Edinburgh Data-Intensive Research - Data-intensive computing

Research topics:

Congratulations: Gary McGilvary wins competition at CERN

Malcolm.Atkinson — Thu, 29 Aug 2013 15:08:20 +0000

Congratulations to Gary, who as his hard-won internship in CERN approaches its end, has won a prize in the lightening presentations given by the 20 interns at CERN yesterday. His talk "The Implementation of OpenStack Cinder and Integration with NetApp and Ceph" was one of the three talks awarded a prize:
http://openlab.web.cern.ch/news/openlab-summer-students-present-their-pr...

We're all looking forward to Gary's return next week, when he will tell us what that means; rumour has it that he's enjoyed his internship.
Malcolm

Topic of this submission:

News

Research topics:

Multi-Workflow Systems and Editors

Sandra.Gesing — Thu, 27 Jun 2013 11:39:10 +0000

Speaker(s):

Sandra.Gesing

Presentation Type:

talk

Researchers often need to use workflows that have been developed by other experts in their field to handle specific parts of their work. Sooner or later they find that they want to use workflows from multiple sources that are written in different languages. Enacting multi-lingual workflows (or meta workflows) has been pioneered in a group of European projects. The next step is to be able to change them when they don’t do exactly what you want. But that is not easy if you need to learn a different editor for each workflow language. The talk will conclude with the status of research underway to address this problem.

Attachment	Size
osdc2013WorkflowsSandraGesing.pdf	1.72 MB

Date and time:

Wednesday, 19 June, 2013 - 11:30

Location:

OSDC PIRE Workshop 2013, Edinburgh, UK

Web site:

OSDC PIRE Workshop

Research topics:

Biological/Medical Image Processing

Intuitive interfaces

MoSGrid - Molecular Simulations in a Distributed Environment

Sandra.Gesing — Thu, 27 Jun 2013 11:33:02 +0000

Speaker(s):

Sandra.Gesing

Presentation Type:

talk

Molecular simulations are indispensable methods in areas like material science, structural biology, and drug design. These methods address data-intensive and compute-intensive problems, which demand high-performance computing to allow data analysis in an acceptable time. The project MoSGrid (Molecular Simulation Grid) offers a workflow-enabled grid portal allowing access to molecular simulation tools on distributed resources in an intuitive manner. Users are able to exchange workflows and data via repositories and, thus, to exchange knowledge about the specific application domain. The talk will give an overview on the current MoSGrid portal, the underlying infrastructure, and an outlook on the next steps.

Attachment	Size
osdc2013MoSGridSandraGesing.pdf	4.77 MB

Date and time:

Tuesday, 18 June, 2013 - 12:00

Location:

OSDC PIRE Workshop 2013, Edinburgh, UK

Web site:

OSDC PIRE Workshop

Research topics:

Intuitive interfaces

Supporting Collaborative Scientific Workflow Development:The Dispel Information Registry

Iraklis.Klampanos — Fri, 21 Jun 2013 16:35:55 +0000

Presentation Type:

talk

Large-scale distributed workflow systems for science are nowadays expected to operate in a consistent, predictable way as well as to promote collaboration between researchers or within groups in a unified way. In this talk we will discuss the VERCE Information Registry, which is designed to provide a consistent view of the VERCE ecosystem for seismology along with related architectural requirements, assumptions and interactions with other components.

Attachment	Size
Registry-OSDC13.pdf	1.65 MB

Date and time:

Thursday, 20 June, 2013 - 11:40

Location:

Open Science Data Cloud Workshop 2013, Edinburgh, UK

Web site:

Projects:

Research topics:

Efficient distributed systems

Seismology

The VERCE Architecture for Data-Intensive Seismology

Iraklis.Klampanos — Fri, 21 Jun 2013 16:32:30 +0000

Speaker(s):

Presentation Type:

talk

Typical of the digital revolution, seismology is changing rapidly: the number of deployed seismometers grows rapidly, their performance and connectivity improve and seismologists employ data from other sources, such as LIDAR, satellite images and GPS. But their computational behavior is also changing: they used to just focus on earthquakes, today the use all of the continuous waveform. Just what does this mean for the science and the computational infrastructure to support it? The VERCE project is pioneering this approach and will provide examples during the talk.

Attachment	Size
osdc-verce.pdf	1.97 MB

Date and time:

Wednesday, 19 June, 2013 - 11:00

Location:

Open Science Data Cloud Workshop 2013, Edinburgh, UK

Projects:

Research topics:

Seismology

The VERCE Architecture for Data-Intensive Seismology

Iraklis.Klampanos — Mon, 10 Jun 2013 15:16:29 +0000

Speaker(s):

Presentation Type:

talk

One of the main objectives of the VERCE project (Virtual Earthquake and Seismology Research Community in Europe) is to provide scientists with a unified, Europe-wide, computing environment able to support data-intensive scientific computation. This talk will be mainly about our approach to designing and building this infrastructure. More specifically, I will present the current computing environment, the rationale for designing our solutions around the workflow paradigm as well as the basic components of the architecture and their interactions. I will conclude with research challenges we have encountered and how we propose to address them in the short- and longer-term.

Attachment	Size
cisa_2013-06-10.pdf	1.99 MB

Date and time:

Monday, 10 June, 2013 - 14:00

Location:

CISA Seminar, School of Informatics, University of Edinburgh, UK

Web site:

Projects:

Research topics:

Efficient distributed systems

Seismology

Data-Intensive Research for Efficiency and Enablement in Heterogeneous Systems

Iraklis.Klampanos — Wed, 05 Jun 2013 12:23:15 +0000

Speaker(s):

http://conference2013.chistera.eu

Presentation Type:

poster

Diversity, in every dimension, is a key attribute of today’s data bonanza. Our research takes a holistic view, embracing this diversity and the consequent intricate interactions between users and systems. We created the Dispel data-streaming language to describe complex computation patterns at high levels of abstraction, while providing meta-information for optimisation. Provenance and contextual information must be harnessed to achieve autonomous execution, data placement, energy efficiency and reliability. Research is needed, that builds foundations for comprehensible and tractable paths through the rapidly evolving data landscape. This must be driven by and tested at realistic scales and diversity levels.

Attachment	Size
chist-era.pdf	8.37 MB

Date and time:

Thursday, 13 June, 2013 - 09:00

Location:

CHIST-ERA Conference, Brussels, Belgium

Web site:

Research topics:

Efficient distributed systems

Effective algorithms

Reusable computational models

Towards Supporting Service-Oriented Seismology Research in Europe

Iraklis.Klampanos — Fri, 31 May 2013 15:46:40 +0000

Speaker(s):

Presentation Type:

poster

Modern seismologists are presented with increasing amounts of data that may help them better understand the Earth's structure and systems. However: 1) they have to access these data from globally distributed sites via different transfer protocols and security mechanisms; 2) to analyse these data they need to access remote powerful computing facilities; 3) their experiments result in yet more data that need to be shared with scientific communities around the world.

In the vast majority of cases, the labourious tasks of data management, transfer and execution of scientific codes is handled manually by the scientist and on an ad hoc basis. This hinders seismologists from making full use of the data and tools they have at their disposal for scientific discovery.

The VERCE project is designing and developing a research platform to deliver a seismology community e-Infrastructure, an integrated computational and data environment that presents a coherent virtual environment in which to conduct research.

Actions and deliverables in VERCE are driven by two broad seismology use-cases, selected for being representative of resource usage in the designed infrastructure: 1) ambient noise cross-correlation and 2) forward modelling and inversion. Use-case 1 has inherent data-intensive characteristics, as it requires access to and orchestration of multiple, diverse data sources, combining files, web services, etc. Use-case 2 is less data-intensive but it requires combining data management and signal processing with processing on remote clusters and HPCs.

VERCE solutions are designed around a workflow representation paradigm, employing the stream-oriented Dispel workflow specification language. A VERCE-wide registry will provide specifications of Dispel programming and other research components throughout the infrastructure, while also accommodating research collaborations. Depending on the needs and specifications of the participating sites, VERCE will aim to support different execution engines, such as OGSA-DAI. All of the above will be made available to researchers in a transparent and unified way, through a Science Gateway backed up by data and resource discovery, meta-data and searching functionalities.

In order to meet its goals and objectives, while also creating a sustainable infrastructure that may eventually be part of EPOS - the European Plate Observing System, VERCE is actively collaborating with a number of initiatives and projects. As VERCE spans high-performance and high-throughput computing, represented by PRACE and EGI respectively, as well as it has demanding data management and discovery requirements, addressed by EUDAT, we are participating in a number of pilot case-studies within an EGI-PRACE-EUDAT synergy. In addition, we are collaborating with the SCI-BUS project, in order to analyse and meet our Science Gateway requirements while staying as flexible as possible with regards to middleware interoperability and workflow frameworks.

Attachment	Size
cloudscapeV.pdf	5.94 MB

Date and time:

Wednesday, 27 February, 2013 - 09:00

Location:

CloudscapeV, Brussels, Belgium

Web site:

Projects:

Research topics:

EFFORT: Earthquake and Failure Forecasting in Real Time Forecasting Model Testing Centre

Rosa.Filgueira — Fri, 31 May 2013 15:35:09 +0000

Speaker(s):

Rosa.Filgueira

Presentation Type:

poster

PICO presentation:
Across Europe there are a large number of rock deformation laboratories, each which runs many experiments. Similarly
there are a large number of theoretical rock physicists who develop constitutive and computational models
both for rock deformation and changes in geophysical properties. Here we consider how to open up opportunities
for sharing experimental data in a way that is integrated with multiple hypothesis testing. We present a prototype
for a new forecasting model testing centre based on e-infrastructures for capturing and sharing data & models to
accelerate the Rock Physicists research.

This proposal is triggered by our work on data assimilation in the NERC EFFORT(Earthquake and Failure Forecasting
in Real Time) project, using data provided by the NERC CREEP 2 experimental project as a test case.
EFFORT is a multi-disciplinary collaboration between Geoscientists, Rock Physicists and Computer Scientist.
Brittle failure of the crust is likely to play a key role in controlling the timing of a range of geophysical hazards,
such as volcanic eruptions, yet the predictability of brittle failure is unknown. Our aim is to provide a facility for
developing and testing models to forecast brittle failure in experimental and natural data. Model testing is done in
real-time, verifiably prospective mode, in order to avoid selection biases that are possible in retrospective analyses.
The project will ultimately quantify the predictability of brittle failure, and how this predictability scales from
simple, controlled laboratory conditions to the complex, uncontrolled real world. Experimental data are collected
from controlled laboratory experiments which includes data from the UCL Laboratory and from Creep2 project
which will undertake experiments in a deep-sea laboratory. We illustrate the properties of the prototype testing
centre by streaming and analysing realistically noisy synthetic data, as an aid to generating and improving testing
methodologies in imperfect conditions. The forecasting model testing centre allows to:

1. Data transfer:
• Upload experimental data: We have developed FAST (Flexible Automated Streaming Transfer) tool
to upload data reliable and periodically from RP laboratories to the data repository. FAST allows for
setting up data transfer requirements and selects automatically the most suitable transfer protocol.
Metadata are created and stored in the data catalogue.
2. Web data access:
• Create synthetic data: The users can choose the generator and supply parameters. Each synthetic data
is automatically stored in the data repository which corresponding metadata in the data catalogue.
• Select data and models: The metadata of each data (synthetic or from laboratory) and models are welldescribed
through their respective catalogues accessible by the web portal. Searches over the metadata
are supported.
• Upload models: The users can upload their models and them will be automatically stored in the models
repository. The users have the opportunity to share their models among other users. The web portal
creates and stores the metadata of each uploaded model into the models catalogue.
• Run models and visualise results: Once a user has selected data and a model, the model is submitted
to a High Performance Computational resource, hidden the complex interaction with clusters. Their
results are displayed in accelerated time. The results of the computations are stored allowing retrieval,
inspection and aggregation.

3. Metadata and Data Storage:
• Store automatically metadata in catalogues and data, models and results in repositories.
The forecasting model testing centre proposed could be integrated into EPOS context and its expected benefits are:
• Facilities the prediction of brittle failure and its scalability to natural phenomena.
• Provides fast testing and ideas propagation.
• Increases the impact and visibility of RP and GS research.
• Provides important resources for education and training.

Attachment	Size
Rosa-PICO-EGU2013.pdf	2.76 MB

Date and time:

Monday, 8 April, 2013 - 10:00

Location:

Vienna

Web site:

Projects:

EFFORT

Research topics: