National Integrated Cyber Infrastructure System; SA Research Infrastructure Roadmap

Meeting report

As the Chairperson had Covid-19, Ms D Mahlatsi (ANC) was nominated as Acting Chairperson.

National Integrated Cyber Infrastructure System; SA Research Infrastructure Roadmap
Prof Yonah Seleti, DSI Acting Deputy Director-General: Research and Development Support, led the DSI delegation.

Dr Daniel Adams, DSI Chief Director: Basic Sciences and Infrastructure, commenced the presentation with the definition of Research Infrastructure (RI), which are facilities, resources and services used by research communities to conduct research and foster innovation in their fields. They include: major scientific equipment or instruments, knowledge-based resources such as collections, archives and scientific data, e-infrastructure such as data and computing systems and communication networks and any other tools essential to achieve excellence in research and innovation. The presentation noted the Research Infrastructure Ecosystem; the new approach to RI provision particularly collaborative national platforms and the work done under the SARIR. It also touched on the South African Centre for Digital Language Resources, which aims to develop the eleven official languages to be capable of expressing all forms of knowledge, and drive their use and function in research and development, education, social transformation, trade, economic and scientific development.

Dr Happy Sithole, CSIR Executive Manager : National Integrated Cyber Infrastructure System and in charge of the Centre for High Performance Computing, gave a briefing on the National Integrated Cyber Infrastructure System (NICIS) and dealt with the implementation of the Big Data Strategy for research, development and innovation; converging technologies; highlights of high performance computing (HPC) systems and the South African National Research Network (SANReN); human capital initiatives; National e-Science Postgraduate Teaching and Training Platform achievements; student cluster competition; cyber security challenge; Datathon challenge and cyber infrastructure roadmap.

The achievements of the National e-Science Postgraduate Teaching and Training Platform, to date are: 124 students accepted onto the platform with 27 graduations, five at UniVen. University of Limpopo (UL) student graduations through Wits due to initial programme registration challenges at UL (now resolved). Due to Covid-19 challenges, 9 graduations were expected in September and 21 in December 2021. For the 2018, 2019 and 2020 years, nine students absconded. Many MSc graduates have been employed as junior lecturers, data scientists, data & business analysts, data engineers at well-known organisations. MA graduates have been employed as research interns, data analysts, data content creators, development intervention analysts, and GIS and environmental practitioners.

Discussion
Ms C King (DA) asked how effective is our infrastructure to guard against cyber intrusions. We have seen the recent cyber attack at Transnet and we are still unaware of the cause or the full extent of that intrusion and the damage caused. How do we monitor and perform oversight on research initiatives? Is the investment in research infrastructure and cybersecurity adequate for the country's development trajectory? Lastly, she asked about South Africa's involvement in the cyber infrastructure framework on the continent.

Ms J Mananiso (ANC) said at times people are left behind in knowing what is happening in this particular space. She was pleased there was an improvement. She asked about the emphasis placed on localising information; are there plans to ensure the District Development Model becomes one of the spaces that will be infiltrated to avoid giving a head start only to the science community. This can be a written response for when Members conduct oversight visits.

She referred to the achievement of 11 out of 13 targets – are there plans to achieve the remaining two? On the institutional review, what were the key findings and how are they being addressed? What progress can be reported on suitable strategies to enable DSI to evolve from research generation to research use and impact assessment? She asked for details on the National Research Foundation (NFR) internship and employment opportunities being transferred to the Human Sciences Research Council (HSRC).

Mr T Letsie (ANC) noted that infrastructure for two research sites has been approved for implementation – which ones are these and what are the implementation time frames? Has funding been secured and where will they be located?

On the 2016 additions to the road map, have they been reviewed; if so, what were the outcomes of the review? If not, when will the review be initiated? Which identified research infrastructure has not been established yet and what are the plans and time frames for those ones? Are the resources, framework and cooperation available to implement them? What are the challenges for the ones outlined in the presentation? How is oversight executed given the range of hosts and implementation partners over the RI being implemented? Is South Africa’s investment in RI and cyber infrastructure adequate for its development trajectory given the pace of technological developments?

Mr Letsie asked how many sites have been connected to SANReN to date. What progress has been made in assisting the Tertiary Education and Research Network of South Africa (TENET) in connecting deep rural sites and funded rural community projects? Has the national Big Data Strategy for research, development and innovation been published yet? If not, why not? What are the initial implementation actions and plans for the evolution uptake on high performance computing capability in South Africa? What progress has been made in implementing the fibre link between Carnavon and Beaufort West that will allow Meerkat measurement data to be carried over the new SANReN fibre backbone network from the Square Kilometre Array (SKA) core site to the Centre for High Performance Computing (CHPC) in Cape Town for processing. What are the current plans to increase the CHPC infrastructure? What level of funding will this require and has funding been secured yet? What progress has been made in deploying the 14 petabytes long term storage facility? Have the standards and policy recommendations to regulate the open and ethical use of management and research data been developed yet?

In the effort to build data and ICT infrastructure, has DSI taken the opportunity available now to help emerging black, women and people with disability companies? Generally, there is an over-reliance on American ICT companies. Is there a plan to build our own industry to close this gap?

The Acting Chairperson said in her view the investments made in the country were quite significant and were placing South Africa as a global player to address our local needs. However, what does it mean for the current socio-economic conditions in the country? How does an ordinary South African benefit from these investments? What is the immediate impact from these investments? DSI has been struggling with funding but will funding availability improve the work in the sector? Can the DSI indicate the difference between the big data and normal data? [Ms Mahlatsi lost network connectivity and did not finish asking her questions]

Responses
Dr Sithole, CSIR, replied that it was true one had to be worried about cybersecurity. DSI is putting in significant infrastructure where consolidated data of the country is being transported, processed and stored; hence cybersecurity is critical. We have been operating the cyber infrastructure since 2006. During these 14 years there has been many attempts to get into the cyber infrastructure and it is monitored 24 hours a day, every day of the year. There are tools that survey the infrastructure. For example, hackers are interested in high performance systems as once hackers take over the HP system, besides the data sitting there, you will have very powerful processing capability to do harm in the world. There are thousands of attempts per day and we document them and see from where they originate but there has never been a successful penetration. We look at this on continuous basis because these threats are always evolving.

Our South African Research Network supports all institutions in the country such as universities and now TVET colleges will be added. We also connect most of the research councils. We have a cybersecurity incident response team and its work is to continuously look at vulnerabilities – we assess the profile of the data that is moving across and when there is an unexplained increase in volume to a specific site, attention is directed to it to diagnose the cause. Once that is done, we advise the institutions about what is happening. We also have research programmes looking at cybersecurity such as the newly established E-Research Programmes to develop some technologies locally. There are partnership projects between Tshwane University of Technology (TUT) and CSIR looking at ensuring we use AI to fast track identification of threats.

The presentation outlined the Cybersecurity Challenge and through it we are training personnel across the country in partnership with universities. This programme is also ensuring that there is a workforce in the future. Lastly, we also support some of the beneficiaries within Research, Development and Innovation. For example, when the City of Johannesburg was attacked, a team was made available to assist it.

Dr Adams replied about oversight of research infrastructure. There are three different layers of oversight: from DSI, from the institutions where the infrastructure is located and from steering committees with representation from a broad range of stakeholders. At the lowest level is the advisory structures, which include people from different sectors to ensure that the strategic intent is preserved and to provide oversight on the impact towards the national imperatives and priorities of the country. All these structures and stakeholders have a mandatory role to play at different levels.

On the equity of the cyber infrastructure and SARIR infrastructure – there will never be an adequate system for what is needed now because this is a very dynamic and evolving sector. You have to draw circles in the sand at any point in time to determine what needs attention and what is required to drive research agendas and the priorities of the country. In the current space, we have provided adequate provision for research infrastructure as well as cyber infrastructure but it is an evolving field and we have to refresh, upgrade and replace all the time. The infrastructure at the current moment is sufficient to respond to the requirements within the scientific community.

On the SADC country cyber infrastructure systems, DSI is very involved in the region and the continent. Training and building capacity in the region is stable and a cyber infrastructure strategy was developed with all partners. We have also partnered with private companies to deploy some HPC systems to SKA countries for high performance computing as well as data management, storage and skills.

We do not want to blame Covid-19 restrictions for the challenges that arose – in the research space people had to do field work and the restrictions impacted the mobility of researchers but critical labs were open and skeleton staff was used. This has since been addressed to a large extent as the restrictions eased.

On the implementation of the outstanding RI, the rollout of RI as per the original schedule was mainly constrained by the readiness of the final proposal by the RI. It was a two-stage process which was getting the RI on the roadmap through a short proposal and ticking those boxes in the criteria map. RIs on the roadmap did not guarantee automatic funding and implementation and the next step was to develop a comprehensive research proposal on funding, business model and governance structures. This was subjected to a steering evaluation panel. These are some of the constraints because it requires very broad based consultation across the country to ensure that it carries the inputs of everybody interested in research infrastructure. The RI around solar technology is a complex field and we have put them at the end of the schedule as it would require much interaction and background work. We did finalise the neon and micro manufacturing RI, which has to do with developing technologies and devices in electronics and bio-sensors.

We have completed a process for Exco approval for the first review of the seven RI established in 2016. Each of these had their own start up challenges and we do not want to conduct a full review of the seven RI yet. From 1 April 2022, an evaluation of the first seven RI will take place. Thereafter in five years’ time there will be a proper review of the infrastructure roadmap, which will include a review of each infrastructure that may have to be terminated for various reasons.

We have indicators of the success of the SANReN in terms of number of sites connected. The constant refresh, replace and upgrade – we decided to focus on total bandwidth capacity available in the institutions across the country. We make available more than 5 000 gigabytes per second total bandwidth and the institutions draw from that. The important thing is ensuring that all universities, especially those in remote areas, are connected and participate in national and international collaborations. There are more than 600 sites connected – but that is not a true reflection but rather the total bandwidth.

The Minister has approved the Big Data strategy and it was now in a process of cosmetics. It was already available to the public.

Funding would definitely improve the work in the sector – there is always more that we desire to do but we try to stretch the available funding as much as possible. Funding remains a challenge.

Big Data is data that is too large or complex to be handled by the traditional tools or equipment we use. High level computers are required to do the processing of this kind of data.

Prof Seleti noted that Deputy Director-General Mr Imraan Patel did not prepare for participation and could not respond to a question about the Presidential Youth Employment Programme and requested that DSI submit the response in writing.

Dr Sithole said that besides infrastructure they were also looking at programs for climate modelling and the SADC region climate change approach. At the moment SADC countries use the forecasting models coming from various countries elsewhere. In South Africa we use a unified model coming from the United Kingdom. The grand plan is to develop a regional climate change model and work was already commencing.

Mr Letsie took over as Acting Chair and requested written responses for the unanswered questions. He bid a speedy recovery to the Chairperson who had tested positive for Covid-19.

The Committee adopted its Committee Report on Commission for Gender Equality as well as committee minutes dated 24 and 26 November; 1 and 3 December 2021. Meeting adjourned.