Next-Gen Business Transaction Configuration, Instrumentation, and Java Performance Diagnostics - AppSphere16
2 likes830 views
The document discusses the next-generation configuration and Java performance diagnostics by AppDynamics, outlining their approach to scalable application monitoring through a flexible configuration model. It highlights various features including live traffic introspection, JVM performance diagnostics, and contention analysis, aimed at enhancing monitoring capabilities in microservices architectures. Additionally, it emphasizes the proprietary nature of the materials and the evolution of their product roadmap without guaranteeing future features.
![Allocation rate
• Objects created / [Time]
–Eden Space
–Old Heap
• Tuning Young Heap
• High
–Frequent Young GCs
Copyright © 2015 AppDynamics. All rights reserved. 39
Promotion rate
• Objects Promoted / [Time]
–S2 => Old Heap
• Tuning Young & Old Heap
• High
–Frequent Full GCs
–Premature Promotion](https://image.slidesharecdn.com/t28329-prasad-srini-161126232019/85/Next-Gen-Business-Transaction-Configuration-Instrumentation-and-Java-Performance-Diagnostics-AppSphere16-39-320.jpg)
![Live data
• Lifetime memory
• Release comparison
• Growth impact
– Frequent full GCs
– Longer full GCs
– Higher OOM probability
– Heap fragmentation
Copyright © 2015 AppDynamics. All rights reserved. 40
Free rate
• Objects free / [time]
• GC efficiency
• Tuning impact](https://image.slidesharecdn.com/t28329-prasad-srini-161126232019/85/Next-Gen-Business-Transaction-Configuration-Instrumentation-and-Java-Performance-Diagnostics-AppSphere16-40-320.jpg)
Next-Gen Business Transaction Configuration, Instrumentation, and Java Performance Diagnostics - AppSphere16
- 1. Next-Gen Configuration and Java Performance Diagnostics Aakrit Prasad, Principal Product Manager, APM Vinay Srini, Sr. Director of Engineering, APM
- 2. APPDYNAMICS CONFIDENTIAL AND PROPRIETARY 2 Notice The information and materials included in this presentation (collectively, the “Materials”) are the proprietary information of AppDynamics, Inc. (“AppDynamics” or the “Company”). No part of the Materials may be reproduced, distributed, communicated or displayed in any form or by any means, or used to make any derivative work, without prior written permission from AppDynamics. The Materials may contain product roadmap information of AppDynamics. AppDynamics reserves the right to change any product roadmap information at any time, for any reason and without notice. This information is intended to outline AppDynamics' general product direction, it is not a guarantee of future product features, and it should not be relied on in making a purchasing decision. The development, release, and timing of any features or functionality described for AppDynamics' products remains at AppDynamics' sole discretion. AppDynamics reserves the right to change any planned features at any time before making them generally available as well as never making them generally available. All third-party trademarks, including names, logos and brands, referenced by AppDynamics in this presentation are property of their respective owners. All references to third-party trademarks are for identification purposes only and shall be considered nominative fair use under trademark law. © 2016 AppDynamics, Inc. All rights reserved.
- 3. Agenda • @Scale Configuration – Flexible Model – Live Traffic Introspection – DEMO • JVM Performance Diagnostics – Contention Analysis – Garbage Collection • Q & A
- 4. APM vision
- 5. Our approach We understand transactions from the start
- 6. Our approach We understand transactions from the start
- 7. Copyright © 2015 AppDynamics. All rights reserved. 7 Continued growth in application monitoring requires a scalable configuration
- 8. Scaling your business with AppDynamics
- 9. Software architectural evolution 2000s Layered Monolithic 2010s Microservices 1990s Spaghetti Code
- 10. Evolving software services • Growth from monolithic modular- micro- services • AppDynamics Enterprise growth – 10 100 1,000 of applications • Capability today must evolve to scale for tomorrow Web App Db
- 11. Configuration for scale • Challenges – Hierarchical – No Grouping of Services – Redundant Rules • Solution – Flexible – Groupable Services
- 12. Hierarchical configuration Copyright © 2015 AppDynamics. All rights reserved. 12 Web Back MyTier My Application /prod/check What if I need a new rule for only “My Tier”? Web Back MyTier ● One set of rules applied everywhere ● Simple deployment
- 13. Hierarchical configuration Copyright © 2015 AppDynamics. All rights reserved. 13 My Application /prod/check My Tier /prod/check/special COPY At scale, this is unreliable and error prone. 3 operations to modify any rule! 1. Modify the original rule 2. Remember what are duplicates 3. Modify the rule that was copied
- 14. Copyright © 2015 AppDynamics. All rights reserved. 14 Introducing “Scope” My TierMy Application Web BackMyTier Default Scope Special Scope Scopes can hold combination of Application and Tiers
- 15. Copyright © 2015 AppDynamics. All rights reserved. 15 Attach your rules to scopes My TierMy Application Web BackMyTier /check /prod /special No Duplication Nothing to remember No more errors Default Scope Special Scope
- 16. DEMO : Scope Configuration Model 16
- 17. Effective configuration • Flexible configuration != effective configuration • Challenges today – Matching & splitting conditions not trivial – Mis-configured rules (common for Pojo) – Masked transaction detection – Uninstrumented traffic Auto rule BT Custom Rule BT (*Masked*)
- 18. AppDynamics Rule configuration with live traffic introspection
- 19. Uninstrumented code detection • Find Transactions by Backends • Complete Stack Trace View • Monitor Method Data • Instant Instrumentation
- 20. Live Instrumentation for Pojo Entry Points • Validate Method Invocations – Name – Type (return) – Details • Object Traversal 20
- 21. Preview for masked transaction • Live Transactions Detected from Transient Rules • BT Generated (Unique v. Masked) • Masking BT(s) Information – Name – State – Load • Save and View in BT List
- 22. Introspection via class method browser • Reflect on all Class & Methods • Instant Pojo Instrumentation • See Other Instrument Methods
- 23. Live instrumentation for servlet entry points • All HTTP requests – Headers, Params, Cookies – Matching & Splitting • Aggregated view – URI – Headers – HTTP Params – Hostname – Port
- 24. Live Transaction Instrumentation DEMO 1.Pojo Entry from Uninstrumented Code 2.Splitting High Value Servlet BTs
- 25. BT performance diagnostics Copyright © 2015 AppDynamics. All rights reserved. 25 ! Deep Dive Features for Detection & Diagnostics • Per Transaction : Contention Analysis • All Transactions : GC Optimization
- 26. Java APM Features for Performance Diagnostics
- 27. Thread contention 101 Copyright © 2015 AppDynamics. All rights reserved. 27 Multi-threaded enterprise services with many shared resources
- 28. Contention visibility Copyright © 2015 AppDynamics. All rights reserved. 28 • Critical for Dev-Ops Monitoring when Scaling Services • More Common in Microservices Architectures • Application Performance RCA
- 29. Thread Contention Analyzer Copyright © 2015 AppDynamics. All rights reserved. 29
- 30. BT Call Graph Thread Contention Copyright © 2015 AppDynamics. All rights reserved. 30
- 31. Call Graph Element : Block Time Copyright © 2015 AppDynamics. All rights reserved. 31
- 32. Service Endpoint List View Copyright © 2015 AppDynamics. All rights reserved. 32
- 33. Call Graph Element : Wait Time Copyright © 2015 AppDynamics. All rights reserved. 33 • Thread Time Waiting to Execute • JVM Wait States – Thread.sleep(sleeptime) – Object.wait(timeout) – Thread.join(timeout) • All Wait States Captured
- 34. Transaction Latency from GC • Issues can be Systemic and NOT Isolated to one BT • Common Sources when all Transactions are Latent – Network Congestion – Hardware Scale (VM Paging) – JVM Garbage Collection (App Level) Copyright © 2015 AppDynamics. All rights reserved. 34
- 36. Garbage collection • Benefits – Memory management – Avoid memory leaks – Extensively tunable • Drawbacks – Tuning needed for performance – Monitoring – Ergonomics dependency Copyright © 2015 AppDynamics. All rights reserved. 36
- 37. GC visibility today Copyright © 2015 AppDynamics. All rights reserved. 37 • Young Generation Heap – Eden – S1 & S2 • Old Generation Heap • Perm Generation Heap • Code Cache
- 38. Copyright © 2015 AppDynamics. All rights reserved. 38 http://www.oracle.com/technetwork/java/javase/gc-tuning-6-140523.html • Allocation Rate • Promotion Rate • Free Rate • Live Data Size GC Performance
- 39. Allocation rate • Objects created / [Time] –Eden Space –Old Heap • Tuning Young Heap • High –Frequent Young GCs Copyright © 2015 AppDynamics. All rights reserved. 39 Promotion rate • Objects Promoted / [Time] –S2 => Old Heap • Tuning Young & Old Heap • High –Frequent Full GCs –Premature Promotion
- 40. Live data • Lifetime memory • Release comparison • Growth impact – Frequent full GCs – Longer full GCs – Higher OOM probability – Heap fragmentation Copyright © 2015 AppDynamics. All rights reserved. 40 Free rate • Objects free / [time] • GC efficiency • Tuning impact
- 41. Summary Copyright © 2015 AppDynamics. All rights reserved. 41 Configuration for Scale • Flexible Model • Live Traffic Introspection Performance Diagnostics for Low Latency • Contention Analysis (GA : 4.2.5) • Tuning the GC
- 43. Please give us your feedback—Session T28329 • Complete the online survey you'll receive via email later today or via text at: Text this number: 878787 Text this word: APPSPHERE • Every time you submit a session survey, your name will be entered in a random drawing. We're giving away Amazon Echos to 5 lucky winners! • Thank you for your input 43 Win!
- 44. Thank you