H T	  Technologies	   2013	  
HOST:	  Eric	  Kavanagh	  
 	  	  THIS	  YEAR	  is…	  
INVESTIGATIVE	  ANALYTICS	  =	  ž  Seeking	  previously	  unknown	  patterns	  in	  data	  ž  Extracting	  real-­‐time	 ...
ANALYST:	  Philip	  Howard	  Research	  Director,	  Bloor	  Research	  ANALYST:	  Robin	  Bloor	  Chief	  Analyst,	  The	 ...
INTRODUCING	  Philip	  Howard	  
Exploiting the Internet of Things withInvestigative AnalyticsPhilip HowardResearch Director, Bloor Research
telling the right storyConfidential © Bloor Research 2013Internet of Things+ trains, golf courses, icebergs, ATMs, pipelin...
telling the right storyConfidential © Bloor Research 2013Investigative Analytics" What happened?" Why did it happen? Is th...
telling the right storyConfidential © Bloor Research 2013Some use casesIBM X-Force survey
telling the right storyConfidential © Bloor Research 2013Requirements
telling the right storyConfidential © Bloor Research 2013
INTRODUCING	  Robin	  Bloor	  
INVESTIGATIVE ANALYTICS+THE INTERNET OF THINGS
It Begins With State…People, objects,systems, systemcomponents, etc.Things can reportstateRFID tags, sensors,log files, tw...
Transactional Event Based—  Corresponds to a systemchange—  Process heavy/data light—  Analysis happensdownstream—  Fl...
The Technology March
The Three LatenciesTime to developTime to deployUser experience
Boiling It DownIt is all about TIME TO INSIGHT – aslong as that is followed by ACTION
INTRODUCING	  Don	  DeLoach	  
Infobright: Investigative Analyticsfor The Internet of ThingsDon DeLoach, CEO, Infobrightdon@infobright.com
Internet of ThingsGraphic from Sensor Mania! The Internet of Things, Wearable Computing,Objective Metrics, and the Quantif...
Requirements for Practical Investigative AnalyticsLOW TOUCHHIGH AVAILABILITYAFFORDABILITYTCOAD HOCPERFORMANCE SCALABILITYC...
§ Data management§  Hadoop transforming this area§ Transparent analytic stack§  Operational, investigative, predictive...
Intelligence Not Hardware: Knowledge Grid• Stores	  it	  in	  the	  Knowledge	  Grid	  (KG)	  • KG	  is	  loaded	  into	  ...
Infobright Analytic SuiteInvestigative Analytics for Machine-generated Data:§  High performance ad-hoc query capabilities...
AFTERBEFOREWhat is needed today (and tomorrow)?MACHINEDATAMACHINEDATADATABASEADMINISTRATORSHARDWAREHARDWAREAPPLICATIONAPPL...
Embedded Database for M2M/Internet of ThingsLow Admin: Do not want toforce users to require DBAs tokeep solution runningLo...
Embedded Database for M2M/Internet of ThingsLow Admin: Looking forwould ensure customers havefast access to dataLoad Speed...
Embedded Database for M2M/Internet of ThingsLow Admin: Do not want toforce users to require DBAs tokeep solution runningFa...
Embedded Database for M2M/Internet of ThingsHigh Compression:Projected data growthoutpacing storage capacityAd hoc Query: ...
Momentum in the M2M/Internet of ThingsApplications in the Internet of Things will all require Low Touch, HighCapacity and ...
Thank	  You
The	  Archive	  Trifecta:	  •  Inside	  Analysis	  	  www.insideanalysis.com	  •  SlideShare	  	  www.slideshare.net/Insid...
Hot Technologies of 2013: Investigative Analytics
Hot Technologies of 2013: Investigative Analytics
Upcoming SlideShare
Loading in …5
×

Hot Technologies of 2013: Investigative Analytics

508
-1

Published on

Dr. Robin Bloor and Philip Howard with Infobright
Live Webcast on May 29, 2013

Getting to the bottom of serious situations quickly can separate success from failure in the information economy. Whether you're dealing with customer attrition or dropped phone calls, lost sales or failed machinery, the ability to perform effective root cause analysis can offer tremendous value, especially within critical time windows. This is the domain of investigative analytics – using insights gleaned from complex data sets to identify behavioral patterns, then building predictive models that send the business on a better course.

Register for this episode of Hot Technologies to hear veteran Analysts Dr. Robin Bloor of The Bloor Group and Philip Howard of Bloor Research, as they articulate their vision of what you need to utilize investigative analytics. They'll be briefed by Don DeLoach, who will discuss the Infobright solution's ability to analyze large amounts of data quickly and flexibly, thus enabling the kind of root cause analysis that can solve business issues as they arise. Infobright will focus on how their technology is designed to help businesses harness and gain insight from their machine-generated data, which is increasingly generated by instrumentation, aka the Internet of Things.

Published in: Technology, Education
1 Comment
1 Like
Statistics
Notes
No Downloads
Views
Total Views
508
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
14
Comments
1
Likes
1
Embeds 0
No embeds

No notes for slide

Hot Technologies of 2013: Investigative Analytics

  1. 1. H T  Technologies   2013  
  2. 2. HOST:  Eric  Kavanagh  
  3. 3.      THIS  YEAR  is…  
  4. 4. INVESTIGATIVE  ANALYTICS  =  ž  Seeking  previously  unknown  patterns  in  data  ž  Extracting  real-­‐time  insight  from  machine  generated  data  ž  Being  able  to  query  data  streams,  i.e.,  mobile  data,  web  logs,  geospatial  data,  social  media  data,  etc.    
  5. 5. ANALYST:  Philip  Howard  Research  Director,  Bloor  Research  ANALYST:  Robin  Bloor  Chief  Analyst,  The  Bloor  Group  GUEST:  Don  DeLoach  CEO  &  President,  Infobright  THE  LINE  UP  
  6. 6. INTRODUCING  Philip  Howard  
  7. 7. Exploiting the Internet of Things withInvestigative AnalyticsPhilip HowardResearch Director, Bloor Research
  8. 8. telling the right storyConfidential © Bloor Research 2013Internet of Things+ trains, golf courses, icebergs, ATMs, pipeline networks ….
  9. 9. telling the right storyConfidential © Bloor Research 2013Investigative Analytics" What happened?" Why did it happen? Is this part of a pattern that indicates that itmight happen again?" How are we going to react? If it is part of a pattern how can wecan leverage this for business purposes in the future?
  10. 10. telling the right storyConfidential © Bloor Research 2013Some use casesIBM X-Force survey
  11. 11. telling the right storyConfidential © Bloor Research 2013Requirements
  12. 12. telling the right storyConfidential © Bloor Research 2013
  13. 13. INTRODUCING  Robin  Bloor  
  14. 14. INVESTIGATIVE ANALYTICS+THE INTERNET OF THINGS
  15. 15. It Begins With State…People, objects,systems, systemcomponents, etc.Things can reportstateRFID tags, sensors,log files, tweets, etc.Such snippets of dataare eventsEVERYTHING HASSTATE
  16. 16. Transactional Event Based—  Corresponds to a systemchange—  Process heavy/data light—  Analysis happensdownstream—  Flows as part of abusiness process—  Fast—  Corresponds to a statechange—  Process light/data heavy—  Analysis can happenpre-transaction—  Can be a trigger in abusiness process—  FasterTransactions v Events
  17. 17. The Technology March
  18. 18. The Three LatenciesTime to developTime to deployUser experience
  19. 19. Boiling It DownIt is all about TIME TO INSIGHT – aslong as that is followed by ACTION
  20. 20. INTRODUCING  Don  DeLoach  
  21. 21. Infobright: Investigative Analyticsfor The Internet of ThingsDon DeLoach, CEO, Infobrightdon@infobright.com
  22. 22. Internet of ThingsGraphic from Sensor Mania! The Internet of Things, Wearable Computing,Objective Metrics, and the Quantified Self 2.0,JSAN, Nov. 2102
  23. 23. Requirements for Practical Investigative AnalyticsLOW TOUCHHIGH AVAILABILITYAFFORDABILITYTCOAD HOCPERFORMANCE SCALABILITYCOMPRESSIONLOAD SPEEDS
  24. 24. § Data management§  Hadoop transforming this area§ Transparent analytic stack§  Operational, investigative, predictive§  Machine-generated, text§ User consumption:§  Real-time, interactive visualization & query creationEmerging Data Analytics Stack:Days of One-Size-Fits All Are Gone“Yesterday’s  BI-­‐ETL-­‐EDW  stack  is  wrong-­‐sided  for  tomorrow’s  needs,  and  quickly  becoming  irrelevant.”  Gigamon  
  25. 25. Intelligence Not Hardware: Knowledge Grid• Stores  it  in  the  Knowledge  Grid  (KG)  • KG  is  loaded  into  memory  • Less  than  1%  of  total  compressed  data  size      Creates  informa?on  (metadata)  about  the  data  upon  load,  automa?cally  • The  less  data  that  needs  to  be  accessed,  the  faster  the  response  • Sub-­‐second  responses  when  answered  by  the  KG  Uses  the  metadata  when  processing  a  query  to  eliminate  /  reduce  need  to  access  data  • No  need  to  par??on  data,  create/maintain  indexes,  projec?ons  or  tune  for  performance  • Ad-­‐hoc  queries  are  as  fast  as  sta?c  queries,  so  users  have  total  flexibility  Architecture  Benefits  
  26. 26. Infobright Analytic SuiteInvestigative Analytics for Machine-generated Data:§  High performance ad-hoc query capabilities—enabling real-time information insights atthe speed of business§  Extremely efficient (footprint, compression, data load) analytic engine designed forenterprise software deployments, OEM/embedded configurations and enterprise-readyappliance configurations proven in production§  Install to analytics in hours: Infobright is designed for time to value§  Integrated with the leading Hadoop, BI and ETL playersOperationalSimplicityHighPerformanceEfficientForm FactorInfobright sets the bar forquery performance, formfactor, and analyticsbusiness impact
  27. 27. AFTERBEFOREWhat is needed today (and tomorrow)?MACHINEDATAMACHINEDATADATABASEADMINISTRATORSHARDWAREHARDWAREAPPLICATIONAPPLICATION
  28. 28. Embedded Database for M2M/Internet of ThingsLow Admin: Do not want toforce users to require DBAs tokeep solution runningLoad Speeds: Ingestion ratescontinue to increase, placingheavy burden on solutionsHigh Compression: Want tokeep longer histories in lessspaceLower TCO: Resulting inbetter value for customers,better margins for providersStripped Away “DBA” taxrequirement required byprevious versionsIngesting over 1TB/Hour,with significant headroombeyond thatOver 3X the retention periodand a 5X simultaneousreduction in storagerequirementLower TCO for users,higher margins for JDSULittle to NoAdminFast LoadSpeeds20:1+CompressionExceptional AdHoc QueryPerformanceVery Low TCOREQUIREMENTS EXAMPLE: JDSU
  29. 29. Embedded Database for M2M/Internet of ThingsLow Admin: Looking forwould ensure customers havefast access to dataLoad Speeds: Handleprojected 70% growth rate inmobile messagingHigh Compression: Need toincrease data stored withoutincrease in storagerequirementsLower TCO: Competitiveflexibility of lower cost withhigher value-add servicesNo indexes, data partitioningor manual tuning. No need fordedicated DBAs.100,000 records per secondat peak making data availablefor analysis within minutesIncreased to 90 days of datastored in less hardware dueto drastic compressionTCO only 20% of the cost ofcompetitors. Major wirelesscarrier wins with this solutionLittle to NoAdminFast LoadSpeeds20:1+CompressionExceptional AdHoc QueryPerformanceVery Low TCOREQUIREMENTS EXAMPLE: MAVENIR
  30. 30. Embedded Database for M2M/Internet of ThingsLow Admin: Do not want toforce users to require DBAs tokeep solution runningFast Query Performance:Customers depend on thisanalysis to tune networksHigh Compression and FastLoad Speeds: Need to meetbusiness growth projectionsLower TCO: Resulting inbetter value for customers,better margins for providersLow touch administrationreduces friction and latencyfor queriesSub-second web-basedqueries critical to customersto tune the networkHigh data compressionrates and load speed allowfor projected growth rate ofdata volumeLow OPEX = better marginsand more confidence planningcapacity to meet growthLittle to NoAdminFast LoadSpeeds20:1+CompressionExceptional AdHoc QueryPerformanceVery Low TCOREQUIREMENTS EXAMPLE: POLYSTAR
  31. 31. Embedded Database for M2M/Internet of ThingsHigh Compression:Projected data growthoutpacing storage capacityAd hoc Query: Utilities wantto drive customer participationin efficiency-related programsFast Load Speeds: Need tointegrate several data streamsquicklyLower TCO: Solution needsto affordably meet businessneedsNo additional hardware ormanual set-up in the form ofdata indexing or partitioningFast flexible reporting (20Kreports in first 3 months) helputilities better drive businessBetter business answersdue to combined analysis ofbehavioral, demographic andlog dataLow TCO translates to betterpricing and strongercompetitive positioningLittle to NoAdminFast LoadSpeeds20:1+CompressionExceptional AdHoc QueryPerformanceVery Low TCOREQUIREMENTS EXAMPLE: OPOWER
  32. 32. Momentum in the M2M/Internet of ThingsApplications in the Internet of Things will all require Low Touch, HighCapacity and High Density; and Low Cost deploymentsSmart GridsSmartVehicles,Smart CitiesMobile HealthOthers..BEFOREMACHINEDATADATABASEADMINISTRATORSHARDWAREAPPLICATIONAFTERMACHINEDATAHARDWAREAPPLICATION
  33. 33. Thank  You
  34. 34. The  Archive  Trifecta:  •  Inside  Analysis    www.insideanalysis.com  •  SlideShare    www.slideshare.net/InsideAnalysis  •  YouTube    www.youtube.com/user/BloorGroup  THANK  YOU!  

×