Berlin Buzzwords 2016
2016-06-05..2016-06-07
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: Inspiring the Next Generation ... #bbuzz | Diane Mueller-Klingspor | 45m |
A rural program converts a 1965 LJ trailer into a mobile makerspace stocked with 3D printers, a human-sized turntable and an Xbox Kinect to create 1965 LJ… | ||
| Help I need a stream processor - learning how to chose ... | Andrew Psaltis | 31m |
Deploys a smart‑meter example where commercial meters jump from one record per month to ~8,500/month and residential to ~2,500–3,000/month, then compares… | ||
| Application development and data in the emerging world ... | Neha Narkhede | 50m |
Argues that stream processing becomes core application logic and shows a concrete stack: Apache Kafka as a durable pub/sub and storage layer, log compaction… | ||
| Migrating a data stack from AWS to Azure #bbuzz | Dániel Molnár | 39m |
Migrates a full AWS data stack to Azure by replacing Hadoop/EMR and Redshift-centered ETL with Unix tooling: make plus Python, a homegrown Junior Beaver… | ||
| BM25 demystified #bbuzz | Britta Weber | 37m |
Lucene and Elasticsearch change the default ranking from TF/IDF to BM25 and BM25 replaces unbounded term frequency with a k parameter that makes term-frequency… | ||
| Predictive Maintenance: From POC to Production With Spark | Heloise Nonne | 43m |
Proof of concept runs in 10 weeks (2 for discovery, 8 to build) and moves to a six‑month pilot; models use random forests and experiments with neural networks… | ||
| Community & Commercialization: How to Build an Open Source Company | Will Hayes | 39m |
Lucidworks builds a commercial business on Apache Lucene/Solr, contributing 70–80% of the codebase and employing about 40% of active committers, with Solr… | ||
| Towards consensus on distributed consensus #bbuzz | Flavio Paiva Junqueira | 38m |
Explains when distributed consensus is actually required versus cheaper coordination, using ZooKeeper, etcd/Consul and Raft examples and two production… | ||
| Business Intelligence in Microservice architecture #bbuzz | Debarshi Basak | 32m |
Bol.com moves BI from a monolithic warehouse to a hybrid microservice model using Hadoop/HBase (26 nodes, ~1.5 PB), Flink for streaming, and Sqoop/Pig for… | ||
| Leveraging blockchain technologies for the internet of things. | Sam Bessalah | 30m |
Argues that blockchains can secure machine-to-machine IoT interactions by running Ethereum smart contracts on constrained hardware, using the Go geth… | ||
| Scaling Yelp’s Logging Pipeline with Apache Kafka #bbuzz | Enrico Canzonieri | 37m |
Legacy scribe aggregators get replaced by Kafka as the single source of truth, with a scribe-to-Kafka connector called secretary buffering producers on hosts… | ||
| Learning the learner: Using machine learning to track performance | Ira Cohen | 31m |
Anodot builds an automated monitoring layer that uses anomaly detection to track dozens of live models, claiming it instruments 30 algorithm types and watches… | ||
| Household INFOSEC in a Post-Sony Era #bbuzz | Steve Loughran | 30m |
Home security hardens around DD‑WRT router and subnet isolation but still fails: an LG smart TV leaks channel viewing, phones and apps (British Airways… | ||
| Shaping Applications for Docker, CoreOS, Kubernetes & Co. | Thomas Fricke | 37m |
Real projects use Docker, CoreOS and Kubernetes to run a messaging service (IMMA) with Java, Python and Ruby; environments enforce a strict stateless vs… | ||
| Level Up: Everything you need to know about search, ... | Grant Ingersoll | 40m |
Argument: modern search moves beyond keywords by combining Lucene/Solr, Spark, Mahout, NLP and machine learning to boost relevance and cut infrastructure… | ||
| Keyvi the Key Value Index Optimized for Size and Speed #bbuzz | Hendrik Muhs | 32m |
Keyvi is a finite-state key value index used by Cliqz to serve ~6,000 requests/min with ~60 ms average latency and 99% availability. It stores 1.8 billion URLs… | ||
| Running High Performance And Fault Tolerant ElasticSearch ... | Rafał Kuć | 22m |
Shows how to run Elasticsearch inside Docker with concrete commands, constraints and failure handling: use Docker run with environment vars and -m 2G memory… | ||
| Real Time Marketing with Kafka, Storm, Cassandra & Spark #bbuzz | Volker Janz | 35m |
A realtime CRM pipeline sends game events through Kafka into Storm where JavaScript handlers update per-day player profiles in Cassandra and emit targeted… | ||
| Streaming analytics and CEP ... #bbuzz | Fabian Hueske, Till Rohrmann | 38m |
Apache Flink claims a single engine can handle both low-latency CEP and high-throughput stream analytics, running millions of events per second at sub-second… | ||
| A Data Streaming Architecture with Apache Flink #bbuzz | Robert Metzger | 43m |
Shows converting a batch ETL pipeline into a streaming ETL built with Apache Flink reading from Apache Kafka and writing to Elasticsearch, Cassandra and a… | ||
| Learning to Rank: where search meets machine learning #bbuzz | Andrew Clegg | 38m |
Introduces Ranking SVM applied to web search: trains on pairwise training examples derived from click logs, learns feature weights (tfidf, metadata, CNN image… | ||
| Live-Hack: Analyzing 7 years of Buzzwords at Scale #bbuzz | Christoph Tavan | 39m |
Scrapes 2010–2016 Berlin Buzzwords sites with Scrapy, stores line‑delimited JSON on S3, and analyzes it in Databricks + Spark using regex cleanup, a stopword… | ||
| SMACK Stack - Data done Right #bbuzz | Stefan Siprell | 41m |
SMACK combines Spark, Mesos, Akka, Cassandra and Kafka into pipelines that aim for second‑level reaction on large context rather than microsecond trading… | ||
| Google Dataflow: New open model for batch & stream processing | Felipe Hoffa | 41m |
Explains that Google Dataflow becomes an open model now incubating as Apache Beam and shows a working pipeline reading Wikipedia edits from Pub/Sub, counting… | ||
| Gain speed and space with NLP in Solr #bbuzz | Tobias Kässmann | 20m |
With Apache Solr at Shopping24 an index-time pipeline drops 60 million plus product noise to scale from 3M toward 100M by storing only significant terms. A… | ||
| Graphs as Streams: Rethinking Graph Processing in ... #bbuzz | Vasia Kalavri | 19m |
Argues that graphs do not need full storage to be analyzed and demonstrates a working prototype built on Apache Flink called jelly stream that runs… | ||
| Stream Processor as a Database: Building Online Applications | Stephan Ewen | 41m |
Claims that serving live aggregates directly from the stream processor removes the KV-store bottleneck: prototype uses Apache Flink + Kafka with a Queryable… | ||
| Wikimedia Content API: A Cassandra Use-case #bbuzz | Eric Evans | 40m |
Wikimedia stores roundtripable HTML generated by Parsoid and cached by Restbase as durable blobs in Cassandra to serve low-latency editors and mobile clients… | ||
| Acceptably Inaccurate: Probabilistic Data Structures #bbuzz | James Stanier | 39m |
Compares naive in-memory Java sets (≈264 MB heap for ~1M UUIDs) and multisets (≈234 MB) to compact probabilistic structures: Bloom filters via Guava give ~0.9… | ||
| The Surprising Course of Knowledge and Innovation #bbuzz | Ellen Friedman | 27m |
Argues that disciplined measurement, record‑keeping and asking “what changed” beat vague theory: historical case shows Matthew Maury aggregates of ship logs… | ||
| Apache Kudu (incubating): Fast Analytics on Fast Data #bbuzz | Todd Lipcon | 40m |
Apache Kudu is a C++ tabular storage engine that uses Raft-replicated tablets (default three replicas) and removes the HDFS layer so clients talk directly to… | ||
| Fast Cars, Big Data - How Streaming Can Help Formula 1 #bbuzz | Ted Dunning | 40m |
A physics-based Spa‑Francorchamps simulator generates realistic F1 telemetry and streams it through MapR Streams (Kafka-compatible) into MapR‑DB, visualized… | ||
| What We Talk About When We Talk About Distributed Systems | Alvaro Videla | 43m |
Survey of consensus and replication literature that argues FLP impossibility means no deterministic consensus in a fully asynchronous model even with one… | ||
| The world wide 60B transaction per day journey | Frank Conrad, Ranbir Chawla | 36m |
Claim: AudienceScience scales its Helios stack to 1,000,000 transactions per second and handles 80 billion requests per day across six global data centers. The… | ||
| OSv: Probably the best OS for Cloud workloads you've ... | Roman Shaposhnik | 41m |
OSv is a POSIX-focused unikernel that boots as a single C++11 image on KVM/VirtualBox/VMware, requires only rebuilding apps as a shared object and packs a… | ||
| Parallel SQL & Streaming Expressions in Apache Solr 6 | Shalin Shekhar Mangar | 39m |
Solr 6 exposes Parallel SQL compiled to a Streaming Expressions plan and served over /sql and /stream with a thin JDBC client; queries run in two modes —… | ||
| Real-time analytics with Flink and Druid #bbuzz | Jan Graßegger | 20m |
A streaming-only pipeline connects Kafka -> Flink -> Druid to deliver sub-second multidimensional queries at 70,000 events/sec with records carrying up to 50… | ||
| Improve your marketing reach using ... | Nina Bertrand, Matthieu Vautrot | 24m |
Case study using a Hadoop/Spark cluster on 6,000,000 clients builds a daily test-and-learn pipeline: Hive/Pig for ETL, Apache Spark / Spark ML for model… | ||
| Efficient Data formats for Analytics with Parquet and Arrow | Julien Le Dem | 46m |
Compares Parquet (disk-optimized) and Arrow (memory-optimized) with concrete tradeoffs: Parquet uses columnar storage, statistics and encodings like dictionary… | ||
| Introducing Kafka Streams, new stream processing library ... | Michael Noll | 37m |
Kafka Streams is a native Java library that runs as part of Kafka (released with Kafka 0.10), not a separate cluster: you scale by launching identical app… | ||
| Measuring 2.0 – Building the next generation webanalytics ... | Niels Basjes | 37m |
Retail site replaces JS-only analytics with a hybrid server+client pipeline that tags every page element, ships events via Kafka and writes session files to… | ||
| ExpAn: A Python library for advanced statistical analysis ... | Dominic Heger | 32m |
ExpAn is an open source Python library (MIT, pip, GitHub) that creates an experiment object from metadata, features and KPIs and returns a pandas DataFrame… | ||
| Big Data, Small Code ... #bbuzz | David Whiting | 34m |
Claims that Java 8 plus Apache Crunch (now shipping the crunch-lambda wrapper in 0.14.0) cuts MapReduce pipeline boilerplate, keeps execution on Hadoop (or… | ||
| Real-World NoSQL Schema Design #bbuzz | Tugdual Grall | 41m |
Converts the MusicBrainz relational model into denormalized NoSQL documents and column families, cutting MusicBrainz -> 138 tables to 4 by embedding aliases… | ||
| Event Sourcing in Yammer | Michal Rutkowski, Dmitry Stratiychuk, Philipp Fehre | 41m |
Yammer prototypes event sourcing for a year, then replaces a Kafka-based pipeline with a managed Azure Event Hubs backend and an in-house RX Java SDK (built on… | ||
| Shipping at Scale: ChatOps at GitHub #bbuzz | Georgi Knox | 18m |
GitHub uses Hubot to automate deploys, incident response and everyday ops across a 550-person company and a global workforce; repositories trigger CI and chat… | ||
| Apache Lucene 6: What's coming next? #bbuzz | Uwe Schindler | 40m |
Lucene 6 introduces points (dimensional values) as a new, codec-backed data type replacing NumericRangeQuery for faster 1–8D queries (IPv6/128‑bit ranges… | ||
| Causality is simple #bbuzz | Carlos Baquero | 33m |
Argues that causality is simple by reducing it to sets of unique event IDs (a causal histories model) and showing how those sets compress into vector clocks… | ||
| Building a real-time news search engine #bbuzz | Ramkumar Aiyengar | 42m |
Bloomberg replaces a legacy backend with a custom real‑time news search built on Solr/Lucene, indexing ~500 million stories, ingesting peaks of 500… | ||
| nothing matches | ||