Life of a Performance Specialist Symposium on Software Performance - - PowerPoint PPT Presentation

life of a performance specialist
SMART_READER_LITE
LIVE PREVIEW

Life of a Performance Specialist Symposium on Software Performance - - PowerPoint PPT Presentation

Into the wild Life of a Performance Specialist Symposium on Software Performance (SSP 2015) Munich, 5.-6.11.2015 Torsten Hellwig, Dipl.-Inf., Managing Partner @ QMETHODS Business & IT Consulting GmbH 1 You dont need to walk to


slide-1
SLIDE 1

1

“Into the wild…” Life of a Performance Specialist

Symposium on Software Performance (SSP 2015) Munich, 5.-6.11.2015

Torsten Hellwig, Dipl.-Inf., Managing Partner @ QMETHODS Business & IT Consulting GmbH

slide-2
SLIDE 2

2

You don‘t need to walk to Alaska…

slide-3
SLIDE 3

3

Who am I – how did it start

  • first programming in 1986
  • Master in computer science from TU Dresden in 1998
  • software engineer and architect, Java, PHP, JS, …
  • in 2003 established testing / QA at an insurance company
  • „moved to the other side“: leading Ops team – no system monitoring!
  • 2010 first encounter with Applicaton Performance Management: Dynatrace
  • Performance Consultant ever since…
slide-4
SLIDE 4

4

What does an application performance guy do ?

  • Troubleshoot application performance, yay !
  • Educating and evangelising APM
  • Understand customers maturity, organisational structure, processes
  • map requirements and expectations to a solution
  • help with tooling decisions: demonstrate tools, prove value
  • set up and establish tooling – develop customers maturity
slide-5
SLIDE 5

5

A little theory on APM

  • it‘s a lifecycle thing !
  • get grip and control over your environment
  • performance requirements (NFRs to FRs)
  • develop performance aware or –driven
  • white-box testing
  • holistic monitoring
  • automation !
slide-6
SLIDE 6

6

What you need to look at…

  • organisational
  • classic siloed IT organization
  • DevOps
  • outsurcing
  • skill levels
  • maturity
  • driving blind – end user experience, monitoring
  • reactive, problem-driven
  • aware and effective
  • optimized process, continuous improvement
  • technical environment
  • tooling
slide-7
SLIDE 7

7

Collaboration

slide-8
SLIDE 8

8

Troubleshooting

  • pro-actively
  • new application to be rolled out – does it perform ?
  • QA or users constantly complain
  • Dev has no idea anymore
  • Ops has no idea but ‚more hardware!‘
  • Business needs a solution
  • war room – the house is on fire
  • outtages
  • KPIs bad (conversion rate, revenue)
  • project deadline was … yesterday
slide-9
SLIDE 9

9

Example 1: Troubleshooting

  • larger customer wants to introduce a new HR software
  • big bang rollout planned and communicated to >1000 employees
  • large peak in load expected on go-live
  • vendor and hosting provider did some improvised load tests with jMeter
  • more hosted hardware would drive up costs, still no guarantee
slide-10
SLIDE 10

10

Approach

  • first: wire up APM to the test environment which is used by manual testers – do

End-to-End analysis over a whole day

  • understand the app and user behaviour
  • see the infrastructure
  • check functional health
  • second: repeat a properly defined load test under APM diagnosis
  • proper definition: user behaviour patterns from above + capabilities of the load driver
  • estimated virtual users: 1 VU = 25 real users
  • system was able to sustain 300 real users…
slide-11
SLIDE 11

11

Read the signs…

slide-12
SLIDE 12

12

Example 2: business-driven, strategic-thinking

„We have no problems, we just want to stay ahead of the rest…“

  • strong growth – want to limit growing costs
  • high peak to normal traffic ratio – elastic cloud
  • business stakeholders want insight
  • better collaboration wanted – software vendor and hosting provider are external
  • application performance management is the key

„looks like a walk in the park…“

slide-13
SLIDE 13

13

What‘s the deal then ?

  • understand their real situation
  • full outsourcing ? uhm…
  • a Bank, with demanding stakeholders
  • really high traffic
  • propose a solution and get it up and running
  • integration+automation
  • hotline
  • incident management
  • big data analytics
  • collaboration process with both external parties
  • skill transfer
slide-14
SLIDE 14

14

Example 3: Out there in the wild…

The Mainframe

slide-15
SLIDE 15

15

Active APM user…

  • it‘s about coverage
  • you can‘t start everything at the same time
  • complexity of project
  • cost
  • limited resources
  • customer with classic IT organization does APM in Dev, Test and Ops
  • started in the online business branch
  • technology coverage left out the backend: Mainframe
  • separate tooling – common architecture
  • again – growth (mobile frontend)
  • cost limiting: Mainframe pays per CPU usage !
slide-16
SLIDE 16

16

The task…

  • Assess requirements
  • propose a solution – extend APM into the Mainframe
  • win the Mainframers
  • evaluate tooling

How ?

  • get it up and running and FIND SOME PROOF
  • the trick: correlation of distributed and mainframe on transaction level
slide-17
SLIDE 17

QMETHODS - Business & IT Consulting GmbH Stargarder Straße 10 10437 Berlin Telefon : +49 (30) 22 19 86 51 Telefax : +49 (30) 22 19 86 53 E-Mail : info@qmethods.com Internet : http://www.qmethods.com

Thanks for your patience!

Questions ? Torsten Hellwig | Management Partner +49 (178) 838 0450 torsten.hellwig@qmethods.com