Into the cloud, you better fly by sight

Slide 1

Slide 1 text

Cloud Native Migration INTO THE CLOUD, YOU BETTER FLY BY SIGHT Dr. Josef Adersberger

Slide 2

Slide 2 text

1. Overrun of costs, efforts, timelines 2. Unexpected severe portability issues 3. Wrong technology decisions 4. Lack of security, privacy & compliance 5. ... Things can go terribly wrong on a cloud native journey

Slide 3

Slide 3 text

3 A successful cloud migration ❏ In time: All 600+ web applications migrated to public cloud and in production within 4 years ❏ In budget: no overrun & strong business case generated ❏ Full advantage: All applications security-hardened and modernized to containerized 12-factor-apps, higher availability, more agile teams

Slide 4

Slide 4 text

4 #1 LESSON LEARNED: No blind flight into the cloud

Slide 5

Slide 5 text

It’s all about visibility

Slide 6

Slide 6 text

It’s all about visibility Source Application Landscape Migration Process Target Application Landscape Cloud Native Readiness Check Effort Estimation Cartography (Applications, Dependencies, Technologies) Cloud Cost Optimization Compliance Sign-off Hypercare Observation Migration Controlling Production Readiness Monitoring

Slide 7

Slide 7 text

Migration Data Warehouse SONARQUBE EAM TOOL OWASP Scanner jQAssistant Establish a single source of truth QUESTIONNAIRES STATIC ANALYSIS DYNAMIC ANALYSIS LEANIX Iteraplan XLS JEE Migration Tool Monitoring Tools Inventory Infrastructure 1

Slide 8

Slide 8 text

8 Provide rich data analysis and visualization capabilities 2 Migration Data Warehouse Business Intelligence Graph Search & Analysis Dashboards & Reports Issue Tracking

Slide 9

Slide 9 text

9 Business Intelligence (Tableau) ❏ What technologies are used? ❏ What libraries are integrated in which version? ❏ What protocols are used? ❏ How much code? ❏ How much over-complex code? ❏ How many code anomalies per LOC? ❏ ...

Slide 10

Slide 10 text

10 Graph Search & Analysis (neo4j) ❏ Which deployment units contain a specific class? ❏ What deployment units have a transitive dependency to a specific class or library? ❏ What are all code-level dependencies between our applications? ❏ ...

Slide 11

Slide 11 text

MATCH (app:WebApp) -[:CONSISTS_OF_JAVA_CLASS]->(class:JavaClass) -[:IMPORTS]->(dependency:JavaClass) WHERE dependency.name =~ '.*security.*' OPTIONAL MATCH (dependency) -[:PROVIDED_BY_JAR]-> (jar:JarOrModule) RETURN app, dependency, jar class LIMIT 50; 11 Graph Search & Analysis (neo4j) Which security packages are used? Interactive exploration Refactoring automation with Spoon

Slide 12

Slide 12 text

12 Dashboards & Reports Migration status Health status Quality status Information radiator Incentives :-)

Slide 13

Slide 13 text

13 Issue Tracking

Slide 14

Slide 14 text

INTO THE CLOUD, YOU BETTER FLY BY SIGHT INSTRUMENTS Dr. Josef Adersberger

Slide 15

Slide 15 text

15 HOLOWARE: Collaborative visualization, exploration, and analysis of software landscapes in XR Research project by QAware and University of Erlangen-Nuremberg funded by BMWi / ZIM

Slide 16

Slide 16 text

Slide 17

Slide 17 text

Slide 18

Slide 18 text

Slide 19

Slide 19 text

QAware GmbH Aschauer Straße 32 81549 München Tel.: +49 89 232315-0 E-Mail: info@qaware.de | sales@qaware.de qaware.de twitter.com/qaware linkedin.com/company/qaware-gmbh xing.com/companies/qawaregmbh github.com/qaware slideshare.net/qaware

Slide 20

Slide 20 text

BONUS SLIDES

Slide 21

Slide 21 text

21 Slicing & dicing multi-dimensional time series data with the Software ECG ❏ Nodes: Compute nodes ❏ Namespaces: Application namespaces ❏ Pods & Services: Containerized Applications ❏ Processes: Application processes ❏ Metrics: e.g. runtime metrics, cloud costs Dimensions: ❏ Tool for analyzing multi-dimensional time series as collected from system behaviour monitoring ❏ Capable of analyzing trillions of data points in seconds ❏ Internally used in QAware projects at BMW, Deutsche Telekom, Allianz etc,