On The Social Impedance Mismatch in Data Storage

Data Storage affects the way systems are modelled. This slide set shall give a short justification why the data storage and -processing layer needs an overhaul to overcome its current limitations.

Transcript

1. What is “social”? Data in a personal context.
2. I have a suspicion: Data Store Software is not social Martin Scholl <martin@infinipool.com> @zeit_geist
4. Column -Store
5. Column -Store Row-Store
8. •Notes • are a fact. can get copied. • consist of immutable & absolute entities • have a fixed beginning and ending • music sheet = music essence; “Music’s NoSQL DB” •Music • Making is a process. You can record but not copy music. • flows with the rhythm • lives by the interactions • is uniquely determined in space and time: the Music’s context
9. Music = Music Sheet + Context Music Sheet
10. Music = Music Sheet + Context Music Sheet ?
11. Music = Music Sheet + Context Music Sheet ?
13. Social Context Data
14. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB
15. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB 1. Social Interaction: Data + Context }
16. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB 1. Social Interaction: Data + Context } 2. just Data }
17. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB
18. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB just Data }
19. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB context-freed Data } just Data }
20. •Data Stores • store facts. • facts are fix and absolute • facts are uniquely determined by key / ID • Data Stores are the source of “truth” • contain what has happened.
21. •Notes • are a fact. can get copied. • consist of immutable & absolute entities • have a fixed beginning and ending • music sheet = music essence; “Music’s NoSQL DB” •Data Stores • store facts. • facts are fix and absolute • facts are uniquely determined by key / ID • Data Stores are the source of “truth” • contain what has happened.
22. Lose Information w/ your fav. Data Store • Data in a Data Store gets de-contextualized. • You don’t get to know the origin of data but just the fact itself. • irrecoverable information loss! • There is a severe social impedance mismatch
23. Lose Information w/ your fav. Data Store • Data in a Data Store gets de-contextualized. • You don’t get to know the origin of data but just the fact itself. • irrecoverable information loss! • There is a severe social impedance mismatch
24. How can we fix the social impedance mismatch?
25. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB Data } Data + Context }
26. Webserver Webserver Webserver Webserver API-Endpoint API-Endpoint API-Endpoint API-Endpoint Graph-DB Graph-DB Graph-DB Flock-DB Data } Data + Context } Data + Context Logic } Context- Engine Context- Engine Context- Engine
27. Context Engine Requirements • must have a flexible programming model • must be scalable and resilient • must be able to integrate and process data from high velocity data sources
28. Nathan Marz’s Storm • has a flexible programming model • is scalable and resilient • integrates and processes data from high velocity data sources
29. Nathan Marz’s Storm • implemented in Clojure + Java • was Backtype proprietary • OpenSource’d Sep 2011 • is Eclipse Public License licensed • http://github.com/nathanmarz/storm
30. What does Storm? • it’s like M/R but for real-time computation • works over streams • communicates tuples in a cluster Spout Bolt Bolt Bolt Bolt Bolt
31. What does Storm? • Local Development mode or distributed • Starts JVMs (workers) • at-least-once message processing guarantee • Storm’s contributions: scalability, resiliency and processing guarantee Spout Bolt Bolt Bolt Bolt Bolt
32. Some Use-Cases • Analysis on Event-Streams: • Filtering, Counting, Aggregation • Monitoring, etc. etc. • Parallel and Distributed RPC • Contextualization Spout Bolt Bolt Bolt
33. Message Processing Guarantee Spout Acker Bolt Bolt ID V Bolt
34. Message Processing Guarantee Spout Acker Bolt Bolt Tuple(id=42) ID V (id=42) Bolt
35. Resiliency Spout Acker Bolt Bolt Tuple(id=42) ID V 42 42 Bolt
36. Resiliency Spout Acker Bolt Bolt Tuple(id=42) ID V 42 42 Bolt
37. Resiliency Spout Acker Bolt Bolt ID V 42 42 Tuple(id=40) Bolt Tuple(id=4)
38. Resiliency Spout Acker Bolt Bolt ID V 42 42^40^4 Tuple(id=40) Bolt Tuple(id=4) (id=[40,4])
39. Resiliency Spout Acker Bolt Bolt ID V 42 42^40^4 Tuple(id=40) Bolt Tuple(id=4) (ack=42)
40. Spout Acker Bolt Bolt ID V 42 40^4 Tuple(id=40) Bolt Tuple(id=4) Message Processing Guarantee
41. Spout Acker Bolt Bolt ID V 42 40 ^ 4 (id=40) Tuple(id=40) Bolt Tuple(id=4) Message Processing Guarantee
42. Spout Acker Bolt Bolt ID V 42 4 (id=4) Bolt Tuple(id=40) Tuple(id=4) Message Processing Guarantee
43. Spout Acker Bolt Bolt ID V 42 0 Bolt ack(id=42) Message Processing Guarantee
44. Resilience • a centralized component coordinates deployment and starts worker (Nimbus) • Workers run distributed & are supervised • Online State is persisted into Zookeeper • Every component may fail Nimbus ZK ZK ZK Worker Worker
45. Use-Case • Use-Case: Online A/B Testing • Contextualization: determine Clique (A | B) online • Reconfigure A/B-Test really quick Spout Clickstream ∑ New Configuration User User
46. Use-Case • Use-Case: Social Graph Update Propagation • Send E-Mail to B • Update Recommendation Matrix for A (and B) Spout ‘A follows B now’ A A Bolt B B Bolt New Configuration New ML Model Send EMail
47. Contextualization with Storm • Contextualization ✓ • Store Users’ context in-memory using Bolts • Continuously persist state into stable storage • Towards real-time context to every request Spout Consolidated Event-Stream User User User Recom- mender Trending Stuff / global stats Anti- Spam
48. On Storm • Storm is not a silver-bullet • Rather Storm is petri dish for real-time computation and coordination tasks • Topology changes: stop-start-cycle required • There is no Pig Latin / Hive for Storm • Advanced Topics are added with every release (e.g. Transactional Semantics)
49. Lessons Learned • De-Contextualization is a bad thing. • Your data store won’t help you. • You have to add some magic to your stack. • Storm has the potential to become the Next Big Thing after Hadoop • Use Storm to fix the Social Impedance Mismatch Issue
50. Want to change the world with real-time data? contact me: Martin Scholl <martin@infinipool.com> @zeit_geist
52. Data Stores (DBMS, NoSQL) Event Systems (e.g. Storm, S4) Model Queries Data Focus Dataset Size Domain Pull Push Run Once Run Continuously Historic Live Retrieval & Storage Format Efficiency Throughput & Latency 10^9 10^6 Volume Velocity
53. A Note on Time • Real-Time: milliseconds - seconds • Near Real-Time: seconds-minutes • Batch: minutes-