AppDynamics Unified Monitoring

Slide 1

Slide 1 text

Uniﬁed Monitoring with AppDynamics Dus$n Whi*le @AppDynamics

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

52% of Fortune 500 firms since 2000 are gone

Slide 4

Slide 4 text

Slide 5

Slide 5 text

No content

Slide 6

Slide 6 text

No content

Slide 7

Slide 7 text

Up7me is cri7cal for enterprises and consumers

Slide 8

Slide 8 text

Performance impacts the bo=om line

Slide 9

Slide 9 text

How fast is fast enough? § Performance is key to a great user experience -‐ Under 100ms is perceived as reac$ng instantaneously -‐ A 100ms to 300ms delay is percep$ble -‐ 1 second is about the limit for the user's ﬂow of thought to stay uninterrupted -‐ Users expect a site to load in 2 seconds -‐ ADer 3 seconds, 40% will abandon your site. -‐ 10 seconds is about the limit for keeping the user's a*en$on § Modern applica7ons spend more 7me in the browser than on the server-‐side

Slide 10

Slide 10 text

Who cares about performance?

Slide 11

Slide 11 text

No content

Slide 12

Slide 12 text

No content

Slide 13

Slide 13 text

No content

Slide 14

Slide 14 text

How many enterprise monitoring products would you es7mate your IT org owns?

Slide 15

Slide 15 text

The war room response team

Slide 16

Slide 16 text

The problems with monitoring tools § Root cause isola7on is elusive as monitoring lives in silos -‐ Infrastructure -‐ Hardware + Logs + Network + Storage + Containers + VMs -‐ Applica$on -‐ Load Balancers + Web Servers + App Servers + File Servers -‐ Databases + Caches + Queues + Third party services -‐ End Users -‐ CDN -‐ Web + Mobile § Metrics lack the context of impact

Slide 17

Slide 17 text

Monitoring lacks the business context

Slide 18

Slide 18 text

The struggle of modern monitoring § Organiza7ons focus on availability + raw metrics and not end user experience / impact § Complex apps built on micro-‐services in containers living in elas7c cloud environments § Too many graphs from too many metrics -‐ Understanding the signal from the noise is diﬃcult -‐ No topology awareness. No transac$onal visibility. No root cause. § Alert storming with too many false alarms -‐ Aler$ng is based on sta$c thresholds — lacks intelligent anomaly detec$on + correla$on -‐ Lack of historical context or rela$onships between metrics and events § A single pane of glass across the performance stakeholders -‐ Not able to quan$fy the impact of performance degrada$on — Not self-‐service

Slide 19

Slide 19 text

No content

Slide 20

Slide 20 text

Context is king: Uniﬁed Monitoring

Slide 21

Slide 21 text

Breaking down the silos

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

No content

Slide 24

Slide 24 text

No content

Slide 25

Slide 25 text

Situa7on-‐aware data and views Web Ops App Owner Server Admin DBA IT Ops

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

No content

Slide 29

Slide 29 text

No content

Slide 30

Slide 30 text

No content

Slide 31

Slide 31 text

Monitor the end user experience § Real User Monitoring vs Synthe7c Monitoring -‐ Synthe$c tests provide 24/7 assurance -‐ RUM provides insights into actual users -‐ Mobile device segmenta$on -‐ Unexpected behavior/trends § Real User Monitoring -‐ Naviga$on Timing API -‐ Resource Timing API -‐ User Timing API -‐ Javascript Errors

Slide 32

Slide 32 text

Metrics + logs help, but intelligence is be=er

Slide 33

Slide 33 text

No content

Slide 34

Slide 34 text

No content

Slide 35

Slide 35 text

No content

Slide 36

Slide 36 text

Moving from reac7ve to proac7ve § Resolving before the red = ﬁxing in the yellow -‐ Automa$c runbook automa$on integrates with your devops stack § Intelligent anomaly detec7on across end-‐user, applica7on, database, server metrics -‐ Automa$cally calculates dynamic baselines for all of your metrics, which, based on actual usage, deﬁne what is "normal" for each metric -‐ Smart aler$ng based on any devia$on from the baselines § Understand trends and pa=erns in failures -‐ automa7cally learn from the past -‐ Understand what are the most impacaul issues to resolve -‐ Ocen $mes external services are the root cause with limited visibility -‐ Enforce SLAs

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

Moving from reac7ve to proac7ve -‐ Automa7c discovery of environment and applica7on changes -‐ New APIs, transac$ons, services, clouds § Leverage analy7cs to be smarter about using the data you already have -‐ System Logs, Metrics from events and infrastructure stats -‐ Transac$ons with request parameters + User state from cookies/sessions § Performance monitoring isn’t just about the tech -‐ Visibility into the impact of business -‐ aler$ng when revenue is down

Slide 39

Slide 39 text

AppDynamics leverages and embraces open-‐source

Slide 40

Slide 40 text

Leading companies invest in performance § Etsy = Kale = Statsd + Skyline + Oculus (stats collec$on + anomaly detec$on/correla$on) § Nealix = PCP + Vector + Servo + Atlas (dashboards, data collec$on, root cause analysis) § Twi*er = Zipkin (distributed tracing)

Slide 41

Slide 41 text

Recommenda7ons § Treat performance as a feature -‐ Create a performance budget with milestones, speed index, page speed -‐ Capacity plan and load test the server-‐side -‐ Op$mize and performance test the client-‐side § Monitor performance in development and produc7on -‐ Instrument everything -‐ Measure the diﬀerence of every change -‐ Understand how failures impact performance § Make monitoring cri7cal and test in your con7nuous delivery process § Connect the exec/dev/ops performance perspec7ves to align on business impact

Slide 42

Slide 42 text

Go back and inves7gate how your company can break down the monitoring silos and be more impac]ul with applica7on intelligence.