For Sharp HealthCare, cloud technology comes with autonomy
Sharp HealthCare's cloud strategy spans public, private and SaaS platforms to reduce vendor dependency and emphasizes knowledge transfer with its deployment partner.
Sharp HealthCare is working through a multi-phase, hybrid cloud deployment that has a bit of everything: AWS infrastructure, a private cloud, hundreds of applications and an electronic medical record system migration.
The project marks the San Diego-based healthcare system's first foray into public cloud infrastructure. The key motivations driving the healthcare cloud technology initiative include the potential for cost savings, the ability to automate IT resources and a faster path to technology adoption, said Thomas Gorrie, Sharp's director of core technologies.
"We were leveraging SaaS services, but we realized we really needed to take a look at commercial cloud [platforms]," he said. "We didn't have expertise in that at all."
Sharp hired Chicago-based IT service provider Ahead to guide its cloud project, which got underway about 18 months ago. In the first phase of that effort, Ahead reviewed Sharp's application portfolio, considered migration options and came up with an end-state model for what it would look like to run that portfolio in the cloud.
The project aims to deliver the technical nuts and bolts of a hybrid cloud. A parallel thrust, however, is to preserve Sharp's IT autonomy. The health system's cloud architecture, for example, is designed to prevent vendor lock-in, which hinders customers from moving workloads to alternative cloud providers. In addition, Sharp's relationship with Ahead emphasizes knowledge transfer to avoid dependence on the service provider. To that end, Ahead is coaching the health system's IT team on how to set up and manage a cloud environment.
"[Ahead] could have built it and operated it for us if we had wanted them to," Gorrie said. "But that's not the way we do things. We worked with their experts to essentially train us to feed ourselves."
Sharp HealthCare's hybrid cloud aims to provide cost, AI adoption and self-management advantages.
Building for multiple clouds and self-management
Sharp's cloud deployment is built to accommodate multiple cloud vendors. Ahead worked with Sharp to design a cloud network that uses Megaport's network-as-a-service offering to link the health system's public cloud provider, AWS, to its private cloud, which is based on Cisco Unified Computing System hardware and a VMware hypervisor. Megaport's service supports several cloud platforms, including AWS, Google Cloud and Microsoft Azure.
"We don't want to be dependent on one particular public cloud provider," Gorrie said.
With the cloud network in place, Ahead began the ongoing process of training Sharp's IT team to construct an AWS environment. One workshop covers best practices for an AWS landing zone, which provides a launch pad for deploying workloads and applications in the AWS cloud. Other workshop topics include continuous integration/continuous delivery (CI/CD) best practices and the fundamentals of both GitHub Actions and Hashicorp's Terraform.
GitHub Actions, a CI/CD platform, lets users create workflows for building, testing and deploying software. TerraForm automates cloud tasks such as infrastructure provisioning and orchestration.
The workshops include consultant-led briefings on each topic and practical scenarios that reinforce the concepts, Gorrie said. In the case of Terraform fundamentals, Ahead provided pre-packaged Terraform modules to help IT personnel quickly apply the newly learned concepts in Sharp's AWS environment, he added. Sharp is now extending its Terraform knowledge, techniques and tools to the private cloud, so operations can be unified across the hybrid setup, Gorrie said.
Another third-party provider, which Gorrie didn't identify, built and now operates an AWS environment for Sharp's data analysts and web teams. But once Sharp's cloud-trained personnel build a self-managed AWS environment, they will take on the provider's responsibilities. Gorrie said Sharp will be able to self-manage the cloud for about half the cost of contracting out the management duties. That transition is scheduled for completion by Feb. 1, 2025.
Healthcare clients seek cloud reskilling
Healthcare organizations moving to the cloud are looking within to facilitate adoption.
"The common pattern we often see with most healthcare systems going down the cloud modernization journey is leaning in heavy on the enablement and reskilling of existing IT operations staff," said Andy Sajous, field CTO for healthcare at Ahead.
Running a smooth cloud operation often involves a mindset shift for the entire IT department, he noted. Ahead ran into similar challenges at Sharp, but Sajous said strong communication from the healthcare system's leadership and top-down support helped make the transition successful.
Sorting apps, speeding time-to-value
In the coming months, Sharp will work with Ahead to rationalize its application portfolio. The organizations will evaluate the apps and determine whether a given system should be migrated to a public cloud, maintained on Sharp's private cloud, replaced with a SaaS application or retired, Gorrie said.
We don't want to be dependent on one particular public cloud provider.
Thomas GorrieDirector of core technologies, Sharp Healthcare
Sharp has about 400 applications. The healthcare system has already migrated many of the largest ones to SaaS or remote hosting providers. One example is Sharp's electronic medical records migration from a Cerner system to a remotely hosted Epic EMR deployment, which went live in March. Ahead played a role in that aspect of the project, setting up AWS to extract data from the Epic database, move it into an AWS account and then load it into a Snowflake data warehouse for analytics.
But there are plenty of applications the organizations have yet to sort out.
"We have a handful of pretty significant applications that are potential candidates for a commercial cloud," Gorrie said. "And then we have a whole bunch -- on the order of 200 applications -- that are pretty small. And that's where the difficult work is going to be. A lot of those applications are not amenable to moving into the commercial cloud."
Such applications can be migrated, but only if the infrastructure is duplicated in a private or public cloud, which Gorrie said becomes expensive.
As the rationalization process continues, Sharp will seek opportunities to optimize cloud consumption to trim expenses.
"Our overall goal is to try to get the commercial cloud at a cost-neutral position," Gorrie said. "And then, as we go forward, we hope to be able to show cost reductions."
Beyond cost cutting, another advantage of cloud is time to value, Gorrie said. Emerging technologies such as AI are more readily accessible through the cloud, he noted. In one case, Sharp deployed its own version of ChatGPT in AWS. The rollout took about 60 days, from concept to production.
"The only way we could deploy it was through cloud services," Gorrie said.
John Moore is a writer for TechTarget Editorial covering the CIO role, economic trends and the IT services industry.
Dig Deeper on IT applications, infrastructure and operations
