Big Data and Space Travel
- 1. BIG DATA and your mission to the Red Planet MARS NO JUNK MAIL
- 2. 2 Copyright ? 2013 CA. All rights reserved. Big Data VARIETY Structured to unstructured VELOCITY Batch to streamed VOLUME Terabytes to Petabytes Data Analyst Boffin Dude, we gotta have one of those! But maybe explain it to me another way... Aspiring CIO Simply put ¨C the collection and analysis of large, diverse data sources to create a competitive advantage
- 3. Saturn Big Data V 6,200,000 pounds Thrust -7,648,000 pounds-force; 20k per hour (give or take) 3,000,0000 parts in a labyrinth of fuel lines, pumps, gauges, sensors, circuits, and switches ¨C each of which had to function reliably Consolidated into one beautiful piece of technology That¡¯s one giant leap for our strategic and competitive advantage THAT¡¯S ONE BIG ROCKET Value Visibility Variety Velocity Volume Copyright ? 2013 CA. All rights reserved. 3
- 4. BUT LET¡¯S NOT GET TOO EXCITED ABOUT OUR MISSION WE HAVE JUST A FEW, WELL UM ¨C TECHNICAL CHALLENGES TO OVERCOME 4 Copyright ? 2013 CA. All rights reserved.
- 5. 5
- 6. ¡°There¡¯s a whole lot of stuff to test¡± ¡°Just like in IT¡± ¡°and you can¡¯t have everything all at once¡± I¡¯d like 50 fully assembled rockets to test please You know, it is actually called rocket science for a good reason Chief Rocket Guru Junior Rocket Scientist 6 Copyright ? 2013 CA. All rights reserved.
- 7. On September 23, 1999 NASA lost the $125 million Mars Climate Orbiter spacecraft after a 286-day journey to Mars. Miscalculations due to the use of English units instead of metric units apparently sent the craft slowly off course ¡ª 60 miles in all ...but quality issues in BIG projects can have a negative impact... before, during and after lift off Tell me honestly... did you test that GPS.... No¡ but in the big scheme of things 60 miles isn¡¯t that far.... That¡¯s a miss 7 Copyright ? 2013 CA. All rights reserved.
- 8. Service What? Legacy Infrastructure Internal External Cloud Partners everyone can have everything ¨C all the time! Service Virtualization It¡¯s a Service Jim, but not as we know it Propulsion dynamics Telemetry & Communications Contractor systems Power systems There is an answer 8 Copyright ? 2013 CA. All rights reserved.
- 9. 9
- 10. CONTROL ACCESS AUTHENTICATE PROTECT DATA MANAGE IDENTITIES ...and your Smart card Your Big Data mission...is, well, let¡¯s say hush hush Password Please... ...your eyes are such beautiful colour 10 Copyright ? 2013 CA. All rights reserved.
- 11. Wow ¨C these guys throw out a lot of really useful stuff for a space pirate like me ... I can learn a lot by sifting through it And as you blaze across the cosmos... You need to protect what goes in¡ ...and also what you can leave behind in your dust... Staff Moral Product Strategy Merger Plans 11
- 12. Yeah ok, but ours has a better paint job ...after all, you don¡¯t want your data to become someone else¡¯s BIG DATA 12 Copyright ? 2013 CA. All rights reserved.
- 13. 13
- 14. Like rockets, your Big Data is fed by a complex mix of ¡®fuel¡¯ or applications ¨C poor app performance = no escape velocity ...Big Data V depends on the quality and performance of your ¡®fuel¡¯ 14 Copyright ? 2013 CA. All rights reserved.
- 15. Big Data will put similar stresses on your infrastructure ¨C especially performance¨C plan proactively ...and of course Big Data V infrastructure needs to be pretty fault tolerant N1 Lunar Rocket Complex combined forces, which ripped apart propellant lines and turbines as well other dynamic problems were not discovered and worked out before flight. 15 Copyright ? 2013 CA. All rights reserved.
- 16. 16
- 17. Same in IT ¨C you¡¯ll never have enough capacity... So you¡¯ve got to use what you have carefully and find it from other sources... how...where? ...Getting Big Data V off the ground is the least of your problems... Fuel for a roundtrip to Mars might be a bit of a challenge 400 to 450 days 17 Copyright ? 2013 CA. All rights reserved.
- 18. ...Gas Clouds of Course! SPARE PARTS INFRASTRUCTURE FOOD & WATER COMPUTE/ STORAGE CAPACITY HEAT & AIR-CON APPS Harvesting and reusing resources will depend on automation for speed and agility...ok, Big Data survival! 18 Copyright ? 2013 CA. All rights reserved.
- 19. 19
- 20. But in IT you wont have this luxury, so consolidate functions using automated processes¡ so that... ...Your Big Data V mission control elements are crucial to success! FLIGHT SURGEON Monitoring the heath of your most important assets BOOSTER SYSTEMS Monitoring performance of propulsion systems NETWORK ENGINEER Network of ground stations relaying telemetry and communications ¡AND MANY MORE¡ 20 Copyright ? 2013 CA. All rights reserved.
- 21. Help your business cost-effectively gain insights and advantage from your massive and diverse information assets ¨C BIG DATA ...your rocket scientists ¨C ok, data analysts ¨C can do what they do best 21 Copyright ? 2013 CA. All rights reserved.
- 22. Here¡¯s to your successful Big Data Mission! 22 Copyright ? 2013 CA. All rights reserved.
- 23. BIG DATA and your mission to the Red Planet MARS NO JUNK MAIL
Editor's Notes
- #9: Another key capability ¨C and a rapidly growing discipline ¨C is the area of Service Virtualization. This is a solution to the typical constraint that happens in the end-to-end development and delivery cycle. When it comes to testing especially, the typical refrain is ¡°you can¡¯t do anything until I have everything¡± ¨C meaning, you cannot even start to properly test a new application until you have all the components of that application. [CLICK ¨C circles appear showing application components - External, Partners, Cloud, Legacy, Internal] This is especially true of complex applications that combine components from internal and external sources, partners, cloud systems, and legacy platforms. [CLICK ¨C transparent boxes appear around multiple components] Not everyone can have all of those systems at once; and you cannot ¡®steal¡¯ capacity from production to do testing. So you end up in a stalemate, where you simply cannot test all your code before it goes out the door. The result is high error rates, SLA failures, and cost overruns. [CLICK ¨C small packages appear next to each of the transparent boxes ] With service virtualization, you can clone the behavior of these constrained systems, package them up, and create a simulated version of the whole application , or just some parts of it, including the components and the load patterns you choose. [CLICK ¨C the user types appear at right, and the packages move along the paths to the different users ] You can then deliver these simulated service containers to whoever needs then ¨C developers, acceptance testing, unit testing etc. whenever they need them. With service virtualization, everyone can have everything ¨C all the time! This means you can deliver new services faster because there is less contention and waiting around for resources, but you also deliver them better, because you can actually run through more test cycles to find and fix more bugs.