The Transformation of your College, University & Research Institution data center strategy can ultimately play a significant role in improving student success. Many factors contribute to the need to transform. These include rapid data growth, exchange of information within applications and multiple providers, and the need to maintain privacy standards, resiliency requirements, backup, synchronization, latency, and connectivity within the multi-cloud environment. Our experts will assist in developing a plan to address these issues and more.
What's driving the change
The challenges facing information technology needs in Higher Education are growing exponentially. Historically, facility support has largely led to multiple data centers serving Academic, Administrative, and other distributed Departmental and Research needs. The very nature of the academic mission led to this diversity of facilities, generally housed in buildings where competition for space to support that mission made expansion or remediation of data centers difficult and costly. Additionally, the distributed nature of older facilities keeps operating costs high and limits the adoption of new technologies.
Some of the many challenges and observations that Colleges and Universities are facing includes:
Facility budgets under pressure and have difficulty competing for the capital and operating expense budgets needed to support a growing, diverse demand.
Growing costs for third-party services to help manage and operate older data centers.
Existing data centers, those 10 to 20 years old, that are part of a potential consolidation plan, do not have a clear path to a solution that is institution-wide in consensus for change. Operating costs continue to be high due to older design and energy efficiency capabilities.
Out of date governance and operations procedures impede rapid deployment or do not enforce standards needed to safely operate all types of data centers.
Moving workloads to the Cloud or to Software as a Service is often done in an uneven fashion and do not take advantage of good decision analysis for their use. Many times, these decisions are done at the departmental or application level and do not take advantage of the potential for total institution scale. Expected cost savings may not be achieved or may not take into consideration the collateral effect on cost to the data centers that supported the migrated workloads.
“Ghost IT and Cloud” use (casual, unauthorized, or unknown use of cloud or on-premise use of space for IT assets) represents a cost and risk that can be difficult to capture but can be critical to understanding total costs.
The need for High-Performance Computing (HPC) has expanded beyond the traditional mathematics and scientific communities. The need to support “Big Data” is a growing part of every curriculum and research area. HPC demands higher density capabilities for power and cooling that most existing Higher Ed infrastructure cannot accommodate.
Alternative data center sourcing strategies need to be seriously considered. Often the use of Colocation facilities that might not be in close physical proximity to the institution are difficult to gain consensus for even though considerable cost savings may attach, and there isn’t a critical, confirmed, requirement for daily on-site access.
Ability to cope with special circumstances such as the Covid 19 lockdown:
Older on-premise data centers were not designed to be low touch or lights out and rely on facility and IT personnel for daily operations.
The requirement to support online delivery of courseware is challenging network, compute, and storage capabilities.
Increased need for media production facilities for distance learning.
Preventing “guard down” security attacks due to volume of network activity or launching of new services.
How can you meet these requirements?
Solutions to data center facility and associated challenges are complex but they can be faced through focused campus-wide analysis and short/long term master planning. It is common to react with tactical solutions that are not useful in the long term. A Data Center Strategy meets the objectives of addressing the challenges detailed above by developing solution scenarios that vary in cost, time, complexity, and risk to implement. It starts with a rapid assessment of your current environment and pivots to the needs of your future state IT, application, and cloud / SaaS architectures. It should also include an assessment of the network infrastructure inclusive of bandwidth requirements to match any future state changes.
The process itself helps to gain consensus through the inclusion of multiple communities of interest. It helps build a plan that is focused on coping with change rather than being a static solution. It addresses capital and operating costs as equal in importance to the technical solution. It develops total cost of ownership forecasts for each solution scenario to understand the impact of technology, implementation, and growth in scale of any plan.
What services are included?
Present state data center analysis including IT and network architectures, application portfolio and opex/capex budgets.
Evaluation of the technical infrastructures of each legacy data center to understand capacity, resiliency, remediation, and growth potential. Cost and associated risks are used as measures for determining if any legacy facilities are good candidates for modernization. This might include any off-premise data center including the use of exiting colocation deployments and cloud / SaaS environments.
Workshops with the academic. administrative and research communities to understand:
Future state planning.
Key data center technical requirements.
Application portfolio needs and growth.
Security and compliance issues.
Financial guidelines for developing solution scenario opex and capex budgets.
Development of the future state through:
Modeling of IT requirements.
Developing the right mix of on-premise and colocation data center space, use of cloud and SaaS.
Creating alternate strategies with associated costs.
Analyzing best-phased implementation timelines and associated risks.
Develop a set of presentations to assist in the dissemination of results and report outs to gain both technical, financial, and senior-level approval.
University of Iowa
Iowa City, IA
LEED Platinum Data Center Design
One of the University’s old data centers, serving the core Information Technology Services (ITS) group, was outgrowing its space in a 100-year-old building, which was far from ideal for an optimized IT environment. The Health Care Information Services (HCIS) group,
serving the university’s hospitals and clinics, was also outgrowing its home. EYP MCF helped University of Iowa providing the design and commissioning of a Tier III Data Center Facility, this Information Technology Facility became the university's first LEED Platinum building. and is believed to be the first educational institution
data center with that certification.
Confidential University and Research Institution
Data Center Multi-Cloud Strategy
1.5MW Data Hall with HPC
The client was looking to perform a comprehensive Data Center and Multi-Cloud Strategy that included the capacity of the existing facilities, a cloud adoption analysis to quantify the IT infrastructure cost drivers that move to the public cloud, a future state planning and a co-location RFI process and Load requirements. EYP MCF evaluated the condition and capacity of four main data centers and up to 15 server rooms. The project includes an application Cloud Adoption Service, Co-Location selection and a data center hall MEP design of a new High-Performance Data Center. This 1.5 MW data hall will house an HPC computer that will put this university in the top 15 research institutions in the US.
Space, Power, Cooling TCO Strategy
Tier III 80,000 SF
EYP MCF provided a space, power and cooling strategy that included the cost estimating, design, and construction administration of a Tier 3, 80,000 SF data center as part of Indiana University's Cyber Infrastructure Building.
The research portion will be designed so it is expandable to accommodate petascale computers at anticipated load densities of over 300 W/sq. ft.
University of Illinois System
Chicago, Springfield, and Urbana, IL
Data Center Master Planning
EYP MCF provided a master planning that included determining the best options for data center consolidation, feasibility study, programming, data center layout, development of power and cooling loads and high-level design concepts, and ROM construction cost estimating.
The University of Illinois data center building has 81,000 SF and is capable to host almost 30,000 SF of scalable HPC machine rooms.
Rennselaer Polytechnic Institute
Nanotechnology Data Center Design
EYP MCF provided design for a new Computational Center for Nanotechnology Innovation (CCNI). The computational power of the current hardware configuration is rated at over 12 PF peak. The project included the conversion of an existing manufacturing building into a new, state of the art computing facility. The data center will include 5,000 SF of 48" raised floor area at 250-300 Watts/SF.
Louisiana State University
Baton Rouge, LA
Site Evaluation and Data Center Design
EYP MCF was selected by LSU to provide a Feasibility Study that encompasses
the following planning services:
• Site Utilities Infrastructure Capacity Assessment, Gap Analysis and Planning
• Conceptual Design to Develop ROM Construction Cost Estimate
• Assistance with Design and Construction Schedule Development (including identification of major long-lead procurement items)
Data Center Master Plan
EYP MCF conducted group interviews with the research and administrative computing committee, provided a technology and facilities needs analysis, gap assessment, master planning, and facility specifications.
EYP MCF's master plan included current- and future-state data center tier classification and space, power, and cooling requirements; future-state facility, technology, and infrastructure recommendations; conceptual spatial, electrical, and mechanical layouts/specifications and ROM cost estimates
West Lafayette, IN
Data Center Master Plan
EYP MCF collaborated with Purdue to provide an independent analysis and assessment of their entire data center enterprise, encompassing both business and research computing, that involves a current-state assessment, analysis of ITaP’s (IT at Purdue’s) master plan and proposed capital expenditures on facilities an
infrastructure, and a validation of Purdue’s cost estimates and priorities.
These analyses were used to provide improvements to the current facilities infrastructure, as well as move forward with future-state plans and developments to the overall data processing capabilities for the entire university model
University of California at San Diego
San Diego California
Site Evaluation/Master Plan/Retrofit
The San Diego Supercomputer Center (SDSC) at the University of California at San Diego plays a major role in building a national cyberinfrastructure. SDSC engaged EYP MCF for Data Center Assessment, CFD Modeling, and Master Planning services for upgrade/expansion scenarios.
EYP MCF provided the design for upgrades to the data center
based on the findings and recommendations presented in its study.
University of California, Berkley
Energy Efficiency Assessment/Energy Monitoring
Lawrence Berkley National Laboratory (LBNL), home of the Department of Energy’s National Research Scientific Computing Center (NERSC) and the Energy Sciences Network (ESnet), is a global leader in scientific computing and networking.
The University of California and LBNL selected EYP MCF to collaborate on a data center benchmarking project. The project consisted of collecting data from a series of data centers to establish a best practice protocol with respect to energy efficiency. Energy monitoring was used to quantify the energy use of the racks, HVAC systems, and other loads, as well as supply temperatures and flows.
University of Exeter
Site Evaluation/Master Plan
EYP MCF conducted a site survey at Exeter University. EYP MCF determined the extent of the infrastructure build-out required to provide the necessary power and cooling to the computer system. The proposed infrastructure solution took into account the relatively
short-life expectancy of the system (4 years) and provided the necessary power and cooling to suit the exact requirements of the computer equipment, which will allow for the most economic power and cooling systems to be installed.
University of Notre Dame
Notre Dame, IN
Data Center Master Plan
EYP MCF interviewed key stakeholders, users, and executives, including the new VP of Research, and provided strategic consulting services to deliver a high-level assessment of the overall enterprise data center technologies and facilities architecture. Completed a study outlining existing deficiencies and gaps with the current data centers and provided the University recommendations for remedies for moving forward. Determined the best options for data center and server room consolidation and determined the requirements for a
purpose-built data center to accommodate future enterprise and supercomputing needs.
- Data Center Due Diligence
Special Guest: Kevin Sanders
Managing Principal, Data Center Consulting & Strategy EYP Mission Critical Facilities