Dumb and Dumber: how smart is your monitoring data?
•
2 likes•948 views
Big Data is all the rage right now. Everyone from a social media company to your grandmother's online knitting store is suddenly a big data shop. Application monitoring tools are no exception from this trend – they collect gigabytes of monitoring data from your application every minute. But most of this data is useless. It's dumb data. More data isn't better if the data you're getting from your tools isn't helping you do your job – in fact, it's a real problem. In this session AppDynamics will cover how to be smarter about collecting monitoring data, and how to ensure that the data we're collecting is intelligent.
Report
Share
Dumb and Dumber: how smart is your monitoring data?
- 1. DUMB & DUMBER How smart is your monitoring data
- 4. Big data is BAD monitoring Make the move to Smart Data
- 5. Why do monitoring tools exist?
- 6. What is Business Impact?
- 7. Enterprise Data Bloating @ Log Files Email monitoring £ Legal business
- 10. just what you need
- 12. The Digital Hoarders are Winning Data Storage System Performance Network Bandwidth 47% 37% 36% 40-60% STORAGE REQUIREMENT INCREASE YEARLY gartner
- 13. smart data defined actionable, intelligent, information
- 14. IS THIS PERSON PERFORMING WELL? Heart rate! 150bpm! Blood pressure! 165/100!
- 15. are we talking about this person? !15
- 16. OR this person?
- 17. What data could we collect? Attribute Person 1 Person 2 Heart Rate 150 150 Blood Pressure 180/90 180/90 Eye Color Blue Brown Blood Type O+ O- White Blood Cell Count 3.5 3.8 Hair Color Brown Blue Height 180cm 175cm Shoe size 11 10 Weight 180kg 94kg Current activity sitting skating
- 18. IS PERSON 2 PERFORMING WELL? Distance 10,000 metres! Time 12min 44sec! Record time: 12min 58sec baseline
- 19. New Olympic Record Jorrit Bergsma 10,000m winner
- 20. Traditional Monitoring Tools Are Misleading Resource Spikes May or May Not Cause Business Impact
- 21. Good luck !21
- 22. We’ve become addicted to metrics! !22
- 23. What do these charts tell us about application performance or business impact? !23
- 24. average response time with historical baseline
- 25. monitoring platforms should do the heavy lifting User & IT perspective Analytics Correlation Intelligent alerting Resolution path
- 26. Disaster Recovery (DR) vs Problem Recovery (PR)
- 29. Problem Recovery Planning STRATEGY ! MONITORING CONSIDERED CRITICAL ! END USER FOCUSSED ! CULTURAL
- 30. when a problem occurs… Do we have monitoring? ! What kind? ! What are we collecting? ! How long do we have history?
- 31. when a problem occurs… Network LOGS DATABASE INFRASTRUCTURE APPLICATION
- 33. smart data
- 34. MOST APM TOOLS ARE DUMB MOST INFRASTRUCTURE MONITORING TOOLS ARE DUMBER!
- 36. Who is using my application
- 37. End user experience - kpi's
- 38. business transaction - kpi's Oracle .NET Service ESB/MQ Apache Browser(s) Purchase Search Flight Flight Status Login Native Mobile App Network
- 39. application flow - kpi's
- 41. We need to save THE RIGHT data Analytics Aggregation Correlation Control Application Archive
- 42. Don’t be this person… !42
- 43. plan ahead anticipate needs intended purpose !43
- 44. QUANTIFY IMPACT $64,499 per min Application Revenue 10.1 s $11,987 per min Application Response time 100 ms Application Errors * Screenshot from US e-Commerce AppDynamics Customer
- 45. QUANTIFY BENEFIT code release 1 code release 2 code release 3 $58,237 per min $44,499 per min Application Revenue Application Response time 3.1 sec 1.9 s !45
- 46. EXAMPLES
- 48. over 1 million metrics per minute.
- 49. what did he say? 1. 2. 3.
- 50. 1. 2. 3. Big monitoring data slows down MTTR and drives up costs… a lot!
- 51. 1. 2. 3. Focusing on Problem Recovery will help you figure out your architecture, tools, & process… This saves time and money
- 52. 1. 2. 3. Don’t be a digital hoarder!!! Make the move to Smart Data
- 53. stand 1020 Free trial: appdynamics.com/freetrial
- 54. thankyou!