By Steve Sears, director, Cloud and Virtualization Services, Johns Hopkins Medicine
At Johns Hopkins University, IT delivers a set of computing infrastructure services to customers at our diverse institutions through the Enterprise Networking and Computing Technologies organization. In response to growing requirements, our newly organized Cloud and Virtualization Services group is expanding availability of server, storage, and client access solutions to distributed IT organizations and researchers.
Server virtualization and data storage services operate as a large-scale, enterprise-class computing resource. Supporting more than 1,500 virtual servers, comprising 70 percent of the server infrastructure, the organization is aiming to build the next generation of computing environments. Client-access solutions are rapidly evolving to provide virtual desktop solutions and virtualized application services to a larger segment of the user community. Having provided more than 6,000 virtual desktops to the Johns Hopkins clinical community, the organization has learned to scale and mature the systems management practices for these solutions. More importantly, building the support systems needed to create an excellent user experience in a virtual environment has provided the organization with the skills and confidence to build the next generation of user computing.
Today, resources are pooled on storage and virtualization platforms, a degree of elasticity is present, and services are more “on demand” than ever before. Behind the scenes, there is some level of automation present, with many of the tools and processes in place to build a self-service portal. In essence, we are prepared to deploy a cloud computing solution as a natural next step.
From a customer perspective, the most noticeable differences between today’s resource pool and a cloud computing approach are the way resources are created and accessed. Most services require IT consumers to navigate various tools, request methods, and people to get a server up and running.
These interactions are routine and not a major barrier for people inside the central IT organization. However, for a broader audience, getting a final product can be a significant challenge. Cloud computing reduces some of these problems by introducing tools and processes that provide a complete server and storage environment without the need to interact with technical specialists.
One focus of our Cloud and Virtualization Services team is determining the kinds of services useful to distributed IT groups, researchers, and departments. We have defined general services like virtual server hosting, storage, and virtual desktops. To help facilitate adoption, we also define services by how they are used: for example, departmental analytics, researcher desktops, lab and testing environments, training environments, and mobile computing.
We’ve built and identified costs for technical solutions around a common set of themes based on customer need for our infrastructure-as-a-service (IaaS) offerings. Our service catalog clearly labels and packages these choices with the appropriate options and price, making it easier for IT consumers to identify the right services for them. While many service providers indicate levels of service around terms such as “silver, gold, and platinum,” or “basic, advanced, and premium,” we have found it helpful to show what these terms encompass so that decisions and choices are more tangible.
Our catalog offers a variety of computing, storage, data protection, management, networking, and operating systems and services. While many of these services are simply included at no extra cost, others require additional resources and incremental fees. We also offer additional tools, such as load balancing, data replication, and systems management, which have raised the overall level of functionality available to many more systems.
"We’ve built and identified costs for technical solutions around a common set of themes based on customer need for our IaaS offerings."