Prof. Roland Jochem
Technical University Berlin, Germany
Jochem is the Head of Chair Quality Science,
Institute of Machine Tools and Factory
Management (IWF), Technical University
From 1987-1989 he worked as a project engineer in a medium-sized engineering company. Then he worked on research projects at the IWF, TU Berlin, and consulting projects at Fraunhofer-Institute for Production Systems and Design Technology (IPK). In the period 2002-2005 Prof. Jochem performed optimization projects in the area of product creation processes at Bosch Siemens Hausgeräte GmbH. From 2005-2010 he hold the chair on Quality Management at University of Kassel, Germany. Since 2010 he is Extraordinary Professor for Quality Management at Stellenbosch University, South Africa.
His research and consulting focuses are model-based process-oriented quality, standardized quality processes, quality management, quality excellence, organizational design of quality departments, innovation and requirements management. Prof. Jochem is author of eleven books; Author of more than 80 Articles published in Journals, Conference Proceedings or Edited Books.
Assoc. Prof. Leila Ismail
UAE University, United Arab Emirates
Dr. Leila Ismail
is the Founder and Head of the Distributed
Computing and Systems Research Laboratory at
College of Information technology of the
United Arab Emirates University (UAEU), and
Associate Professor at the Department of
Computer Science and Software Engineering
that she joined in 2005. Dr. Ismail
completed advanced studies (DEA) in
distributed computing and systems at Joseph
Fourier University (Grenoble I) and ENSIMAG
Engineering School, France. She earned her
Ph.D. from the National Polytechnic
Institute of Grenoble (INPG), France, and
INRIA, the French National Research
Institute for digital science and
technology, in September 2000 with
very-honorable degree. Dr. Ismail has a vast
industrial and academic experience at Sun
Microsystems Research & Development Center,
worked on the design and implementation of
highly available distributed systems, and
participated in the deposit of a US patent.
She served in teaching at Grenoble I,
France, Assistant Professor at the American
University of Beirut, and has been serving
as an Adjunct Professor at the Digital
Ecosystems and Business Intelligence
Institute Curtin University, Australia. Her
current research interests include
Distributed Computing and Systems,
Artificial Intelligence (AI), Mobile Agents,
Healthcare, Energy Efficiency and Green
Computing, Smart Cities, and Pandemic
Control using AI.
She won the IBM Shared University Research (SURA) and the IBM Faculty Awards, very competitive world-wide, and won funding for major projects.
Dr. Ismail has international collaborations and publishing her research results in prestigious journals and international conferences. Her most recent work on energy savings in cloud computing data centers is published in ACM Computing Surveys and Internet of Things journals and on tailoring machine learning models for a better precision of Covid-19 infection prediction is published in Computational and Structural Biotechnology journal. She is currently member of ACM and IEEE. She served as an Associate Editor of the International Journal of Parallel, Emergent and Distributed Systems for several years, served as chair, co-chair and track chair for many IEEE and ACM international conferences, including being a General Chair for IEEE DEST 2009 and a General Chair, Technical Program Chair and Organizing Committee Chair for the 11th International Conference on Innovations in Information Technology 2015 (IIT’15) for which Dr. Ismail got the support of the IEEE Computer Society (HQs) technical sponsorship. Dr. Ismail is the Editor of the Information Innovation Technology in Smart Cities, published by Nature Springer @2018.
Assoc. Prof. Raju Shrestha
Oslo Metropolitan University, Norway
Shrestha is a computer scientist and
engineer, currently working as an Associate
Professor in Computer Science at the OsloMet
- Oslo Metropolitan University, Norway. He
holds a Ph.D. degree in Computer Science
from the University of Oslo, Norway, and
received a European Erasmus Mundus M.Sc.
degree in Color in Informatics and MEdia
Technology from the three European
Universities: University of Jean Monnet,
France, University of Granada, Spain, and
NTNU in Gjøvik, Norway. Dr. Shrestha has
several years of professional and industrial
working experience in the field of
Information Technology. He received
professional training in Japan, Singapore,
and the USA. Dr. Shrestha is a member of
IEEE, ACM, and IS&T. His area of research
interest includes AI (machine learning &
deep learning), data science, assistive
technology, image processing & analysis, and
Assoc. Prof. Su-Cheng Haw
Multimedia University, Malaysia
Su-Cheng Haw is
Associate Professor at Faculty of Computing
and Informatics, Multimedia University,
where she leads several funded research. She
is also the research center chairperson of
Center for Web Engineering (CWE), which is
multidisciplinary and encompasses diversify
research from modeling and tools,
implementation, testing and evaluation, and
application in the areas such as Databases
and Information Retrieval, Service Oriented
Computing, IoT, e-Learning, and Statistical
Her research interests include XML databases, data warehousing, semantic web & ontology, data modeling, and recommender system. She has published around 120 articles in reputable journals and conferences. She serves in several editorial boards and participated as technical committee member and reviewer boards for several international conferences and journals. Besides, she also received several ad-hoc invitations to review journal/conference articles. She is currently a member of IEEE and MBOT professional membership.
Speech Title: Retailer-based Recommender Systems: Challenges and Opportunities
Abstract: Recommender system has been a useful tool especially in the area of Big Data. There is a need for people to filter the information to search for their needs efficiently and relevantly. E-commerce website such as Amazon and Netflix have been using recommender system to build and boost their sales through the personalization recommendation. In addition, it is crucial that a Recommender System is able to provide an unified approach to visualize the interactions level of data, i.e., filter out the unnecessary, and provides the most relevant processed data for decision making purpose. In this talk, some existing recommender system will be discussed focusing in retailer and e-commerce domain. In addition, some possible research direction will be discussed based on the current trends and problems.
Assoc. Prof. Ehsan Atoofian
Lakehead University, Canada
is an associate professor at the Electrical
Engineering Department, Lakehead University,
Thunder Bay, Canada. His research interests
include computer architecture with emphasis
on many-core processors, parallel
programming models, and power-aware
micro-architectures. He received his B.Sc.
and M.Sc. in computer engineering from
University of Tehran, Tehran, Iran, in 2000
and 2003, respectively. He received his
Ph.D. from University of Victoria, Victoria,
Canada in 2008.
Speech Title: Reducing Energy in GPGPUs
Abstract: GPGPUs are throughput-oriented devices and are able to execute thousands of threads simultaneously. Compared to multi-threaded CPUs that run a few threads, GPGPUs rely on massive thread-level parallelism (TLP) to improve computational throughput. To support execution of a large number of threads, the architecture of GPGPUs differs from CPUs in three different ways. First, GPGPUs exploit a large number of ALUs to execute instructions of multiple threads, simultaneously. As an example, Tesla P100 employs 128 ALUs whereas the Intel Core i7 processor has only 4 ALUs. The other factor that enables GPGPUs running massive number of threads is a large register file. GPGPUs store the state of threads in the register file, enabling fast context switching. This contrasts with CPUs where the memory holds state of threads, resulting in 10s to 100s of CPU clock cycles required for context switching. As an example, NVIDIA Maxwell GPU has 64K 32-bit registers per streaming multiprocessor (SM) which is
much larger than the 336 physical registers per core on the Intel Haswell CPU. In addition to large register file, designers embed large L2 cache in GPGPUs to service memory requests of an increasing number of SMs. Furthermore, in the era of big data, it is expected that the criticality of L2 cache increases as it should service a growing number of memory requests which is the result of increasing data size. To address this challenge, GPGPU designers increased the size of L2 cache in each generation. As an example, Tesla P100 employs 4096KB L2 cache which is 25% larger than L2 cache in Tesla M40.
While a large number of ALUs and registers as well as larger caches increase performance of GPGPUs, they also increase power consumption. ALU, register file, and L2 cache consume 36%, 15%, and 5% of total power, respectively. Simply scaling up resources will not address performance challenge in future GPGPUs due to power constrains. In this talk, we discuss architectural techniques to enhance energyefficiency of GPGPUs. In particular, we offer new techniques to reduce energy of ALU, register file, and L2 cache.