0
Scaling Analytics with     elasticsearch      Dan Noble      @dwnoble
Background• Technologist at The HumanGeo• We use elasticsearch to build social media  analysis tools• 100MM documents inde...
Overview• What is elasticsearch?• Scaling with elasticsearch• How can I use elasticsearch to help with  analytics?• Use Ca...
What is elasticsearch?
Search Engine•   Open source•   Distributed•   Automatic failover•   Crazy fast
Search Engine•   Actively maintained•   REST API•   JSON messages•   Lucene based
Search                  Elasticsearch “Cluster”                           Host                      Index: Articles• Simpl...
Distributed Search                                Elasticsearch “Cluster”                    Host                         ...
Redundancy                          Elasticsearch Cluster              Host                                   Host        ...
Node Auto Discovery                       Elasticsearch Cluster           Host                Host               Host     ...
Failover                          Elasticsearch Cluster             Host                 Host               Host          ...
Querying                    Elasticsearch Cluster        Host                Host                Host     Articles (a)    ...
REST API• JSON query syntax• Developer friendly• Easy to get started
Python Exampleimport raweses = rawes.Elastic(elastic-00:9200)es.get(articles/_search, data={   "query": {     "filtered" :...
Community
Elasticsearch Summary•   Scales horizontally•   Redundancy•   Configures itself automatically•   Developer friendly
Analytics and elasticsearch•   Date Histograms•   Statistical facets•   Geospatial queries•   All with arbitrary search pa...
Use Case: Social Media Analysis• Use social media APIs to search for data on a  topic of interest• 100MM documents indexed...
Sample Documentes.post(articles/facebook, data={   ”date": "2012-09-01 08:37:55",   "tags": {       "sentiment": {        ...
Analytical Queries
Date Histogram for Sentimentes.get(articles/_search, data={   "query" : {      "query_string" : {        "query" : "Mohame...
Date Histogram for Sentiment
Statistical Facet for Sentiment: Queryes.get(articles/_search, data={   "query" : {      "query_string" : {        "query"...
Statistical Facet for Sentiment: Result{    "facets": {       "sentiment_stats": {          "_type": "statistical",       ...
Top Keywordses.get(articles/_search, data={   "query" : {      "match_all" : {}   },   "facets" : {      "search_terms" : ...
Top Search Terms
Geospatial searches.get(articles/_search, data={   "query" : {     "filtered" : {         "filter" : {             "geo_di...
Questions
Upcoming SlideShare
Loading in...5
×

Scaling Analytics with elasticsearch

9,909

Published on

Published in: Technology, Business
2 Comments
15 Likes
Statistics
Notes
  • Hi Dan! Really interesting presentation. Elasticsearch's facets seem to work pretty well for summary statistics, but I'm interested in doing some slightly more involved operations like logistic regression, and I'm not really sure which path to take. Do you have any suggestions or experiences to share along those lines?
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • how about performace?
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
No Downloads
Views
Total Views
9,909
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
92
Comments
2
Likes
15
Embeds 0
No embeds

No notes for slide

Transcript of "Scaling Analytics with elasticsearch"

  1. 1. Scaling Analytics with elasticsearch Dan Noble @dwnoble
  2. 2. Background• Technologist at The HumanGeo• We use elasticsearch to build social media analysis tools• 100MM documents indexed• 600GB+ index size• Author of Python elasticsearch driver “rawes” https://github.com/humangeo/rawes
  3. 3. Overview• What is elasticsearch?• Scaling with elasticsearch• How can I use elasticsearch to help with analytics?• Use Case: Social Media Analytics
  4. 4. What is elasticsearch?
  5. 5. Search Engine• Open source• Distributed• Automatic failover• Crazy fast
  6. 6. Search Engine• Actively maintained• REST API• JSON messages• Lucene based
  7. 7. Search Elasticsearch “Cluster” Host Index: Articles• Simple case: one host• One index containing a set of articles
  8. 8. Distributed Search Elasticsearch “Cluster” Host Host Articles (a) Articles (b)• Too much data?• Add another host• Indices can be broken up into “shards” and live on different machines
  9. 9. Redundancy Elasticsearch Cluster Host Host Articles (a) Articles (b) Articles (b) Articles (a)• Shards can be replicated to improve availability
  10. 10. Node Auto Discovery Elasticsearch Cluster Host Host Host Articles (a) Articles (b) Articles (b) Articles (b) Articles (a) Articles (a)• Say we add a third host• elasticsearch will automatically start moving shards to this new host to distribute load
  11. 11. Failover Elasticsearch Cluster Host Host Host Articles (a) Articles (b) Articles (b) Articles (b) Articles (a) Articles (a)• Say a host goes down• Shards on that host are no longer available for search• Elasticsearch automatically rebuilds these two shards on other hosts
  12. 12. Querying Elasticsearch Cluster Host Host Host Articles (a) Articles (b) Articles (b) Articles(a) Query: “Barack Obama”Can query against Client Search for articles any host (Web Application) Send request to other shards if needed
  13. 13. REST API• JSON query syntax• Developer friendly• Easy to get started
  14. 14. Python Exampleimport raweses = rawes.Elastic(elastic-00:9200)es.get(articles/_search, data={ "query": { "filtered" : { "query" : { "query_string" : { "query" : "Barack Obama" } } } }})
  15. 15. Community
  16. 16. Elasticsearch Summary• Scales horizontally• Redundancy• Configures itself automatically• Developer friendly
  17. 17. Analytics and elasticsearch• Date Histograms• Statistical facets• Geospatial queries• All with arbitrary search parameters• Again: Fast
  18. 18. Use Case: Social Media Analysis• Use social media APIs to search for data on a topic of interest• 100MM documents indexed• Sentiment analysis• Location extraction (“Geotagging”)
  19. 19. Sample Documentes.post(articles/facebook, data={ ”date": "2012-09-01 08:37:55", "tags": { "sentiment": { "positive": 0.36, "negative": 0.10 } "geotags": [{ "term" : "Cairo", "location" : "30.0566,31.2262”, “type” : “geo_point” }], "search_terms": [ "Mohamed Morsi" ] }, "item": { "publisher: "Facebook" "source_domain": "www.facebook.com", "author": "James Smith", "source_url": "http://www.facebook.com/5551231234/posts/414141414141", "content_text": "Mohamed Morsi visits Iran for first time since 1979 ....", "title": "James Smith posted a note to Facebook", "author_url: "http://www.facebook.com/profile.php?id=5551231234" }})
  20. 20. Analytical Queries
  21. 21. Date Histogram for Sentimentes.get(articles/_search, data={ "query" : { "query_string" : { "query" : "Mohamed Morsi" } }, "facets" : { "sentiment_histogram" : { "date_histogram" : { "key_field" : "date_of_information.$date", "value_field" : "tags.sentiment.positive", "interval" : "day" } } }})
  22. 22. Date Histogram for Sentiment
  23. 23. Statistical Facet for Sentiment: Queryes.get(articles/_search, data={ "query" : { "query_string" : { "query" : "Mohamed Morsi" } }, "facets" : { "sentiment_stats" : { "statistical" : { "field" : "tags.sentiment.positive" } } }})
  24. 24. Statistical Facet for Sentiment: Result{ "facets": { "sentiment_stats": { "_type": "statistical", "count": 8825, "max": 0.375, "mean": 0.008503991588291782, "min": 0.0, "std_deviation": 0.021251077265305472, "sum_of_squares": 4.623648343200283, "total": 75.04772576667497, "variance": 0.00045160828493598306 } }, "hits": { "hits": [], "max_score": 1.1120162, "total": 8825 }, "took": 60}
  25. 25. Top Keywordses.get(articles/_search, data={ "query" : { "match_all" : {} }, "facets" : { "search_terms" : { "terms" : { "field" : "tags.search_terms", "size" : 3 } } }})
  26. 26. Top Search Terms
  27. 27. Geospatial searches.get(articles/_search, data={ "query" : { "filtered" : { "filter" : { "geo_distance" : { "distance" : ”20km", "tags.geotags.location" : { "lat" : 30, "lon" : 31 } } } } }})
  28. 28. Questions
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×