Washington University in St. Louis will adopt a Hybrid Cloud strategy that is Cloud First for software and Cloud Preferred for IT platforms and infrastructure.
In line with Washington University’s IT’s “Buy before build” guiding principle, significant cloud adoption efforts will take place over 5 years (2018 thru 2023) to replace existing University applications with cloud-based software and to migrate IT platforms and infrastructure to the cloud.
On premise infrastructure will be gradually reduced but not fully eliminated as some business needs will continue to be better served by on premise technology.
To meet the challenges associated with cloud migration activities, critical staff skills will be developed or acquired to fill new key roles associated with modernized, cloud-powered University community.
The benefits of adopting cloud computing services include streamlined collaboration with peer Universities and strategic partners, access to best-of-breed software that is aligned with industry trends, enhanced resiliency of IT services, and a reduced need for on campus data center space.
Implementation Guidelines
When a business need is presented, WashU will seek IT solutions as high up in the technology stack as possible (SaaS first, PaaS second, IaaS, third)
Key Roles will be filled through a combination of resource reallocation, targeted staff training, professional services engagements, and strategic hiring
WashU will appropriately migrate IT workloads to the public cloud in order to avoid expansion of on campus data centers
A unified cloud governance model will be developed and integrated into the existing IT procurement policy and investment request processes
A unified cloud support model will be developed and integrated into the existing WashU IT organization
For PaaS and IaaS migrations, Microsoft-based workloads will generally move to Azure, Linux/Unix-based workloads will generally move to Amazon Web Services, and some research-centric workloads will move to Google Cloud Platform
WashU will leverage all readily available information security tools and best practices across the range of cloud computing services
Forces and Trends
Information Technology:
Drive toward continuous and rapid innovation
Latest software and tools are frequently only available in the cloud
Higher Education:
Increased need for interdisciplinary and inter-organizational collaboration
Large scale, short-term research computing needs
Washington University in St. Louis:
Need for modernization of enterprise IT systems
Desire to leverage infrastructure investments through the end of their useful life
Drive to enhance IT services while maintaining current levels of IT spending
Adjust to a more opex-focused cost model for IT
Desired Business Outcomes
Avoid significant capital expenses associated with build-out of new data centers
Improve the ability for faculty and staff to focus on teaching, research, and patient care
Streamline core business functions through the modernization of enterprise applications
Leverage improved availability inherent to geo redundant cloud-based solutions
Power advanced research through rapidly scalable cloud compute/storage resources
Enhance both internal and external collaboration
Success Metrics
Reduce on premise datacenter footprint by > 30%
Reduce faculty/staff time spent on IT overhead by > 10%
Increase uptime for critical IT systems by > 25%
Key Roles
Business Analyst: Focuses on understanding how cloud applications can be integrated into institutional business processes
IT Liaison: Works to ensure that IT is included in the product or service decisions taking place within business units on campus
Project Manager: Serves as the liaison between users, technical staff, and vendors
Strategic Sourcing Manager: Experienced in the unique nature of cloud contracts and risk mitigation
Vendor Manager: Maintains an ongoing relationship between the consumer and the cloud provider
Application Administrator: Handles the configuration, management, and access control for SaaS
Cloud Architect: Designs solutions that integrate multiple cloud (IaaS, PaaS, SaaS) and virtualization platforms, including on-premises services and solutions and data sources
Emerging Technologies Analyst: Provides research insights into future cloud technology trends.
Integration Engineer: Implements and maintains cloud integrations
Data Custodian: Manages how institutional data is stored and secured in the cloud Information Security Analyst: Responsible for information security standards and requirements, third-party risk assessments, and mitigations plans
Definitions
Cloud Computing: The practice of storing data or running applications on servers that can be accessed through the Internet
Cloud First: The University strongly recommends using vendor-hosted technology, using on-site technology only if absolutely necessary
Cloud Preferred: The University recommends using vendor-hosted technology first, but may use on-site technology if there is an approved business reason to do so
Private Cloud: Cloud computing services where the supporting technology is owned and operated internally
Public Cloud: Cloud computing services where the supporting technology is owned and operated by an external partner (Microsoft Azure, Google Cloud Platform, etc.)
On-Premise: Cloud computing services where the supporting technology is owned and operated internally (Private Cloud)
Hybrid Cloud: When software, platforms, or services are delivered using a blend of public cloud and on-premise infrastructure
Software as a Service (SaaS): Web-based software that requires little or no on-site IT equipment (Office 365, Box, Canvas, etc.)
Platform as a Service (PaaS): Managed IT platforms designed to simplify and streamline various IT functions (managed databases, web tools, etc.)
Infrastructure as a Service (IaaS): Managed IT building blocks used as a foundation for IT services (servers, networks, raw data storage, etc.)
Approved Business Reason to Continue Using On-Premise Infrastructure
Integration with equipment that must be on site (lab instrumentation, etc.)
Need for extremely low latency (10ms or less) or extremely high throughput (10GB/sec or more)
Ability to comply with legal or regulatory requirements
Hardware appliances (no cloud-based replacement available)
Quote for cloud-based replacement is significantly higher than on premise hardware