Berlin Buzzwords 2012
2012-06-04..2012-06-05
Berlin, Germany
[Berlin Buzzwords 2012, Berlin Buzzwords 2015, Berlin Buzzwords 2016, Berlin Buzzwords 2018, Berlin Buzzwords 2019, Berlin Buzzwords 2020, Berlin Buzzwords 2021, Berlin Buzzwords 2022, Berlin Buzzwords 2023, Berlin Buzzwords 2024, Berlin Buzzwords 2025, Berlin Buzzwords 2026]
| title | author | length |
|---|---|---|
| Keynote: How to Make Your Project Blossom | Leslie Hawthorn | 54m |
Community equals a group working toward a common goal not a set of tools and success hinges on specifics not slogans. Define your landscape with define goals… | ||
| Keynote: Building Spanner | Alex Lloyd | 01h 08m |
Spanner is a global relational storage system built over Paxos replication and a novel TrueTime clock that provides bounded absolute error, developed five… | ||
| Performance and capacity planning in Java based data grids | Mircea Markus | 18m |
Project Radargun is an open source Java benchmarking framework that automates coordinated multi‑node runs with a master and workers, performing validation… | ||
| Scalability Challenges in Big Data Science | Mikio Braun | 30m |
Describes a real-time stream-processing pipeline for social network events that scales to millions of users and spikes of millions of events by using NoSQL… | ||
| Query A Graph with ASCII ART | Michael Hunger | 07m |
Shows a pattern-matching approach to querying property graphs by drawing ASCII-art diagrams that map to node labels and relationship types; the diagrams… | ||
| Realtime Search at Twitter | Michael Busch | 45m |
Twitter runs a modified Lucene engine called Earlybird indexing ~340–400 million tweets/day and serving ~2.3 billion queries/day with <10s indexing latency and… | ||
| Highly Scalable Discriminative Spam Filtering | Michael Brückner | 24m |
Claims that global volume is huge — about 3.2 billion email accounts with 70% of all emails as spam and >10% undetected — and shows concrete pipelines… | ||
| Named Entity Statistics Using Apache Pig | Max Jakob | 11m |
Uses Wikipedia link counts to compute MLEs for surface-form→entity probabilities, implemented as Pig scripts that use Olevia Cazale’s Wikipedia loader to… | ||
| Visualising Public Service: Seeing Inside Birmingham City | Matt patterson | 35m |
Birmingham Civic Dashboard visualizes call centre dumps, ingesting ~10,000 interactions a day from SAP, importing into SQL, aggregating into daily MongoDB… | ||
| Joining in Lucene | Martijn van Groningen | 21m |
Shows how Lucene/Solr implement joins by using block indexing and parent-child pointers to represent products and product_items, with query-time collectors… | ||
| Travis CI | Lukas Kahwe Smith | 12m |
Travis CI runs as cloud-hosted CI and configures builds with a .travis.yml file that declares language, versions and scripts; it executes a matrix testing… | ||
| PHP Content Repository Specification | Lukas Kahwe Smith | 06m |
Ports the Java Content Repository (JCR) spec into PHP as PHP CR and claims concrete reuse of Java CMS data by connecting to Jackrabbit backends and Magnolia… | ||
| From Batch to Realtime with Hadoop | Lars George | 41m |
Hadoop proves useful for long term batch storage but not instantaneous aggregates, so the architecture pairs HDFS MapReduce with HBase and a stream layer to… | ||
| Advanced HBase Schema Design | Lars George | 22m |
Covers concrete HBase choices that trade write throughput for read latency: use composite row keys and move high-cardinality attributes left into the row key… | ||
| ElasticSearch on Rails | Karel Minarik | 09m |
Shows how to wire Elasticsearch into a Rails app using the Ruby tire gem and a rails template that generates a scaffold, seeds data and can download and launch… | ||
| Three Nodes and One Cluster | Karel Minarik, Shay Banon, Lukáš Vlček | 21m |
Live demo of a three-node cluster demo shows Elasticsearch creating an index with three shards, indexing ~100k then 200k documents, and automatically… | ||
| Hadoop Ecosystem | Kai Voigt | 18m |
Explains Hadoop as HDFS plus MapReduce and lists concrete ecosystem pieces: Hive turns SQL into MapReduce jobs, Pig compiles Pig Latin scripts to MapReduce… | ||
| Building a Global Object Storage Platform | Jonathan Wisler | 08m |
Builds a global object store using OpenStack and Python with a Swift-compatible platform integrated into CDN and search and exposed via API. Uses commodity… | ||
| Hydra - An Open Source Processing Framework | Joel Westberg | 33m |
Hydra is an open-source, document-processing pipeline built around MongoDB that reads pipelines from the DB and spins independent Java workers for each stage… | ||
| You Know, for Search. Querying 24 Billion Records in 900ms | Jodok Batlogg | 19m |
Practical lessons from designing and running a 120-node Elasticsearch cluster (6 TB) and related Hadoop integration to serve very large search datasets (24… | ||
| Large Scale Graph Computation with Apache Giraph | Jakob Homan | 24m |
Apache Giraph is an Apache project that implements Pregel-style BSP for graphs as a map-only Hadoop job coordinated by ZooKeeper with one master and many… | ||
| Pangool, Hadoop API made easy! | Iván de Prado | 08m |
Library that implements tuple-based MapReduce on top of Hadoop, available as a Maven artifact and documented on the project site. Claims: simpler API that… | ||
| Apache James: More Than Emails in the Cloud | Ioan Eugen Stan | 20m |
Apache James is an email-server platform that adds a programmable processing layer via Mailets and a scalable HBase-backed store. The HBase mailbox uses a… | ||
| Apache Solr for TYPO3, Search Meets CMS | Ingo Renner | 10m |
Integration of Apache Solr with the TYPO3 CMS that maps 40 TYPO3 database tables into 14 Solr indexes, extracts full text and metadata from files, and pushes… | ||
| Germany's Next Top Data Model | Ian Plosker | 18m |
Argues that traditional normalized schemas fail at scale and NoSQL forces query-driven design, so align storage to application queries rather than forcing… | ||
| Discovery & Analytics with Hadoop, Mahout & Solr | Grant Ingersoll | 46m |
Describes a concrete SDA stack: search served by SolrCloud for sub-second queries and near‑real‑time soft commits, bulk processing and storage on Hadoop, user… | ||
| Serious Network Analysis Using Hadoop & Neo4j | Friso van Vollenhoven | 21m |
Shows a concrete pipeline that turns raw logs into a Neo4j graph: Hadoop + Cascading to extract unique nodes and edges, Hive joins to merge feature aggregates… | ||
| Machine Learning in the Cloud with Mahout and Whirr | Frank Scholten | 22m |
Shows how to launch mahout cluster on public clouds by having Whirr fetch and install the Mahout tarball, set env vars and mount EBS with Apache archives so… | ||
| Castle-enhanced Cassandra | Eric Evans | 20m |
Castle is a Linux kernel module key‑value store that replaces RAID and POSIX filesystems for Cassandra’s column family store, moving memtables off‑heap and… | ||
| Reporting Systems Built on Top of Hadoop & Lucene | Dragan Milosevic | 21m |
Zanox joins tracking (2M sales, 30M clicks, ~1B views daily) with search-cost feeds using Hadoop, HDFS and MapReduce to produce Lucene indices that power… | ||
| How Hypertable Powers one of the World's Largest Email Services | Doug Judd | 25m |
Rediffmail moves metadata off its SAN into Hypertable to stop outages caused when over 75% of storage requests are inbox metadata updates; headers and user… | ||
| Big Data, Big Performance | Doug Judd | 09m |
HyperTable is a C++ BigTable clone with a Thrift API and production deployments at Baidu and Rediff Mail (100 million users) storing metadata and metrics… | ||
| Implementing a Text Classifier based on Lucene & LIBSVM | Christoph Goller | 22m |
Shows a working text-classification pipeline built on Lucene+LIBSVM that indexes training/test docs in Lucene, extracts bag-of-words features via posting… | ||
| Near Real Time Processing of Time Series Data with HBase | Christian Richter | 14m |
Shows how HBase time‑series writes use a salt prefix to avoid hot regions but break range scans, and how implementing MultiSegmentTableInputFormat lets a… | ||
| Using HBase Coprocessors to Implement Prospective Search | Christian Gügi | 30m |
Prototype embeds HBase coprocessors per table, hooks postPut, converts each put into a Lucene in-memory index document, runs all user queries (agents) against… | ||
| Hadoop: Lessons Learned | Torsten Curdt | 47m |
A practical rundown of running Hadoop in production: a 20-node grep vs MapReduce test (1.5 GB per node) shows MapReduce overhead often outweighs benefit; an… | ||
| Hybrid Applications with MongoDB and RDBMS | Chris Harris | 25m |
Shows a concrete hybrid: front-end JSON services store raw JSON in MongoDB for fewer joins and agile schema changes while transactional or legacy data stays in… | ||
| A JCR View of the World: Everything is Content | Bertrand Delacretaz | 19m |
Shows how Adobe CQ5 based on Apache Jackrabbit and Sling models content as nested nodes and properties and stores code, config and assets in one repository… | ||
| Smart Autocompl ... | Anne Veling | 21m |
Shows a concrete system that replaces a single concatenated Solr fuzzy index with a pre-query matrix that tests each input token against specific fields… | ||
| Updateable Fields and other Fun with Codec API | Andrzej Bialecki | 43m |
Claims that Lucene 4’s Codec API can add updateable fields and other on‑the‑fly behaviors, and demonstrates a prototype that writes updates to separate stacked… | ||
| Real-time Analytics with HBase | Alex Baranau | 20m |
Claims an append-only writes pattern that turns HBase from batch-limited into real-time by replacing Get+Put updates with append records placed next to target… | ||
| HCatalog, Table Management for Hadoop | Alan Gates | 35m |
HCatalog exposes the Hive metastore to Pig and MapReduce so jobs use a shared table abstraction instead of hard-coded HDFS paths and formats. It supplies… | ||
| Profiling & Performance-Tuning Your Hadoop Pipelines | Aaron Beppu | 23m |
Etsy runs hundreds of Cascading flows (thousands of Hadoop jobs daily) and finds concrete wins: replacing Scala parser combinators with a left-to-right parser… | ||
| Scalable Web Applications with a Transactional Key Value Store | Nico Kruber | 19m |
Scalaris is a distributed Erlang key value store that provides transactional, strongly consistent updates so a wiki can atomically update revisions, backlinks… | ||
| Multi-tenant HBase Solutions at Facebook | Nicolas Spiegelberg | 43m |
Facebook runs HBase for multiple production systems including Messages which handles 180 billion messages per month. Infrastructure details: HBase on HDFS with… | ||
| Large Scale, Solr & HBase Performance Monitoring | Otis Gospodnetić | 19m |
A hosted SPM monitors Elasticsearch, Solr and HBase by shipping a collectd agent plus JVM instrumentation to a Jetty receiver, feeding a Flume → HBase pipeline… | ||
| Processing High Volume Updates in Hive | Owen O'Malley | 19m |
A production solution handles 1 TB/day with 15% updates by keeping a base file per bucket and storing edits as separate HDFS files, then using a custom record… | ||
| Eventually Consistent Data Structures | Sean Cribbs | 39m |
Explains why Riak, Voldemort and Cassandra need deterministic data types and presents Convergent/Commutative Replicated Data Types that guarantee a single… | ||
| What You Get by Replicating Lucene Indexes on the Data Grid | Sanne Grinovero | 23m |
Stores Lucene indexes as byte-array fragments inside an Infinispan Directory that replicates across the grid, keeps a single-writer master for commits, and… | ||
| Automata Invasion | Robert Muir | 16m |
Lucene integrates finite-state sets (FSA) and maps (FST/wFST) into queries, term dictionaries and analysis, delivering concrete wins: up to 50× speed or memory… | ||
| ElasticSearch: Big Data, Search and Analytics | Shay Banon | 40m |
ElasticSearch stores huge datasets by treating each shard as the unit of scale (default five shards per index), moving shard files between nodes instead of… | ||
| Efficient Scoring in Lucene | Stefan Pohl | 19m |
Describes a concrete scoring algorithm built on Lucene primitives that implements early-termination / max-score style pruning over inverted lists, with code… | ||
| Hadoop Gets Groovy | Steve Loughran | 20m |
Shows how Groovy and a tiny helper library Grumpy run inside Hadoop to process ~10GB of Bluetooth scan logs, with a sliding-window debounce implemented in the… | ||
| Real-/Neartime analysis with Hadoop & VoltDB | Ralf Neeb | 18m |
Combines Hadoop MapReduce for batch with VoltDB in‑memory for streaming to handle many small events and deliver near‑real‑time analytics. Pipeline moves raw… | ||
| Real Time Datamining and Aggregation at Scale | Ted Dunning | 39m |
Claims that Storm and Hadoop can cooperate to deliver sub-5 ms typical latencies and <5 s worst-case for a sample metrics app while keeping years of durable… | ||
| Lightning IO | Tim Lossen | 02m |
A one-day conference organizes ~50 talks of strictly five-minute slots so a poor talk ends quickly and every attendee becomes a presenter because tickets =… | ||
| Is Your Index Reader Really Atomic or Maybe Slow? | Uwe Schindler | 38m |
Lucene 4.0 replaces the old IndexReader model with AtomicReader and CompositeReader and a segment-oriented DirectoryReader, makes all IndexReaders read-only… | ||
| Closing Session | 04m | |
A wrap-up that contains no technical claims or demos, just logistics and thanks: sponsors (gold, silver, bronze) acknowledged, team credited for program… | ||
| nothing matches | ||