Using Cloud Technologies in Geosciences
Within geoscience, the technical workflow is well established. There are clear ways to utilize the geoscientist’s knowledge and domain expertise to deliver subsurface evaluations. From time to time a new method may be incorporated, but overall the workflow remains relatively constant. Compare that to the rate of change of technology across all industries. What if we could harness some of these exciting new technologies and apply them to our workflows? With cloud technology, that possibility starts to become a reality.
Why do I care where my data is stored?
Do you know where every single seismic line that your company owns and licenses is located? Can you find that data and be looking at it in under a minute? If the answer is no, then cloud technology can help you with this.
By storing all your data in the cloud (including all your archived data), you can save both money and, more importantly, time. The cloud enables you to access any seismic line within seconds. This means that your time can be spent looking at the data rather than looking for the data.
OK, my data is stored in the cloud. Now what?
There are two clear paths that you can go down. Firstly, there is the ‘lift and shift’ option. You move your data and your workflows into the cloud and you keep doing what you have been doing. It may save you a bit of money and a bit of time. However, to do that is to ignore the key opportunities associated with the cloud and ignore the technologies that have been created because of cloud computing.
What is cloud computing?
Cloud computing provides global, on-demand services including compute power, storage and databases, all via the internet. Technology companies like Amazon, Microsoft and Google all provide cloud services platforms. It represents a paradigm shift in the way that functionality can be provided to your desktop. The scale of these cloud services platforms is mind-boggling. The total outlay on cloud infrastructure in 2018 exceeded US$80 billion (Canalys, 2019). This investment is necessary to support the increasing adoption of cloud computing by a diverse range of customers. For example, in 2016 one AWS (Amazon Web Services) customer, DigitalGlobe, moved 100 petabytes of data to the cloud. This level of adoption leads to economies of scale that can result in price reductions that all cloud customers, including geoscientists, can benefit from.
So how can the cloud help me?
There are numerous benefits to using the cloud.
From an IT perspective, the benefits include reducing investment in, maintenance of, and deployment time of IT resources, coupled with scalability to meet changing workload and user demands. In addition, you only pay for the resources that you use.
For a geoscientist, the key opportunities are the scalability and connectability of the cloud, as well as its near-limitless compute power.
Tell me more about the scalability of the cloud.
The cloud can scale up and down depending on what you are trying to do. That means that you can test an attribute on a single seismic line, but it also means that you can quickly scale up and run that same attribute over your entire seismic repository. Operations can run in parallel, meaning that even if you have 100 times the amount of data it will not take 100 times as long to get the results. This has exciting potential for data screening and large-scale portfolio evaluations.
How can the near-limitless compute of the cloud help me?
What would you do with your data if you had unlimited compute power? Seismic data is often large and unwieldy. It requires large amounts of compute to do things with the data, and this, in the past, has been a limiting factor. Project durations are often dominated by critical path computation. The current geoscience workflows have been developed around this limitation; however, once your data is in the cloud, that limitation no longer exists. To not think about what you would do differently risks missing the opportunity that the cloud provides. It also means that any lift and shift strategy is only a short-term solution because your current workflows and applications were developed before the cloud.
So, take a moment to think about “What would you do with near-limitless compute?”
What about: standardized processing over your entire portfolio to allow for direct comparisons between seismic volumes? Reprocessing older data using the latest technologies? Compressing your geoscience workflow timeframe? Or you could run machine learning algorithms to identify new opportunities; or augment the knowledge of geoscientists with techniques to remove human bias?
The near-limitless compute of the cloud opens up new possibilities and allows you to ask new questions about your data in a way that was previously not possible.
What do you mean by the connectability of the cloud?
The connectability of the cloud provides you with the option to securely share data with third parties. This allows you to improve your collaboration with both joint venture and academic partnerships to share understandings and develop new insights into your data. The connectability of the cloud can also be utilized internally to enhance the geoscience workflow. Data can be shared globally and viewed within seconds to deliver crowd-sourced interpretations and multiple perspectives on the same data. This helps open your eyes to the diverse range of potential concepts, resulting in a better understanding of the full potential – and also the challenges – associated with a particular asset.
Once your data is in the cloud, you also have the potential to rapidly connect to the latest innovations across multiple industries, which will help to deliver a unique competitive advantage for your business. Data can be transformed ready for use by data scientists and also by deep learning algorithms. You can combine your knowledge and domain expertise with the latest technologies to redefine the existing geoscience workflows, gain new insights and identify new opportunities.
This sounds like a lot of effort.
It isn’t the easy route and there are challenges. The hardest thing is often just getting started. It is an adventure into the unknown and that is what makes it exciting. Geoscientists are known for exploring, being curious and finding meaning amongst the uncertainty. In the end, you should be able to ask more and different questions of your data to gain new insights.
Seismic data is complex and uncertain. The geoscientists are the experts and therefore play a vital role in shaping what geoscience in the cloud could and should look like.
Will this take a long time?
It is actually very quick and easy to get started with the cloud. It is now possible to go from initially opening a cloud account to securely storing, viewing and collaborating on your data in less than one hour.
Geoscience data in the cloud can be integrated with existing geoscience applications to ensure business as usual is maintained as you explore and develop your workflows in the cloud. The technologies and scale of the cloud deliver performance whilst using industry standard data formats such as SEG-Y data. This reduces the complexity of data management by removing the need to create copies of the datasets in multiple data formats for different applications.
OK, I’m interested to know more, where should I start?
Our experience has been that a layered approach to development has helped people to get started.
Start by putting some data into the cloud and then you can learn as you go. Cloud services are on-demand services which means that you only pay for them when you use them, allowing you to explore cloud technologies with no upfront costs.
Explore, learn and investigate to see what the cloud can do for you.