1. Distributed Computations MapReduce/Dryad M/R slides adapted from those of Jeff Dean’s Dryad slides adapted from those of Michael Isard

2. What we’ve learnt so far Basic distributed systems concepts –Consistency (sequential, eventual) –Concurrency –Fault tolerance (recoverability, availability) What are distributed systems good for? –Better fault tolerance Better security? –Increased storage/serving capacity Storage systems, email clusters –Parallel (distributed) computation (Today’s topic)

3. Why distributed computations? How long to sort 1 TB on one computer? –One computer can read ~60MB from disk –Takes ~1 days!! Google indexes 100 billion+ web pages –100 * 10^9 pages * 20KB/page = 2 PB Large Hadron Collider is expected to produce 15 PB every year!

4. Solution: use many nodes! Cluster computing –Hundreds or thousands of PCs connected by high speed LANs Grid computing –Hundreds of supercomputers connected by high speed net 1000 nodes potentially give 1000X speedup

5. Distributed computations are difficult to program Sending data to/from nodes Coordinating among nodes Recovering from node failure Optimizing for locality Debugging Same for all problems

6. MapReduce A programming model for large-scale computations –Process large amounts of input, produce output –No side-effects or persistent state (unlike file system) MapReduce is implemented as a runtime library: –automatic parallelization –load balancing –locality optimization –handling of machine failures

7. MapReduce design Input data is partitioned into M splits Map: extract information on each split –Each Map produces R partitions Shuffle and sort –Bring M partitions to the same reducer Reduce: aggregate, summarize, filter or transform Output is in R result files

8. More specifically… Programmer specifies two methods: –map(k, v) → * –reduce(k', *) → * All v' with same k' are reduced together, in order. Usually also specify: –partition(k’, total partitions) -> partition for k’ often a simple hash of the key allows reduce operations for different k’ to be parallelized

9. Example: Count word frequencies in web pages Input is files with one doc per record Map parses documents into words –key = document URL –value = document contents Output of map: “doc1”, “to be or not to be” “to”, “1” “be”, “1” “or”, “1” …

10. Example: word frequencies Reduce: computes sum for a key Output of reduce saved “be”, “2” “not”, “1” “or”, “1” “to”, “2” key = “or” values = “1” “1”“1” key = “be” values = “1”, “1” “2”“2” key = “to” values = “1”, “1” “2”“2” key = “not” values = “1” “1”“1”

11. Example: Pseudo-code Map(String input_key, String input_value): //input_key: document name //input_value: document contents for each word w in input_values: EmitIntermediate(w, "1"); Reduce(String key, Iterator intermediate_values): //key: a word, same for input and output //intermediate_values: a list of counts int result = 0; for each v in intermediate_values: result += ParseInt(v); Emit(AsString(result));

12. MapReduce is widely applicable Distributed grep Document clustering Web link graph reversal Detecting duplicate web pages …

13. MapReduce implementation Input data is partitioned into M splits Map: extract information on each split –Each Map produces R partitions Shuffle and sort –Bring M partitions to the same reducer Reduce: aggregate, summarize, filter or transform Output is in R result files, stored in a replicated, distributed file system (GFS).

14. MapReduce scheduling One master, many workers –Input data split into M map tasks (e.g. 64 MB) –R reduce tasks –Tasks are assigned to workers dynamically –E.g. M=200,000; R=4,000; workers=2,000

15. MapReduce scheduling Master assigns a map task to a free worker –Prefers “close-by” workers when assigning task –Worker reads task input (often from local disk!) –Worker produces R local files containing intermediate k/v pairs Master assigns a reduce task to a free worker –Worker reads intermediate k/v pairs from map workers –Worker sorts & applies user’s Reduce op to produce the output

16. Parallel MapReduce Map Input data Input data Reduce Shuffle Reduce Shuffle Reduce Shuffle Partitioned output Master

17. WordCount Internals Input data is split into M map jobs Each map job generates in R local partitions “doc1”, “to be or not to be” “to”, “1” “be”, “1” “or”, “1” “not”, “1 “to”, “1” “be”,“1” “not”,“1” “or”, “1” R local partitions “doc234”, “do not be silly” “do”, “1” “not”, “1” “be”, “1” “silly”, “1 “be”,“1” R local partitions “not”,“1” “do”,“1” “to”,“1”,”1” Hash(“to”) % R

18. WordCount Internals Shuffle brings same partitions to same reducer “to”,“1”,”1” “be”,“1” “not”,“1” “or”, “1” “be”,“1” R local partitions R local partitions “not”,“1” “do”,“1” “to”,“1”,”1” “do”,“1” “be”,“1”,”1” “not”,“1”,”1” “or”, “1”

19. WordCount Internals Reduce aggregates sorted key values pairs “to”,“1”,”1” “do”,“1” “not”,“1”,”1” “or”, “1” “do”,“1” “to”, “2” “be”,“2” “not”,“2” “or”, “1” “be”,“1”,”1”

20. The importance of partition function partition(k’, total partitions) -> partition for k’ –e.g. hash(k’) % R What is the partition function for sort?

21. Load Balance and Pipelining Fine granularity tasks: many more map tasks than machines –Minimizes time for fault recovery –Can pipeline shuffling with map execution –Better dynamic load balancing Often use 200,000 map/5000 reduce tasks w/ 2000 machines

22. Fault tolerance via re-execution On worker failure: Re-execute completed and in-progress map tasks Re-execute in progress reduce tasks Task completion committed through master On master failure: State is checkpointed to GFS: new master recovers & continues

23. Avoid straggler using backup tasks Slow workers drastically increase completion time –Other jobs consuming resources on machine –Bad disks with soft errors transfer data very slowly –Weird things: processor caches disabled (!!) –An unusually large reduce partition Solution: Near end of phase, spawn backup copies of tasks –Whichever one finishes first "wins" Effect: Dramatically shortens job completion time

24. MapReduce Sort Performance 1TB (100-byte record) data to be sorted 1700 machines M=15000 R=4000

25. MapReduce Sort Performance When can shuffle start? When can reduce start?

26. Dryad Slides adapted from those of Yuan Yu and Michael Isard

27. Dryad Similar goals as MapReduce –focus on throughput, not latency –Automatic management of scheduling, distribution, fault tolerance Computations expressed as a graph –Vertices are computations –Edges are communication channels –Each vertex has several input and output edges

28. WordCount in Dryad Count Word:n MergeSort Word:n Count Word:n Distribute Word:n

29. Why using a dataflow graph? Many programs can be represented as a distributed dataflow graph –The programmer may not have to know this “SQL-like” queries: LINQ Dryad will run them for you

30. Job = Directed Acyclic Graph Processing vertices Channels (file, pipe, shared memory) Inputs Outputs

31. Scheduling at JM General scheduling rules: –Vertex can run anywhere once all its inputs are ready Prefer executing a vertex near its inputs –Fault tolerance If A fails, run it again If A’s inputs are gone, run upstream vertices again (recursively) If A is slow, run another copy elsewhere and use output from whichever finishes first

32. Advantages of DAG over MapReduce Big jobs more efficient with Dryad –MapReduce: big job runs >=1 MR stages reducers of each stage write to replicated storage Output of reduce: 2 network copies, 3 disks –Dryad: each job is represented with a DAG intermediate vertices write to local file

33. Advantages of DAG over MapReduce Dryad provides explicit join –MapReduce: mapper (or reducer) needs to read from shared table(s) as a substitute for join –Dryad: explicit join combines inputs of different types –E.g. Most expensive product bought by a customer, PageRank computation

34. DAG optimizations: merge tree

36. Dryad Optimizations: data- dependent re-partitioning Distribute to equal-sized ranges Sample to estimate histogram Randomly partitioned inputs

37. Dryad example: the usefulness of join SkyServer Query: 3-way join to find gravitational lens effect Table U: (objId, color) 11.8GB Table N: (objId, neighborId) 41.8GB Find neighboring stars with similar colors: –Join U+N to find T = N.neighborID where U.objID = N.objID, U.color –Join U+T to find U.objID where U.objID = T.neighborID and U.color ≈ T.color

38. SkyServer query u: objid, color n: objid, neighborobjid [partition by objid] select u.color,n.neighborobjid from u join n where u.objid = n.objid

39. (u.color,n.neighborobjid) [re-partition by n.neighborobjid] [order by n.neighborobjid] [distinct] [merge outputs] select u.objid from u join where u.objid =.neighborobjid and |u.color -.color| < d

41. Another example: how Dryad optimizes DAG automatically Example Application: compute query histogram Input: log file (n partitions) Extract queries from log partitions Re-partition by hash of query (k buckets) Compute histogram within each bucket

42. Naïve histogram topology Pparse lines D hash distribute S quicksort C count occurrences MSmerge sort

43. Efficient histogram topology Pparse lines D hash distribute S quicksort C count occurrences MSmerge sort M non-deterministic merge Q' is:Each R is: Each MS C M P C S Q' RR k T k n T is: Each MS D C

44. RR T Q’Q’ MS►C►D M►P►S►C MS►C Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge R

45. MS►C►D M►P►S►C MS►C Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge RR T R Q’Q’Q’Q’Q’Q’Q’Q’

46. MS►C►D M►P►S►C MS►C Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge RR T R Q’Q’Q’Q’Q’Q’Q’Q’ T

47. MS►C►D M►P►S►C MS►C Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge RR T R Q’Q’Q’Q’Q’Q’Q’Q’ T

48. Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge MS►C►D M►P►S►C MS►C RR T R Q’Q’Q’Q’Q’Q’Q’Q’ T

49. Pparse linesDhash distribute SquicksortMSmerge sort Ccount occurrencesMnon-deterministic merge MS►C►D M►P►S►C MS►C RR T R Q’Q’Q’Q’Q’Q’Q’Q’ T

50. Final histogram refinement 1,800 computers 43,171 vertices 11,072 processes 11.5 minutes