Content about Data management

Opinion: What would Linnaeus do?

Carl Linnaeus on his wedding day, holding in his right hand a specimen of Linnear borealis — his favorite plant, said Magnus Lidén, curator of the Uppsala University Botanic Garden. Image courtesy Uppsala University Art Collection

Elizabeth Leake of TeraGrid — the high-performance, distributed computing network in the USA — was a guest at the EGEE User Forum, the recent conference in Europe on high-throughput computing. Here, she gives her impressions about a technology challenge faced by both types of computing: Long-term, persistent storage.
 
One of the highlights of visiting the conference venue in Uppsala, Sweden, was learning about Carl Linnaeus.
Born in nearby Råshult in 1707, Linnaeus  moved to the college town of Uppsala to study, and quickly became a cornerstone of the university. He died in 1778 and was laid to rest in Uppsala Cathedral — largest in all of Scandinavia.
Linnaeus was fam

Link of the Week: EGEE Screencasts on grid computing

Just in time for the User Forum at Uppsala, the Direct User Support group  produced some educational screencasts that answer common questions about the grid. They are available for viewing here on YouTube and at Vimeo.
Alternatively, you can download them directly to your laptop.
A few samples are
[1] Introduction to Grid Computing[2] Credential management[3] Job submission[4] Data management(An index of such “Use Cases” can be found here.)
 
 
 

Opinion - Challenges to exascale computing

Irving Wladawsky-Berger retired in May 2007 after 37 years at IBM. Today he is a visiting lecturer at MIT’s Sloan School of Management and Engineering Systems Division, a senior fellow at the Levin Institute of the State University of New York, and an adjunct professor in the Innovation and Entrepreneurship Group at the Imperial College Business School.

Supercomputing has been a major part of my education and career, from the late 1960s when I was doing atomic and molecular calculations as a physics doctorate student at the University of Chicago, to the early 1990s when I was general manager of IBM's SP family of parallel supercomputers.
The performance advances of supercomputers in these past decades have been remarkable. The machines I used as a student in the 1960s probably had a peak performance of a few million calculations per second or megaflops. Gigaflops (billions) peak speeds were achieved in 1985, teraflops (trillions) in 1997,

Announcement - Gumboots for the Data Deluge workshop, 12 Feb, Melbourne, Australia

Image courtesy Shannon Pifko, stock.xchng

Places are still available (but not a lot) for “Gumboots for the Data Deluge: Defining and describing collections for the Australian Research Data Commons.”
This half-day workshop will be held in association with VALA 2010 in Melbourne on Friday, 12 February 2010.
Background: The notion of defining and describing data collections for Research Data Australia is a new and challenging one, and part of the task of creating an Australia Research Data Commons that will enable data sharing and re-use of valuable research data collections.
ANDS is building Research Data Australia as a set of interlinked, user-friendly web pages to display this rich descriptive information (metadata), harvested from a range of trusted online sources and held in a federation of online registries.
This workshop is designed to introduce the concepts and practices behind definin

Feature: Grids and clouds - reaching for the next phase

A composite image of the Cat’s Eye Nebula with data from NASA’s Chandra X-ray Observatory (blue) and Hubble Space Telescope (red and purple). This cloud of dust and gas is about 3,000 light-years from Earth. The hybridization of grids and clouds seems considerably closer than that. Image courtesy Smithsonian Astrophysical Observatory

“This is not a replacement technology,” says Ignacio Llorente, “This is the next phase of evolution for grids.”
Llorente coordinates ‘virtual machine’ management  for RESERVOIR, an EC-supported project standing for Resources and Services Virtualization without Barriers, that works to enable deployment and management of complex IT services across different administrative domains. This project began collaborating with EGEE in June 2009 to marry the advantages and practicalities of cloud computing with grid technology.
After starting in February 2

Link of the Week - LarKC: The “Large Knowledge Collider”

Image courtesy LarKC 

The Large Knowledge Collider, or LarKC (pronounced ‘lark’), is part of a €10 million project to promote  “massive distributed incomplete reasoning.”
The goal is to go beyond the limited storage, querying and inference technology currently available, and enable semantic computing — in which data is given meaning (semantics) that  enables computers to look up and “reason” in response to, say, user searches. Or, in an example from Wikipedia — at the moment, a person can create a web page that lists items for sale. The HTML of this person’s catalog page can make simple, document-level assertions such as “this document’s title is ‘Widget Superstore,’ ” but there is no capability within the HTML itself to assert unambiguously that, for example, item number X586172 is an Acme Gizmo with a retail p

Feature - Finding a clue in a data-stack

Data-mining techniques can give law enforcement agencies the edge they need to catch criminals and stop terrorists, but they also raise concerns about respecting personal privacy and civil liberties. New research at Rutgers University has lead to the creation of DI-HOPE-KD, a suite of tools that integrate powerful data mining tactics while guarding civil liberties and due process. Image courtesy Jupiter Images; caption information courtesy of the National Science Foundation

Thanks to a distributed data analysis system under development at Rutgers University, New Jersey, the time it takes police to connect the dots between illicit activities across the globe could go from months to minutes.
Imagine that law enforcement officers in Los Angeles are investigating labs that manufacture the illegal drug methamphetamine. Meanwhile, investigators in Chicago are looking into the illegal sale of large quantities of the drug pseudoephedrine, manufactured i

Feature - A grid-enabled workflow management system for e-Science

Image courtesy Jay Lopez, stock.xchng 

The recent trend towards Service Oriented Architecture — collections of services which operate according to a request/reply model — has stimulated the development of Workflow Management Systems (WMSs) which target the composition of services such as Taverna and Triana.
A new, grid-enabled scientific workflow management system, WS-VLAM, developed in the context of the Virtual Laboratory for e-Science, provides a basic set of tools for building workflows by connecting components to each other based on data dependencies.
The WS-VLAM workflow management system is designed to provide and support the coordinated execution of geographically distributed grid-enabled software components, which can be combined into a workflow. The system takes advantage of the underlying grid infrastructure and unites it with a flexible, high-level, rapid prototyping enviro

Feature - Supercomputing code helps develop new solar cells

Image courtesy of Patrick Moore.

If scientists could use simulations to zoom in on the atomic level of solar cells, the insight they gain could launch solar power into the next energy orbital.
Unfortunately, those simulations would require an exorbitant amount of computational power.
“Typically we need to simulate tens of thousands of atoms,” said Lin-Wang Wang, a scientist at Lawrence Berkeley National Laboratory. “For the conventional code, if the number of atoms increases by a factor of ten, the computational load increases by a factor of a thousand.”
In fact, the same problem arises with nano-scale simulations of a wide variety of materials. That’s why Wang and his research team came up with the LS3DF code.
“We were thinking about how to improve the algorithm and have linear scaling,” said Wang. When an algorithm scales linearly, the computational cost increases at the same rate

Q & A - Smart data handling: An interview with Tevfik Kosar

Image courtesy Tevfik Kosar

In e-science, we are constantly striving to improve performance and speed so that we can complete a larger number of more complex computations faster. Tevfik Kosar, a researcher at Louisiana State University, is working on two intertwined projects that could together lead to the sorts of improvements we hope for. Read on to find out what he had to say.
iSGTW: You just received a National Science Foundation grant to work on the Stork Data Scheduler. Can you tell us a little about that project?
Kosar: The funding for the development of Stork Data Scheduler comes from NSF's Strategic Technologies for CyberInfrastructure program. The STCI program funds innovative cyberinfrastructure services which have the potential to significantly advance research capabilities in multiple areas of science and engineering. The grant will provide three years of funding for the enhancement of the Stork Data Scheduler

Feature - Clarity in clinial care: achieving fully interoperable databases

Werner CeustersImage courtesy of W. Ceusters.

Werner Ceusters, director of the  Ontology Research Group at the NYS Center of Excellence in Bioinformatics & Life Sciences, spoke at the Bio-IT World Conference in Boston this past April on the topic of "Do Healthcare IT Standards Hamper the Advance of Science?" 

“Swimming is healthy and contains eight letters.” Absurd? Of course — the real thing is confused with its symbolic representation. Absurdity, then, abounds in current Electronic Health Records (EHRs) according to Werner Ceusters of the New York State Center of Excellence in Bioinformatics and Life Sciences and the University at Buffalo. A polyp by any other name… A significant and universal problem with legacy EHRs, Ceusters says, is that software programs that access them cannot adequately differentiate between diagnoses for a given patient and link them prope

Feature - Poker, parallelism and payoff

Image courtesy of PSC

You won’t find them at the Vegas casinos and what they do is hard to call gambling, but it’s fair to say that Tuomas Sandholm and grad student Andrew Gilpin of Carnegie Mellon’s School of Computer Science are professional poker players. Last July in Chicago — at the AAAI (Association for the Advancement of Artificial Intelligence) Computer Poker Competition, involving 19 programs from six countries — they walked away with no pile of cash but, nevertheless, were the biggest winners. Their field, game theory, in which Sandholm’s work is internationally recognized, describes conflict in which the payoff is affected by actions and counter-actions of intelligent opponents. Head-to-head poker between two players is a prominent example of what’s called two-person zero-sum games: One player wins what the other loses.
In recent years, poker has emerged as an AI challenge much as chess was for