Five Oracle for Research Customers Advancing Sustainability with Cloud Technologies

Oracle for Research is a global community that is working to address complex problems and drive meaningful change in the world. The program provides scientists, researchers, and university innovators with cloud technologies, participation in Oracle research user community, and access to Oracle’s technical support network. Through the program’s free cloud credits, users can leverage Oracle’s technology and infrastructure while keeping research-developed IP private and secure.

Today, we are spotlighting five Oracle for Research customers that are driving sustainability with Oracle Cloud technology.

1. University of Montana uses Oracle Cloud to help fight forest fires before they start

As forest fires continue to devastate the U.S., researchers at the University of Montana are taking action, harnessing Oracle Cloud HPC to better forecast weather conditions.

The Weather Research & Forecast (WRF) model is an advanced and ever-evolving numerical weather prediction model used for atmospheric research and forecasting around the world. Because WRF users greatly benefit from high-performance cloud computing to produce these forecasts, few organizations can leverage this valuable service at present.

The University of Montana's (UM) Autonomous Aerial Systems Office (AASO) is working to create an autonomously-running WRF model with Oracle Cloud Infrastructure (OCI). This will have multiple applications, first for the Arctic and then for western Montana wildland firefighters and managers. The latter will greatly benefit from higher resolution weather forecasts to aid in prioritizing incident response to better protect human lives and forest ecology.

In preliminary tests, OCI HPC ran the WRF model at high resolution successfully, reducing compute time by a significant margin in comparison to the other high powered computing resources available. OCI reduces the time it takes to complete a high-resolution weather forecast to such an extent that these projections can be completed and shared within the same timeframe for which they are valid.

2. University of Reading launches Fruitwatch.org to monitor UK climate change

Recent research suggests that plants in the UK are flowering too early to achieve healthy pollination. This discrepancy could have devastating consequences – with less pollen and nectar available, insects in the UK may go hungry, hurting their chances of a sustained population. A lack of pollination also means flowers will reproduce less than in recent decades.

To help study this phenomenon further, Oracle for Research teamed up with University of Reading researchers to create FruitWatch.org. Hosted on Oracle Cloud and compatible with both desktop and mobile devices, this website is an APEX autonomous database application that enables UK citizens to report when and how their fruit trees are flowering.

The team at Reading is comparing these flowering dates to local pollinators records to help them understand whether fruit tree flowering and bee flight dates are in sync across the country, enabling a new, UK-wide assessment of pollination trends. We’re pleased to report that the website has captured over 4,500 records since February 21st, with every country in the UK taking part.

3. The World Bee Project, University of Reading, BeeHero and Oracle for Research enable sustainable crop pollination

Bees play a vital role in pollinating many food crops and The World Bee Project has teamed up with University of Reading, BeeHero and Oracle for Research to optimize the pollination of fruit crops while also ensuring a healthy and sustainable environment for bees and other insects.

The Oracle Cloud provides a wide range of analytics and data science capabilities, enabling collaborative research across these teams. The three-year project aims to deliver a better understanding of the delicate balance between farming practices, pollinators and the local environment, as well as new tools and guidelines to support this.

The project benefits from the latest in-depth academic research from Reading University, combined with the cutting edge sensor technology provided by BeeHero, who provide pollination services to commercial farms. The project requires both traditional on-site survey data, which is collected manually across the UK, as well as large volumes of IoT sensor data, which is automatically captured every 15 minutes. The sensor data includes hive temperature, humidity and in-hive acoustics, as well as data related to more than half a million trips the bees make every day. The project also collects hourly third-party weather data as well as satellite images.

After the data has been processed by World Bee Project partner BeeHero, it’s transferred into an autonomous database in the Oracle Cloud, which uses analytics tools including AI and data visualization. This provides researchers at the University of Reading with new insights into the bees and their journeys from their hives to the different crops they pollinate. It is hoped this collaborative research project will help the global agricultural community to better understand pollinators and the vital role they play in pollination.

4. Carnegie Mellon University helps autonomous vehicles go the distance

In recent years, the likes of Uber and Lyft have seemed well positioned to provide an electrical, autonomous vehicle service in the near future. Research from Carnegie Mellon University (CMU)  identifies two crucial disadvantages associated with vehicle automation and electrification, however. The first is the aerodynamic sensor drag introduced by the autonomous vehicle sensor stack and the second is the additional energy used by these vehicles’ computing. It's now clear these factors must be addressed before electrical vehicles can reach price parity with internal combustion engine vehicles.

CMU is therefore working to minimize this aerodynamic drag by developing a framework that enables the optimal placement of sensors on autonomous vehicles. The team is using Oracle Cloud bare metal A100 GPUs to run a series of airflow simulations to train a machine learning model that can describe the spatiotemporal dynamics associated with fluid flow. With this new machine learning technique, the CMU team hope to help enable the efficient, long range autonomous vehicles of the near future.

Researchers at Royal Holloway, University of London, have an interesting idea on how to help drastically reduce atmospheric carbon emissions. They posit harnessing carbon capture sequestration, a process whereby carbon, under pressure, is injected 3,000 feet or deeper into porous rock formation. Here, in its final resting place, the carbon is rendered out of commission – and can no longer contribute to climate change.

Executing, of course, is harder than theorizing. To effectively leverage carbon capture storage, scientists must have a reliable way to identify suitable underground “warehouse sites.” To find these, they need to look closely at geological reservoirs and assess how much carbon can be deposited there over time. 

Royal Holloway’s team developed open-source software to simulate carbon storage into the subsurface. They also leveraged 3-D microtomography images and Oracle Cloud to process massive data sets. The results? Scientists can now predict where to find suitable carbon storage sites more quickly, more accurately, and at a much lower cost than previously thought possible, arming us with a viable tool to combat climate change.

So there you have it – Oracle for Research is proud to be enabling research that drives sustainability around the globe. Are you and your team of researchers passionate about a specific area of environmental research? Find out how to accelerate your research on OCI and create the discoveries that drive sustainability.