Fast Feather Track - ApacheCon EU 2014

The Fast Feather Track is a series of short talks, ~15 minutes in length, covering things that are new / interesting / exciting / recently changed / etc. It's a great chance to learn about what's hot, what's new, and what's coming soon! Whether that's a new project entering the incubator, something shortly to graduate, or an amazing new area of technology that'll change everything when it's done, the Fast Feather Track is the place for it!

We've welcomed submissions from new speakers, as the Fast Feather Track can be a great way to gain experience in presenting at ApacheCon in advance of a full slot in future years. We're also very happy for old faces to give talks too!

All talks are 15 minutes in length, with 5 minutes change-over between sessions. Speakers can use as much or as little of their slot for questions as they like!

The Fast Feather Track is taking place in Krudy, which in the group of rooms to the left as you come up the stairs from reception (about as far from the exhibition space as possible!)

Julien Nioche - Web crawling on Storm

I'll present the storm-crawler project, which is a SDK for building scalable, low-latency, customisable web crawlers on Apache Storm and explain how it compares to existing resources such as Apache Nutch and how it integrates with other projects from the big data ecosystem.

The presentation will contain descriptions of how the project is used by several companies."

Projects: Apache Storm, Nutch, Tika. Search-related projects such as SOLR or ElasticSearch. Storage (HBase/Cassandra). Queues (Kafka)

Rohit Yadav - Citizen Cloud Computing with CloudStack

How one can use CloudStack to build a low-cost Cloud for themselves and their friends and share dedicated server for the win! The talk consists of mostly demoing CloudStack and CloudMonkey (the CLI of CloudStack) and sharing a new idea of Citizen Cloud Computing that empowers small group of friends and startups to manage and run a Cloud of VMs on a dedicated server that is low-cost (say lower than even DigitalOcean).

Projects: Apache CloudStack

Justin Mclean - Creating your own Open Source Hardware

It eases that you think, with very little experience and skills, to design and fabricate your own PCB to create your own OSHW Arduino Shield and to even scale that up to manufacture in low volumes. All you need is the the right tools and a tiny bit of knowledge.We'll go from breadboard to prototype to making a few working prototypes to making in small volumes (10 boards) to low volume manufacture (100 boards) with minimal effort and cost. None but other open source technologies are used. Perhaps the question is why isn't Apache more involved in this space?

= Shahbaz Memon - Accessing supercomputers with UNICORE=
UNICORE (unicore.eu) is an open source middleware that provides scientists with a seamless, secure and standards based access to a wide range of computational resources, in both high performance computing (HPC) or high throughput computing (HTC) deployments. UNICORE is based on service-oriented architecture, therefore most of the UNICORE's significant capabilities such as job management and data access functions are exposed as a set of web services.

Projects: Ant, Maven, XML-Beans, Commons, CXF, Geronimo

Nick Burch - How Big is Big : Tall, Grande, Venti Data

Apache has a wide range of Big Data projects, some suitable for smaller problem sets, some which scale to huge problems. Today though, that one label "Big Data" can cause confusion for new users, as they may struggle to pick the right project for the right scale for their problem.

Do we need new titles for different kinds of Big Data? Does the buzz and VC funding cause confusion? Is the humble requirement dead? Or can we help new users better find the right Apache project for them?