Speed was of the essence when NCI asked Octo and its partners to build a cloud-based image repository to support COVID-19 researchers.
It did not take federal agencies long to put their time, energy and dollars towards the effort against the novel coronavirus pandemic when it became a national emergency in early March.
The spending started to level off in late May but stood at almost $14 billion as of Monday, according to a dashboard from market intelligence and supply chain analytics firm 202 Group.
How did that fast response on the government’s part translate to industry? One example can be seen in a National Institutes of Health project that sees federal IT company Octo Consulting, pathology software provider Indica Labs, and bioscience firm Axle Informatics collaborate to stand up and run a cloud-based virtual image repository.
The repository has been in use for about seven weeks by researchers and other health care professionals in 75 countries since it went live. Given that theme of quick turns on requirements, I recently asked Octo’s Vice President of Enterprise Solutions James Farley the nature of how the team got to work on the effort.
“Our cloud architect that supports NCI (NIH’s National Cancer Institute), as soon as he got the call from his counterpart, he started working on wire frames, what we think the initial page needs to look like of how we are going to capture the slides and give the international community a place to load that imagery,” Farley said. “We also had to figure a way to ensure that the intake page was secure and it was human-readable, not a bot or a machine trying to make the image uploads.”
Now for the second aspect of my question: how soon did it all come together?
“Realistically, within the first day we had the wire frames sketched out,” Farley told me. “Within the first three days, we created the back end that’s all micro services-based and it’s all containerized that can be scaled depending on how many people are actually using the environment.”
Farley also explained that under the repository’s hood, it does not use a traditional “spinning” technique that makes things run almost all the time. Instead, the first page a user hits spins up those micro services and activates the system’s back end for the imagery to be loaded. Once the imagery gets into that repository, the system spins down resources the user needs again.
“It’s not constantly costing NCI or NIH by having a machine spun up and running all the time,” Farley said, adding the team is using the Amazon Web Services cloud infrastructure for the repository.
Images that users put in the repository are of tissues of organs such as lungs, liver, kidney and heart in patients infected with COVID-19 and other similar types of coronaviruses that cause respiratory illnesses. NCI and NIH want the images, annotations and metadata to be used as a reference set for research and future clinical trials to hopefully thwart COVID-19, for which there is no proven treatment or cure.
That novel and still uncertain nature of COVID-19, even as states and localities move in between phases of stay-at-home orders, necessitated the federal government’s all-hands-on-deck approach that more than one executive has relayed to us.
Add Farley to that group of leaders that have seen agency missions get rewired or repurposed to at least some degree for the fight against COVID-19. He also sees that mindset reflected in industry and not just for this project. Across the board, Farley sees companies teaming together and sharing ideas.
“Just focused really on the interactions that everybody is having, they’re all addressing the greater common good, they all know this is a problem that is not going to be solved in ones or twos or individually,” Farley said. “This is going to be a team effort and it’s going to take organizations, departments, administrations and companies doing things that they do collaboratively. For us locally: (it’s) doing things in common that are commonly done. We have to be able to address that together.”