Big Data poses challenges to software developers - Tampere University of Technology

Big Data poses challenges to software developers

Shuvra S. Bhattacharyya does not identify himself as a world healer. The ever-changing field raises new challenges on a daily basis, and he seeks to meet them by developing useful new design methods, tools, and applications for the needs of the industry

In a FiDiPro project by the TUT Department of Pervasive Computing, effective measures are developed for designing embedded systems that process vast data masses. This endeavour is driven by the extensive use of Big Data, ubiquitous sensor technology, and increasingly complex processing devices.

Shuvra S. Bhattacharyya’s FiDiPro term started in January 2015, but the cooperation between the Electronics Professor of the University of Maryland and Finnish researchers set off over a decade ago.

“I have visited TUT in several short-term periods, and this current project builds upon our previous work at the Department of Pervasive Computing. The three-year funding by Tekes allows us to develop more extensive systems and software tools that bring major improvements to the processing of vast data masses and the data flow generated by information networks,” Bhattacharyya explains.

Shuvra S. Bhattacharyya, 47 years

  • FiDiPro Professor at the Department of Pervasive Computing
  • Full Professor at the Department of Electrical and Computer Engineering at the University of Maryland, College Park, USA; joint appointment at the Institute for Advanced Computer Studies (UMIACS), University of Maryland.
  • Education: PhD in Electrical Engineering and Computer Sciences (EECS), University of California at Berkeley, USA, 1994.
  • Research interests: signal processing, embedded systems, electronic design automation, wireless communication and wireless sensor networks.
  • Family: Wife Arundhati, son Arpan (22 years), daughter Arushi (14 years).
  • Hobbies: Reading, cooking.

Big Data accumulates quickly, it is extensive in volume and it is versatile in structure. This means that refining it for useful application necessitates effective processing. The data is not the researchers’ only problem, however.

“What poses our biggest problems are the complex systems that need to react dynamically to the changing data and operational requirements. Also, we must continuously take the effective use of heterogeneous processing resources into account and minimize the energy consumption of the system throughout the design process.”

Evolution in electronic devices occurs rapidly, which is why design processes must facilitate efficient retargetability across different types and generations of hardware platforms.

Industry expects results

“TUT’s strong history of innovation in signal processing methods and digital systems is an apt combination when it comes to solving Big Data problems.”

And indeed, solutions are eagerly awaited. This has been very clear in the project steering group which includes representatives from several companies utilizing or developing technologies for Big Data.

 “Some of our research results benefit the companies quickly, others in the longer term. It is often relatively easy for a company to incorporate a new type of hardware or software subsystem in an application, for example, but the deployment of new design processes and tools requires larger efforts from companies, and that does not happen overnight,” Bhattacharyya explains.

In addition to embedded systems, the data processing structures and methods produced in the project can be used in cloud computing and scientific computing applications.


Updated by: Merja Jaaksi, 22.12.2015 10:44.
Content owner: Laukkanen Tuuli
Keywords: science and research, services and collaboration, working at tut, fidipro