Loading…
This event has ended. View the official site or create your own event + mobile app → Check it out
This event has ended. Create your own
View analytic
Tuesday, April 29 • 5:20pm - 6:10pm
Use-Case Power Management Optimization: Identifying & Tracking Key Power Indicators - Patrick Titiano, BayLibre

Sign up or log in to save this event to your list and see who's attending!


Embedded platforms performances keep increasing, quicker than power and thermal budgets, making Power Management (PM) an essential element. This presentation describes the importance of identifying and tracking KPI (Key Performance Indicators: C-States statistics / OPerating Points (OPP) Usage / Interrupts Rates / MPU Load / Memory Usage / Running Clocks / Power Status / ...) on a per use-case basis, to optimize power consumption at system-level (battery), but also to maintain temperature within limits. Following a short introduction to standard KPI, a methodology will be illustrated with practical examples extracted from critical use-cases (MP3 AV Playback, OS Idle, 3D Gaming) run on a real product. The need for dedicated and automated power management tooling (embedded power measurements, kernel instrumentation, user-space tooling) will be also highlighted.
Speakers
avatar for Patrick Titiano

Patrick Titiano

Embedded Power Management Engineer, BayLibre
Patrick Titiano has 14 years of engineering experience in embedded technologies. Patrick spent the last 9 years at Texas Instruments as a Power Management (PM) Expert. After 3 years as OMAP3 SW PM Architect (architecture definition, requirements, early prototyping), Patrick switched to a System PM Optimization role, in charge of tracking PM system issues (OMAP-Vox, OMAP[4-5]) for critical use-cases (OS Idle, MP3, Voice Call, Video Playback, 3D Gaming). Patrick also developed embedded...
Read More →

Tuesday April 29, 2014 5:20pm - 6:10pm
Salon III

Attendees (27)

Attendance numbers do not account for private attendees. Get there early!


Remove this from your schedule?
This session is full and you may not be able to get back in.
Remove
Cancel