM inference serving . Join us to find out about some open source tools to make managing larg e modern stacks easier. \;\n \;\nDate: Thursday\, May 14 \;\nTi me: 5:30–7:30 PM \;\nLocation: Reperio Health\, 4784 SE 17th Ave Suit e 120\, Portland\, OR \;\n \;\nRecording: Talks are typically reco rded (opt-in by speakers) \;\n \;\nA big thank you to Microsoft fo r sponsoring food &\; beverage\, and to Reperio Health\, our venue ho st. \;\n \;\nDrasi\, a new take on Change Driven Architectures: Am an Singh\, Microsoft \;\n \;\nModern cloud-native systems constant ly generate data changes\, and applications often need to react to them. Building change-driven solutions that respond to specific changes in di stributed data is challenging. This talk introduces Drasi\, a CNCF Sandb ox project that simplifies the design and implementation of change-drive n architectures using Graph Queries and pluggable components. For exampl e\, with Drasi you can declaratively write automation to detect and resp ond to running containers with newly identified vulnerabilities across p ods and deployments in a Kubernetes cluster. Join us for a walkthrough o f real-world use cases that show how Drasi’s approach brings structure a nd responsiveness to complex distributed environments - without writing custom code. \;\n \;\nDynamo: Large Scale Distributed Inference 3\;\nDavid Zeir\, Director\, DL System Software\, Nvidia \;\nNeelay S hah\, Distinguished Engineer\, Nvidia \;\n \;\nThis talk introduce s Dynamo\, NVIDIA's open-source Kubernetes-native distributed inference platform. We'll cover the problem space\, walk through Dynamo's architec ture — disaggregated prefill/decode\, KV-cache-aware routing\, and a tra nsport layer that moves KV blocks directly between GPUs — and dig into t he Kubernetes integration for scheduling\, autoscaling\, and graceful fa ilure handling. We'll close with a demo of Dynamo serving a real workloa d.\n\nTags: kubernetes\, containers\, cloud\, intrastructure\, devops\, LLM\, inference\n\nImported from: http://calagator.org/events/1250482554 URL:https://www.meetup.com/cloud-native-pdx/events/314391212 SUMMARY:Cloud Native May Meetup: Change-driven Architecture and Serving I nference at Scale LOCATION:Reperio Health: 4784 SE 17th Ave Suite 120\, Portland OR 97202 U SA SEQUENCE:1 END:VEVENT END:VCALENDAR