Miss Olayide Adebimpe Arodola, a UKZN masters student in Pharmaceutical Chemistry, was selected in a lucky draw, the winner of the CHPC User Satisfaction Survey. She has won a tribal chess set and a travelling laptop bag.
Her publication and MSc dissertation are as follows:
Her dissertation is not yet published.
Last Updated on Thursday, 30 October 2014 10:13
Bioinformatics Service Platform Sub-committees: Education and Training, Services and Support
The Bioinfomatics Service Platform would like to call for nominations for members of the following sub-committees: Education and Training, Services and Support.
The Department of Science and Technology (DST) has invested in a national Bioinformatics Service Platform (BSP) with the aim of providing a range of user-centred and well-coordinated computational bioinformatics services to the South African biotechnology industry and life science researchers. The BSP service shall encompass access to bioinformatics expertise, computational infrastructure, human capital development activities and research partnership opportunities. DST has appointed a BSP steering committee, based on a nomination process, to oversee the high level strategy or business plan of the BSP.
The sub-committees will focus on advocacy, consultation and outreach projects, as well as propose plans to address the high-level strategy areas proposed by the steering committee. The Services and Support sub-committee will review requests for hosting BSP staff and the Education and Training sub-committee will review requests for specialised courses and trainings. These committees will meet regularly to develop the strategic direction of the platform under these two activities. The BSP manager will be involved in these meetings and will be responsible for implementing these strategies. The chairpersons for these sub-committees will be appointed from the steering committee. The members of each of the sub-committees will be five, including the chairperson.
Education and Training committee will provide:
Services and Support Committee will provide:
Guidance on service models,
Guidance on support needs; and
Advice on infrastructure development (equipment).
The candidate must exhibit a firm understanding of the needs of bioinformatics services and/or training to non-bioinformaticists and clearly demonstrate at least five years of experience in one or more of the following criteria:
The committees’ composition will take into consideration ethnicity, gender and geographical representation.
The closing date for nominations of all above committees is 31October 2014.
Last Updated on Friday, 10 October 2014 11:26
South Africa has successfully defended its championship of the International Student Cluster Competition, winning the overall prize in Leipzig, Germany.
The announcement made on 25 June 2014, shocked many attendees of the International Supercomputing Conference since it is the first time a team wins the competition two years in a row. The South African Team of six was also the youngest as it was made up of undergraduate students only.
The High Performance Computing (HPC) Advisory Council Student Cluster Challenge is an opportunity to showcase the world's brightest computer science students' expertise in a friendly, yet spirited competition. Team preparations prior to the competition includes working with supervisors and vendor partners to design and build a winning cluster from commercially available components without exceeding the set power limitations, and to learn the predetermined HPC applications. The Dell computing company was the vendor partner behind the South African Team, proving all equipment. Dell also sponsored a trip to its headquarters in Austin, Texas and to the Texas Advanced Computing Center where the team received a tour of the facilities and mentoring from technicians that manage the centres. NVIDIA sponsored Graphics Processing Unit (GPU) accelerators for the team and Mellanox sponsored the interconnect.
South African team members:
· Eugene de Beste - University of the Western Cape
· Nicole Thomas - University of the Western Cape
· Saeed Natha - University of the Western Cape
· Warren Jacobus - University of the Western Cape
· Pieter Malan - University of Stellenbosch
· Ellen Nxala - University of Fort Hare
David Macleod – CHPC
Vernon Nichols – DELL
Nicholas Thorne – CHPC
Head Node (Dell R320):
· Intel E5 v2 Processor
· 8GB RAM
· 400GB RAID1 Array
· 2.4TB RAID0 Array
· Infiniband FDR
8x Compute Nodes (Dell R720):
· 2x Intel E5-2660v2 Processors
· 64GB RAM
· NVIDIA 40m GPU
· Infiniband FDR
High Performance Computing Challenge
High Performance Conjugate Gradient (a surprise application)
Lowest power consumption to complete a task (Surprise Task)
The challenge also featured High Performance LINPACK (HPL) as a side competition. The HPL score does not count towards the overall competition. To win HPL teams have to build a cluster specifically for it by using more GPUs per node and fewer nodes. This however, has a negative impact for the main competition because the team would have fewer CPU cores.
The team that won HPL had two GPUs per node and only 4 nodes. This means they had enough power to use all their GPUs and some of their CPU cores to run LINPACK. Team SA also had 8 GPUs but double the nodes. This means the team only had enough power to run LINPACK on the GPUs. All the CPU cores sat idle.
The South African cluster was deemed better balanced than those of other teams.
Teams entered into the competition included:
· Ulsan National Institute of Science and Technology (UNIST), South Korea
· A combined team of students from Massachusetts Institute of Technology (MIT), Bentley University, Northeastern University (NEU), United States
· The University of Edinburgh (EPCC), United Kingdom
· Chemnitz University of Technology, Germany
· University of Hamburg, Germany
· University of São Paulo, Brazil
· University of Colorado at Boulder, United States
· University of Science and Technology of China (USTC), China
· Two additional university teams from China
CHPC begins its Winter School next week where the selection for the 2015 team will begin.
Last Updated on Monday, 30 June 2014 10:14
Tianhe-2, a supercomputer developed by China's National University of Defense Technology has retained its position as the world's No.1 system for the third consecutive time. Tianhe-2 has performance of 33.86 petaflop/s (quadrillions of calculations per second) on the Linpack benchmark, according to the 43rd edition of the twice-yearly TOP500 list of the world’s most powerful supercomputers.
Please click here for more information.
Last Updated on Monday, 23 June 2014 14:28