Is your mobile app up to speed softwaresymposium
Download as PPTX, PDF0 likes737 views
Mobile data usage has grown exponentially and will continue to do so, putting pressure on networks and draining batteries. To improve performance, developers must optimize mobile apps using techniques like minimizing file sizes, caching content, grouping connections, and closing connections properly. The free AT&T Application Resource Optimizer tool allows developers to test apps and identify issues in order to build high-quality, efficient mobile experiences.
Is your mobile app up to speed softwaresymposium
- 1. Is Your Mobile App Up to Speed? The Art of Mobile Performance Doug Sillars Mobile Optimization Lead AT&T Seattle, WA, USA ckr.com/photos/photophilde/8108861388/
- 2. How Fast is Your Mobile App?
- 5. Mobile Data Visualized 1500 1000 500 0 The Internet 2000 Mobile 2012 In 2012, mobile data usage was 12x higher than ALL Internet usage in 2000. http://www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns705/ns827/white_paper_c11-520862.html
- 6. Mobile Data Visualized 1500 1000 500 0 The Internet 2000 Mobile 2012 If you count mobile data offloaded to Wi-Fi, it is closer to 18! http://www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns705/ns827/white_paper_c11-520862.html http://timesofindia.indiatimes.com/tech/tech-news/internet/Over-50-of-internet-users-are-mobile-only-Report/articleshow/24121500.cms?
- 7. The Future 25000 20000 15000 10000 5000 0 2000 2012 2017 2017: 21 Exabytes of Mobile Data (46% Wi-Fi offload)
- 8. Explosive Growth • Faster Networks • More Powerful Devices •Apps – Video – Music – Social Media 2005 2005 Apps are revolutionizing the internet revolution! 2004
- 9. Mobile Expectations • 85% of users prefer apps – Perceived as: – More convenient (55%) – Faster (48%) – Easier to browse (40%) • 56% had a problem with an app – Crash (62%) – Slow (47%) – Did not work (40%) http://offers2.compuware.com/APM_13_WP_Mobile_App_Survey_Report_Registration.html
- 11. Expectations: Performance Very or Somewhat Important 60% 50% Very Important 40% Somewhat Important 30% Neither Important Nor Unimportant 20% Not That Important 10% Not At All Important 0% Very Important Somewhat Important Neither Important Nor Not That Important All Important Unimportant Not At http://offers2.compuware.com/APM_13_WP_Mobile_App_Survey_Report_Registration.html
- 12. http://offers2.compuware.com/APM_13_WP_Mobile_App_Survey_Report_Registration.html
- 13. Battery Life
- 16. Testing Performance • Learn How Mobile is Different • Ensure You Are Optimized: – Learn the best practices – Test your application – Fix issues before your customers discover them • AT&T’s Application Resource Optimizer (ARO) – http://Developer.att.com/ARO
- 17. Mobile Data IS Different Radio Resource Control (RRC) State Machine Goal: Balance Fast Network Connectivity Battery Usage
- 18. Radio Resource Control (LTE) Click Me! Continuous Reception IDLE Goats! Goat! 10s Radio OFF ON Data Transfer Timer Expiration Battery Data Transfer 100ms Tail
- 19. It’s All Connected •Data traffic triggers RRC State transitions: • RRC State transitions control – Speed/User Experience
- 20. It’s All Connected •Application traffic triggers RRC State transitions: • RRC State transitions control – Speed/User Experience – Device battery life
- 21. It’s All Connected •Application traffic triggers RRC State transitions: • RRC State transitions control – Speed/User Experience – Device battery life – Network efficiency
- 22. AT&T’s Application Resource Optimizer (ARO) • • • • • • Free diagnostic tool Open Source Test any Platform Emulate any carrier network Analysis of both Native Apps and the Mobile Web Winner of the 2013 GSMA Smartphone Application Challenge award at MWC • 2013 American Business Award: Tech Innovation of the Year • Allows YOU as a developer to SEE exactly how data is flowing from your app!
- 23. How Do I Test My App? • All Platforms: Pcap/tcpdump network trace • Ex: Wi-Fi Hotspot (like Connectify) and Wireshark/NetMon to collect • iOS: Native Collection tethered from Mac coming 03/2014! • Android and Windows 8: Native ARO Collector • Note: Android collector requires root (non-root collector in development)
- 24. How Did I Do? Transfer Trace Files • Test your app – Active Use – IDLE (for Ads/analytics) – Background Data ARO Analyzer (PC/MAC/Linux) •Applies Radio and Power models to data •Graded against 25 Best Practices •Learn how your application ACTUALLY behaves!
- 25. ARO Analyzer • All network traffic mapped to processes • Stats about trace
- 26. Best Practices
- 27. Best Practices –File Download • Make Stuff Smaller • Download Less Stuff
- 28. Make Stuff Smaller: Text Files • Gzip – Smaller files download faster and use less data – Best for files > 800 bytes
- 29. Make Stuff Smaller: Text Files • Minification: – Removing human readable formatting reduces size – Separate the wheat from the chaff http://www.flickr.com/photos/11795120@N06/3832256313
- 30. Make Stuff Smaller: Images 119 KB PNG (1024x1024) www.Smushit.com (lossless compression) 78 kb 33% savings!) 720 Pixels wide
- 31. Best Practices –File Download • Make Stuff Smaller • Download Less Stuff
- 32. Download Less Stuff: Caching • 17% of HTTP content on mobile is duplicate data – Reading from Cache is 75-99% faster than downloading again 4.8 MB duplicate! 75% possible savings! (data AND speed) 6.7MB “Web Caching on Smartphones: Ideal vs. Reality”, http://www.research.att.com/~sen/pub/Caching_mobisys1 2.pdf 6.7MB
- 33. Caching Support by Library Android Connection Libraries Android & Safari iOS Libraries Test Name UC HUC HC WV HRC T20 NSUR Basic caching Revalidation Non-caching directives Expiration directives URL with query string Partial caching Redirection ASIH R AB SB Even if caching IS supported – it is OFF by default!!
- 34. Caching Methods (How do I do it?) ETags • Each file has a Unique Tag • Revalidated on server for each request – High Performance Web Sites: Rule 1 – Make Fewer HTTP Requests – Adding a connection drains battery, adds 500-3,000 ms latency (1) Cache Control Headers • Important to carefully assign Max-Age times • App will not check file on server until Max-Age is reached – Retrieval is strictly file processing time (1) http://developer.yahoo.com/blogs/ydn/posts/2007/04/rule_1_make_few/
- 35. Comparing ETag vs. Cache Control Network Usage NO Cache ETag Cache Control Battery Drain Data Usage Speed YES entire file downloaded Full network connection made High Slowest Yes – ETag in cache is validated on server Full network connection Low Slow No N/A N/A Fast Any Caching is better than no caching, but for mobile, Cache Control is preferred
- 36. Caching: Worth the Effort? Android 4.0: public void onCreate(Bundle savedInstanceState) { super.onCreate(savedInstanceState); setContentView(R.layout.main); Add this! //establish a cache try { File httpCacheDir = new File(getCacheDir(), "http"); long httpCacheSize = 10 * 1024 * 1024; // 10 MiB HttpResponseCache.install(httpCacheDir, httpCacheSize); // } catch (IOException e) { Log.i(TAG, "HTTP response cache installation failed:" + e); } Don’t leave older devices in the cold: Consider adding reflection for older versions of Android http://developer.android.com/reference/android/net/http/HttpResponseCache.html
- 37. Download Less Stuff: Combine Files • Minimize # of Downloads and Connections – Rule #1 for Faster Websites: Make Fewer HTTP Requests • Mobile Web: – 5% > 10 CSS requests – 45% > 10 JS requests – 27% > 50 images • Inline and Combine CSS and JS when Possible • Consider Spriting images http://stevesouders.com/hpws/rules.php HTTP Archive: Personal results 09/2013 data
- 38. Best Practices: Connections • Fewer Connections: faster and use less battery
- 39. Fewer Connections Imagine an app that: 1. Downloads an image every 60s 2. Downloads an Ad every 60s 3. Sends Analytics to a Server every 60s Meet chicks! ?Chickens=awesome 35s of radio time every 60s (38J!)
- 40. Optimized Connections Now, look what happens if your connections are grouped together: 1. Downloads an image every 60s 2. Downloads an Ad every 60s 3. Sends Analytics to a Server every 60s Meet chicks! ?Chickens=awesome 15s of radio (16J) every 60s (58% savings!)
- 41. How Do I Group Connections if (Tel.getDataActivity() >0){ if (Tel.getDataActivity() <4){ //ok, we are passed the minimum time to check //and we found network activity//download the image here using image getter imagegetter(counter, numberofimages); //and show the ad AdRequest adRequest = new AdRequest(); adRequest.addTestDevice(AdRequest.TEST_EMULATOR); adView.loadAd(adRequest); // Initiate a generic request to load it with an ad adView.loadAd(new AdRequest());
- 42. Grouping Connections for Speed • Threading file downloads vs. serial download
- 43. Grouping Connections for Speed • Not only can you save battery, but you can speed up your application by managing connections properly: • Removing redirects to files 2 sec { Redirection adds ~2-3 seconds for each request
- 44. Grouping Connections for Speed • Not only can you save battery, but you can speed up your application by managing connections properly: • Pre-fetching files that are used often
- 45. Closing Connections • > 80% of applications do NOT close connections when they are finished!!
- 46. Closing Connections: Example 38% more power on LTE! (18% more power on 3G)
- 47. Closing Connections: CODE • MultiRes Sample app from Android SDK – Modified to download images HttpURLConnection getimagecloseconn = (HttpURLConnection) urln.openConnection(); getimagecloseconn.setRequestProperty("connection", "close"); getimagecloseconn.connect(); String cachecontrol = getimagecloseconn.getHeaderField("CacheControl"); InputStream isclose = getimagecloseconn.getInputStream(); bitmap = BitmapFactory.decodeStream(isclose); getimagecloseconn.disconnect(); https://github.com/attdevsupport/ARO/tree/master/2013DevSummitTurbocharge
- 48. Fewer Connections: Periodic • Background data traffic: – 27% of data – 55% of network time http://www2.research.att.com/~fengqian/paper/screen_imc12.pdf
- 49. Connections: 30x Redirects • Redirection to another server can add 200600ms on each connection • Request xyz.com – Redirects to www.xyz.com – Redirects to xyz.mobi – Redirects to www.xyz.mobi • ARO has Waterfall visualization of packets
- 50. HTTP Response Codes • Look for error codes: 404s Error messages should be short and sweet
- 51. Best Practices: HTML • Order Matters • Hiding Code
- 52. HTML Tricks • Javascript files block rendering when downloaded before CSS: – CSS Before JS – JS Asynchronous load • CSS “Display:none” – Hides images from user – Images still downloaded
- 53. Everyone Wins With ARO • • • • Faster Apps Efficient Battery Use Better App Reviews More Efficient Network Traffic • Happier Customers! Customers Developers Carriers Highly Performing apps require Development Art beyond an awesome UI.
- 54. ARO • Free • Open Source • Test all Platforms • Developer.att.com/ARO • KnowMyApp.org
Editor's Notes
- #12: Redo graphs
- #14: http://www2012.wwwconference.org/proceedings/proceedings/p41.pdfhttp://gizmodo.com/5992917/battery-life-is-the-only-spec-that-mattershttp://mobithinking.com/blog/greedy-apps
- #16: Learn How Mobile is DifferentEnsure You Are Optimized:Learn the best practicesTest your applicationFix issues before your customers discover themAT&T’s Application Resource Optimizer (ARO) http://Developer.att.com/AROIn 2012, <2% of tested apps required no optimizationhttp://www.flickr.com/photos/photocapy/43580405/sizes/o/in/photostream/
- #18: Funny picture??
- #30: Separating the wheat from the chaff