BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:PDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:PST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260618T032902Z UID:E6FAEAB1-E029-42BF-AE10-8EF3E65B09D5 DTSTART;TZID=America/Los_Angeles:20260617T190000 DTEND;TZID=America/Los_Angeles:20260617T200000 DESCRIPTION:The San Francisco Bay Area chapter of the IEEE Computer Society invites to our free and open Virtual Tech Talks (no IEEE membership requi red):\n\nSpeaker: Lakshmana Rao Koppada ([Connect on LinkedIn](https://www .google.com/url?q=https://www.linkedin.com/in/lakshmana-koppada/&sa=D&sour ce=calendar&ust=1781490549069271&usg=AOvVaw11bsPeQi6ednCxDVhnMelu))\n\nTit le: GPU-Free Real-Time Utility Asset Anomaly Detection with IBM Granite TS Pulse-R1\n\nAbstract: Utility enterprises operate geographically distribut ed critical assets—such as pumps\, tanks\, flow meters\, and substations that demand continuous health monitoring to avert service disruptions\, e nvironmental hazards\, and substantial economic losses. Conventional centr alized cloud-based Enterprise Asset Management systems incur high latency\ , excessive bandwidth consumption\, and reliance on GPU acceleration\, the reby impeding real-time response in connectivity-constrained environments. This paper introduces a GPU-free edge-to-cloud architecture for real-time anomaly detection that exploits the compact IBM Granite TSPulse-R1 time-s eries foundation model (approximately 1 million parameters). The framework executes lightweight CPU-only inference on edge gateways and transmits on ly concise anomaly summaries via Message Queuing Telemetry Transport (MQTT )\, realizing over 99.9% bandwidth reduction relative to raw data transfer . A novel sensor-weighted reconstruction error scoring mechanism prioritiz es the most discriminative sensors\, enhancing zero-shot multivariate dete ction performance. Rigorous evaluation on the Water Treatment dataset\, a real-world industrial control system benchmark featuring 51 sensors and a 12.53% anomalous timestep ratio\, demonstrates an average inference latenc y of 710.3 ms per 512-timestep chunk\, a ROC-AUC of 0.717\, and tolerance- adjusted recall exceeding 0.945 at ±20 timesteps. Scalability analysis re veals near-linear latency growth with increasing sensor counts and chunk s izes\, confirming feasibility on resource-constrained edge devices. The pr oposed architecture delivers a scalable\, cost-effective\, and resilient p redictive maintenance solution for critical infrastructure\, providing a p ractical GPU-independent alternative to traditional cloud-centric paradigm s while supporting robust operation in distributed utility networks.\n\nBi o: Lakshmana Rao Koppada (IETE Fellow\, Senior Member IEEE\, Professional Member of BCS) is a Digital Transformation Leader and Technical Architect at PwC with over 16 years of experience delivering large-scale enterprise modernization programs. Specializing in IBM Maximo\, cloud platforms\, and Red Hat OpenShift\, he integrates AI/ML\, IoT\, and predictive analytics to improve asset reliability and operational efficiency. He has led global digital transformation initiatives for Fortune 500 organizations across u tilities\, pharmaceuticals\, manufacturing\, energy\, and data center indu stries. A published researcher and certified IBM Maximo/MAS architect\, he actively contributes to international technology conferences and innovati on initiatives in AI-driven enterprise systems.\n\nSpeaker(s): Ram Sekhar Bodala\n\nVirtual: https://events.vtools.ieee.org/m/563073 LOCATION:Virtual: https://events.vtools.ieee.org/m/563073 ORGANIZER:ruben.glatt@ieee.org SEQUENCE:14 SUMMARY:Tech Talk: GPU-Free Real-Time Utility Asset Anomaly Detection URL;VALUE=URI:https://events.vtools.ieee.org/m/563073 X-ALT-DESC:Description: <br /><p>The San Francisco Bay Area chapter of the IEEE Computer Society invites to our free and open Virtual Tech Talks (no IEEE membership required):</p>\n<p><strong>Speaker:</strong><strong>&nbsp\ ;</strong>Lakshmana Rao Koppada&nbsp\;&nbsp\;(<a href="https://www.google. com/url?q=https://www.linkedin.com/in/lakshmana-koppada/&amp\;sa=D&amp\;so urce=calendar&amp\;ust=1781490549069271&amp\;usg=AOvVaw11bsPeQi6ednCxDVhnM elu" target="_blank" rel="noopener"><u>Connect on LinkedIn</u></a>)</p>\n< p><strong>Title:&nbsp\;</strong>GPU-Free Real-Time Utility Asset Anomaly D etection with IBM Granite TSPulse-R1</p>\n<p><strong>Abstract:</strong>&nb sp\;Utility enterprises operate geographically distributed critical assets &mdash\;such as pumps\, tanks\, flow meters\, and substations that demand continuous health monitoring to avert service disruptions\, environmental hazards\, and substantial economic losses. Conventional centralized cloud- based Enterprise Asset Management systems incur high latency\, excessive b andwidth consumption\, and reliance on GPU acceleration\, thereby impeding real-time response in connectivity-constrained environments. This paper i ntroduces a GPU-free edge-to-cloud architecture for real-time anomaly dete ction that exploits the compact IBM Granite TSPulse-R1 time-series foundat ion model (approximately 1 million parameters). The framework executes lig htweight CPU-only inference on edge gateways and transmits only concise an omaly summaries via Message Queuing Telemetry Transport (MQTT)\, realizing over 99.9% bandwidth reduction relative to raw data transfer. A novel sen sor-weighted reconstruction error scoring mechanism prioritizes the most d iscriminative sensors\, enhancing zero-shot multivariate detection perform ance. Rigorous evaluation on the Water Treatment dataset\, a real-world in dustrial control system benchmark featuring 51 sensors and a 12.53% anomal ous timestep ratio\, demonstrates an average inference latency of 710.3 ms per 512-timestep chunk\, a ROC-AUC of 0.717\, and tolerance-adjusted reca ll exceeding 0.945 at &plusmn\;20 timesteps. Scalability analysis reveals near-linear latency growth with increasing sensor counts and chunk sizes\, confirming feasibility on resource-constrained edge devices. The proposed architecture delivers a scalable\, cost-effective\, and resilient predict ive maintenance solution for critical infrastructure\, providing a practic al GPU-independent alternative to traditional cloud-centric paradigms whil e supporting robust operation in distributed utility networks.</p>\n<p><st rong>Bio:</strong> Lakshmana Rao Koppada (IETE Fellow\, Senior Member IEEE \, Professional Member of BCS) is a Digital Transformation Leader and Tech nical Architect at PwC with over 16 years of experience delivering large-s cale enterprise modernization programs. Specializing in IBM Maximo\, cloud platforms\, and Red Hat OpenShift\, he integrates AI/ML\, IoT\, and predi ctive analytics to improve asset reliability and operational efficiency. H e has led global digital transformation initiatives for Fortune 500 organi zations across utilities\, pharmaceuticals\, manufacturing\, energy\, and data center industries. A published researcher and certified IBM Maximo/MA S architect\, he actively contributes to international technology conferen ces and innovation initiatives in AI-driven enterprise systems.</p> END:VEVENT END:VCALENDAR