Meet your new weather forecaster, the supercomputer
The science of weather forecasting has changed beyond recognition, in fact, today, thanks to supercomputers, it’s possible to make much more accurate predictions about the weather. But, how do these supercomputers make their predictions? First, it makes sense to explain what a supercomputer is. Put simply, it is a large array of smaller computers and processing equipment aggregated to make one large, smart, and very powerful computer. Frequently found in the science, engineering, and business sectors, they can reduce the time taken to solve problems to days rather than months and they’re really demonstrating their mettle when it comes to weather predictions and climate modelling. However, for supercomputers to make weather predictions, they need to obtain data from somewhere and this is proving to be a struggle on our continent. In Africa, according to The Washington Post, we have just one-eighth the minimum density of weather stations recommended by the World Meteorological Organisation, which equates to a problematic lack of data about dozens of countries that are the most vulnerable to climate change. This lack of data has meant imprecise forecasts and poor early-warning systems for people experiencing deadly cyclones, lengthy droughts and powerful floods. Researchers say the lack of data has also led to challenges in measuring the extent of climate change, making it difficult to prove global warming’s impact on the continent.
Fear of job losses is impeding South Africa’s digital transformation
The South African state is, by some distance, the country’s largest single employer. It employs more than 1.3-million people including teachers, nurses, doctors, municipal workers, police officers and defence force members. All too often, the public sector has also been used to cushion the country’s shockingly high unemployment rate. Those employment numbers come with a significant cost too. In the 2021-2022 financial year, the public sector wage bill hit R665.1-billion and is set to rise to R702 billion in 2024-25. Put differently, while the government employs 13% of all workers in the country, it pays 33% of all wages. Of course, ordinary taxpayers wouldn’t mind footing that bill if it meant improved service delivery. But it simply doesn’t. Even where public sector employees are doing their utmost, they are constrained by outdated working models, inefficient IT systems, and the slow rollout of digitalisation. While there are any number of factors at play when it comes to these constraints, one of the most alarming is a pervasive fear that digitalisation could lead to job losses. But with the right skills in place, that needn’t be the case.
A new breed of highly integrated devices for negative output DC-DC converters
Electronic equipment uses predominantly positive voltage rails for power. Occasionally, some negative voltage rails are also used. For this reason, negative (or inverting) output DCDC converter solutions are not as common as their positive output DC-DC counterparts. Nevertheless, when powering high performance devices in factory automation, building automation and communications systems such as high-speed DACs, Op-Amps, RF power amplifiers, AFEs, GaN FET gate drivers, IGBT gate drivers, etc., a negative voltage rail is needed.
Designers face a big challenge looking for a negative voltage solution where most legacy devices require external level shifter circuits with which to communicate. They are also outdated, inefficient, complex and bulky. The article discusses in detail the drawbacks of legacy solutions, and then investigates a new breed of highly integrated devices that addresses the deficiency and offers a compact, easy-to-use, and highly efficient negative output DC-DC solution.
Demand for datacentre capacity on African continent underestimated
When Africa Data Centres, a division of the Cassava group, announced last year it plans to build ten hyperscale data centres in ten countries throughout Africa, including the North African countries of Morocco, Tunisia and Egypt, questions were asked if that would not create an oversupply of capacity. EngineerIT asked the same question during a one-on-one interview with the newly appointed CEO, Tesh Durvasula. He said that in his nearly 30 years of working in the datacentre world the industry made a career of underestimating the demand for datacentre services. Gartner recently made a prediction that by 2030 Africa, would require a capacity of 1 gigawatt and between 500 to 700 data centres to meet the forecasted uptake of datacentre services. “We are continually underestimating the requirement, so no, I am not concerned that our expansion would create an oversupply in the market. The pressure is on building infrastructure whether it is in Nigeria, South Africa, or Morocco, the opportunities are there.
