A HPC (high performance computer) is a computer which uses the aggregate power of a several individual computers to create a high performing system in order to solve a complex problem. The individual computers in this cluster (group) are referred to as nodes.
I found another analogy, based on the phenomena of emergence as a way to further understand just how HPCs work. Think of the cells in your body. A single cell, on its own, isn’t very useful. It is greatly limited in its ability in terms of feeding/protecting itself in order to lengthen it’s life. When groups of cells come together, they form tissues. When tissues combine, they form organs. A group of organs make up the human body. This network of cells, tissues and organs enable you to do everyday things like walking, which would be too complex for a single cell or organ. Emergence, is basically simple structures coming together to form complex structures.
The increase in the number of ‘smart devices’ creates enormous and valuable data pools which can be used in various ways. (An article in the Economist predicts that a self driving car will generate up to 100GB of data per second). This has brought with it the need to have resources that can process/ refine this voluminous data from its abstract form to a more useful form. HPCs are the refineries.
If you think about the applications of HPC systems in the environment , the sky is the limit. In the healthcare sector for example, a smartphone app can be used to collect data on the population to determine the risk of asthma. The data collected is fed to a HPC in real-time which then analyses and visualizes the data to which parts of the population are at risk of asthma. You can use the same HPC to determine what factors make a certain part of the population more asthma prone by monitoring things like the air quality. Air quality sensors can be installed in smartphones & the phones transmit this data to a HPC which visualizes the data. The HPC can also be used to collect data on how citizens may be polluting the air and putting themselves at risk of asthma.
Analysis and visualization of these large volumes of data by HPC can help cities work towards solutions that reduce air pollution and make them more eco friendly to improve the quality of life for their residents.
Another application of HPC would be in studies and research towards climate resilience, this would go a long way in improving food security in Kenya. HPCs are able to aggregate data collected on changing weather and climate patterns to create models that can be used to accurately predict the weather. These predictions can then be used by government to better advise farmers on how to plan their harvest and planting times in order to mitigate food shortages.
HPCs are not limited to the world of academia, businesses that generate large volumes of data; like telecom companies can utilize the high computing power in the analysis of their data to create products that are better suited for their consumers as well as get feedback on existing products.
Government can utilize HPCs to quantify how large scale infrastructure developments like SGR affect communities and the environments close to them . A clear understanding of both negative & positive impact will help in creating policy that develops our country’s infrastructure, without hurting our environment and local communities.
Now, you’re probably thinking to yourself that HPC is a great resource, why don’t we have one in Kenya? And what would it take?
As a country, we need political goodwill and lot of backing from government as creating a HPC is very costly undertaking. Removing of tax on electronic items is one of the steps we can take in the journey to having our own HPC. Universities need to invest in centres for research which would utilize HPCs to generate relevant information to end users like government. Universities must also create an environment that fosters innovation as well as provide relevant curriculum to its students to enable them to build a super computer.