1. Can I use grant funding to fund the movement of data from physical to cloud infrastructure?
2. Can I use grant funding to convert from physical to cloud infrastructure?
3. Can I use grant funding to support cloud compute and storage infrastructure?

There is a growing trend within universities and research entities towards utilizing cloud compute and storage infrastructure. There are numerous advantages to moving data into the cloud, particularly when it comes to the collaborative nature of research projects and the higher costs associated with physical infrastructure.

Many folks are struggling to understand what this paradigm shift in infrastructure means, especially for a research community that relies heavily on federal grants to support their efforts.

At first glance, the three questions posed at the beginning of this article appear to address the same underlying issue. In other words, the same question appears three different ways. However, there is some nuance in how these questions are phrased that has a dramatic impact on how the infrastructure can be financed through grant funds.

Let's start with Question 3 - Funding Cloud Infrastructure

It makes the most sense to start with question #3... Can grant funds be used to support cloud infrastructure? The simple answer is, absolutely! Grant funders typically remain neutral when it comes to distinguishing between alternative forms of technology to accomplish a particular objective.

If compute and storage infrastructure are justified in the context of a particular research project, then it can be supported through grant funds, assuming there are no funding restriction unique to that particular funder or grant program.

In many cases, federal funders are simply not allowed to display favoritism for one type of technology over another. The primary funders of research, the National Science Foundation (NSF) and the National Institutes of Health (NIH), are well aware of the growing trend towards cloud solutions. In the former’s case, they are in many ways leading the charge in studying and exploring the potential of cloud infrastructure to support research.

If the program is an instrumentation/ infrastructure grant, then the case must be made that inclusion of the cloud resources will support several science use cases (multiple researchers and projects).

If the grant program is targeted towards a particular research area, then the expenditures must be justified in the context of the singular project that is being proposed.

An example justification: A researcher is doing a study on a particular treatment for pediatric cancer patients. He justifies the use of cloud infrastructure as a way to ensure his collaborators across the country (that lack physical storage infrastructure) have access to several large datasets.

Final Answer: As long as the cloud infrastructure can be justified in the context of the research project, federal grant funds can be used to make it happen.

Question 2 - Conversion to the Cloud

While we have established that grant funds can be used to support cloud infrastructure, it really comes down to appropriate justification.

The issue is that most people approach funders with the idea of "converting" from physical to cloud infrastructure without understanding the potential misgivings. The NSF and NIH have invested billions of dollars in physical infrastructure over the past half-century in their roles as the primary government funders of research.

From their perspective, the idea of simply financing your organization's conversion to cloud infrastructure is laughable. At face value, there is no reasonable justification that will convince these organizations to simply fund a conversion from physical to cloud infrastructure.

Many note that cloud is now required for collaboration, however, funders counter that previously funded physical infrastructure should be used via a virtual private network (VPN), or some other mechanisms for collaborators to access the project data.

This notion doesn't necessarily stem from bias against cloud infrastructure (in fact, NSF is one of the primary parties responsible for its proliferation); it is merely what they see as a justified stance to protect their past investments in physical research infrastructure.

Final Answer: You will get nowhere if you simply ask grantmakers for funds to convert your organization from physical to cloud infrastructure.

Question 1 - Funding the Movement of Data from Physical to Cloud Storage

So, as we discovered, cloud infrastructure can be funded through grants, so long as you are not simply requesting that grantmakers do a wholesale replacement of your existing physical infrastructure with cloud technology.

Does that mean only new cloud infrastructure can be funded and replacing older physical storage can never be justified?

While there are no conceivable use cases that would convince grantmakers to fund full scale infrastructure conversion to the cloud, there are instances where replacement can be justified in the context of particular data relevant to a research project.

Imagine a scenario where the NIH funded physical infrastructure so that a cancer researcher could access and analyze datasets that were measured in Megabytes a decade ago. The use of imaging in oncology has dramatically increased the size of datasets in this particular field over that timeframe. The data is now measured in Terabytes and the project has grown to include researchers across the United States, Europe, and Asia. In order to deal with the ballooning size of the data and the collaborative nature of the research, a principal investigator could make a case to the NIH to move the data from their exsisting physical infrastructure into the cloud.

If a researcher strings enough of these use cases together, they may get to a point where they have converted completely from physical to cloud, albeit in a piecemeal fashion.

Final Answer: Scientific use cases specific to a particular research project will dictate whether grant funds can be used to move data into the cloud.

In addressing these three issues related to the funding of cloud infrastructure, we should now understand the beginning three questions as drastically different rather than repetitious.

While the differences may appear subtle at first, each approach will lead to drastically different responses from grant funders. The positive news is that the time has never been more ripe to use grant funding for cloud infrastructure... you just need to ask the right questions.