Are Device Response Times a Neglected Risk of IoT?
Talkat Embedded Linux Conference + OpenIoT Summit North America 2017 · on Wednesday, 2017-02-22· at 16:20· Galleria South·
summary
Claim: connecting embedded devices to cloud services creates a real operational risk when device response times degrade and teams lack end-to-end visibility. Recommends built-in instrumentation (CPU, memory, app traces, crashes), a scalable backend and unified console that correlates device-to-cloud traces, and automated alerts so MTTR drops from days to hours. Cites fixes: RFID readers spamming GPS updates that caused backend overload and a firmware/backend patch, and consumer panels slowed by backend timeouts. Targets 1 second target, 100 ms ideal, 2–5% runtime budget, bytecode instrumentation.
AI-written summary · about
chapters (17)
- 0:00Intro
- 0:41About me: Balwinder Kaur
- 9:43Response Time Expectation
- 12:26Growing Complexity
- 13:56What's Becoming More Come?
- 15:51Latency Management
- 17:39Risk Assessment
- 18:22Risk Assesment - After Deployment
- 19:03Performance Impact
- 20:38Where can problems happen?
- 22:01Poor MTTR
- 24:38End to End Monitoring Solution
- 26:23Device Side Instrumentation
- 27:16Enterprise Grade Performance Monitoring Solution
- 32:34Guidelines to choosing an agent
- 34:54Best Practices - For the loT Cloud Service Provider
- 36:04Guidelines to choosing a Management Console
related talks
33mThe IoT for Real - Beyond Simplistic Thinking and Models
16m"Performance Testing: ..."
47mResolving performance issues in practice
43mThe Invisible Door: Reliability Gaps in the Front End
40mDon't Forget the Network: Your App is Slower Than You Think
50mCommon Strategy to Test IoT for Different Industries
30mAvoiding software fails: few metrics to improve application reliability
30mLightning in the Cloud: A Study of Transient Bottlenecks on n-Tier Web Application