Let the benchmark wars begin! :-)
(Friendly competition, of course.)
On 11/22/13 5:35 AM, Till Westmann wrote:
We are not alone ...
Begin forwarded message:
From: "Ioana Manolescu" <[email protected]>
Date: November 22, 2013 at 4:21:15 PST
To: [email protected]
Subject: [Dbworld] Engineer position in Inria Saclay, France
Reply-To: [email protected]
PAXQuery integration into Stratosphere
Context
The Oak team (http://team.inria.fr/oak/), located in Orsay
(south of Paris), brings together Inria and University of Paris Sud
faculty working on models, languages, and algorithms,
for efficiently processing complex data.
In particular, our recent research has focused on efficient
algorithms for massively parallel processing of Web data.
To scale up the processing of vast volumes of Web data, programming
models built on top of shared- nothing architectures such as MapReduce
offer interesting opportunities. However, MapReduce is very simple
both in its computational capacity and in the data types it supports,
making it a rather poor match for complex operations. This has lead
to the development of richer parallel data processing stacks, among
which Stratosphere (http://www.stratosphere.eu), developed at
TU Berlin. S
tratosphere is an analytics platform for high-volume, high-velocity
heterogeneous data, with the unique selling proposition to automatically
optimize and parallelize complex data analytics programs that involve
user-defined functions, iterative processing and shared-state.
PAXQuery
Over the last two years, we have collaborated with the Stratosphere
team within the EIT ICT Labs Europa activity.
We have developed PAXQuery, a system for processing complex structured
data such as XML documents by compiling XQuery queries into parallel
programs executed within Stratosphere. PAXQuery is currently at the
level of a prototype, capable of translating from an internal algebraic
representation of XQuery, into the internal task description dialect
that the Stratosphere platform handles.
Job description
For the PAXQuery project, we currently seek to hire for one year an
Inria engineer which will be responsible for the following tasks:
- Consolidating and extending our PAXQuery prototype implementation,
notably to accept as input standard XML query syntax (XQuery).
- Developing a thorough test platform, raising the code maturity level so
that it can be released in the open-source distribution of Stratosphere.
- Writing the platform documentation, providing enough resources to
kick-start a community of users around it.
Candidate profile.
The ideal candidate is able to work independently and has strong problem
solving skills. You are motivated by quality, excellence and results.
You want to contribute to an open-source project at the forefront of big
data analytics, and you are willing to engage with an active community
ready to use the result of your work.
Employment period and salary
The position is from January to December 2014; we may consider shifting
the period by 1-2 months to suit a particularly interesting candidate.
The gross salary is 2500 Euros/month.
Requirements
- BS/MS in Computer Science (or equivalent).
- Strong Java skills.
- (Essential:) good knowledge of compiler design and construction.
- Good knowledge of Linux and bash shell scripting.
- Familiar with version control systems (SVN/GIT), deployment tools
(ANT/Maven) and unit testing frameworks (JUnit) is a plus.
- Previous experience with Hadoop or Stratosphere is a plus.
How to apply
Send a resume highlighting relevant knowledge and experience to:
jesus.camacho [email protected] and [email protected].
_______________________________________________
Please do not post msgs that are not relevant to the database community at
large. Go to www.cs.wisc.edu/dbworld for guidelines and posting forms.
To unsubscribe, go to https://lists.cs.wisc.edu/mailman/listinfo/dbworld
