When Migration Looms: from Business Model Changes to End of Hardware Life
Data center migrations are an inevitable part of the technology lifecycle. They can be triggered by a number of business change events including:
- Current data center no long meets business needs for space and power
- Data center is outdated and needs to be remodeled
- Spending time and resources on the internal data center is negatively impacting the core mission
- Hardware has reached end of life
- There aren’t enough internal resources to manage an internal data center
- Digital transformation is in process and requires a different model for IT
But no matter the cause, one thing is certain: getting equipment from one location to another can be a difficult undertaking regardless of whether you’re considering colocation or fully managed hosting services.
Luckily, DataBank can help, by drawing on a long history of migrations and deep institutional knowledge on the best practices for making a migration successful.
Avoiding the Minefields of Migrations
The potential for downtime is often the greatest cause for concern when a migration is impending. It can affect the ability of customers, employees and partners to do business, and seriously damage reputation and brand. Without a proper plan, it’s not possible to move your production site without incurring downtime.
Determining whether the migration will be a “lift and shift” or a “progressive path” is a second key step. In some cases, lift and shift is easiest. The entire environment will come down at the same time, then back up. But in other cases, a multi-staged migration may be necessary as not all businesses are able to move their entire environment at once. These progressive path scenarios involve a series of smaller migration phases and can present a number of challenges of their own. For instance, if a particular app needs to access to a database in order to function, a planned database replication and cut-over may be needed or the application will fail.
Where applications and infrastructure need to meet stringent compliance requirements, migrations can become even more complex and additional planning, process, experience, and staffing are critical. While maintaining a compliant environment is one thing, migration to and from one often requires internal resources to work outside their primary competencies. Having staff or a partner with the experience to plan each step in the migration is critical to avoiding downtime and maintaining compliance.
Tapping Experienced Partners For Migration Success
A physical equipment migration isn’t usually standard expertise for an IT professional. When it comes to a colocation environment, moving equipment on your own, especially without the proper hardware certifications, can go awry easily. It’s a whole different skillset than typical racking and stacking.
This is where a proven data center partner that specializes in migration becomes critical. It’s not just about facilities that can meet customer needs for scalability. It’s about having the mix of services to aid in a migration and the experience to de-risk one. With a full complement of colocation and cloud options, DataBank has all the bases covered from a platform standpoint. Contract and revenue portability among our services means we can deploy whatever services are needed to facilitate a migration, even if the eventual mix of services changes. Our deep mix of managed services means customers can rely on DataBank for server patches, application support, database administration or network support during a migration. And our security and compliance certifications ensure a migration is conducted the right way, with a failback plan in place.
“No one foresaw that it would be necessary to become a part of the digital world in the way that the market is demanding today. Everyone’s asking themselves, who do I work with, and where, to make the digital transformation journey?”
-Bill Pratt, VP, Managed Services & Channel Sales
DataBank’s Purpose-Built Migration Process
For both colocation and cloud migrations, DataBank has developed a proven methodology to maximize success. Colocation migrations require the physical move of equipment and often hardware vendors require certification, and sometimes insurance, for moving equipment because doing so without them can result in a voided warranty. For these migrations, DataBank enlists the help of specialized third-party providers who can provide a full team to handle the physical aspects of the migration. The process starts with a detailed equipment inventory and then the team consults with both DataBank and our customers to develop a layout of the new proposed space with rack elevation and placement. This makes it easy to pre-build the environment, from racks and cabling to power cords.
Migrating from a physical environment to cloud, or from cloud to cloud, requires a different type of skillset. For this specific purpose, DataBank has an entire team and toolset exclusively dedicated to migration. In a sense, migration is effectively a DR event in that it requires failover from one site to another. Our migration team utilizes proven DR tools and best practices for making the move which fall into five migration steps:
- Plan
- Design
- Build
- Test
- Go live
“The fact that we actually have our own project management and migration teams is certainly a differentiator for our customers. Data center providers who only offer colocation aren’t going to be a good fit for a hybrid environment with 500 servers that need to be moved.”
-Justin Sevier, Director of Hosting Solutions, DataBank
Preparation and Process are the Keys to Migration Success
The most important aspect of migration, whether physical or to the cloud, is preparation. If you’re not thorough it’s practically a guarantee you will have problems. A successful migration is not just about getting from one environment to another, it’s about considering every element of the change process and the impact to every aspect of your business and applications. For instance, in what order do elements need to be activated in order for applications to come back online? If you were to boot your database server before your domain controllers are online, will your application be able to authenticate? And what are your business hours? 8-5? 24/7? When are your systems least utilized? Can you afford downtime? These are the planning factors that have to be taken into account and having a dedicated migration team results in a comprehensive understanding of the environment, as well as what’s going to be involved to make the move.
If you are facing a data center migration and you’re looking for feedback or support, give DataBank a call and we’ll show you how we leverage world-class platforms, process and people to make the move successfully.
Understanding the Unique IT Infrastructure Needs of Universities as Super Computing Becomes a Key Focus Area
The IT infrastructure needs of universities are very specific to higher education, and demand for upgrades is rising as technology continues to play a more integral role, both in the classroom and critical research initiatives. When it comes to higher education, especially engineering and scientific research, the pursuit of information is never-ending. The only way to analyze and answer complex questions that arise along the way is with high performance computing systems. Consequently, astounding amounts of data are being gathered and stored; researchers working for universities are generating billions of files and petabytes of data every day. This data often carries the progress of incredibly important initiatives: climate change, cancer research, and more.
The computing needs of scientists, engineers and researchers will only continue to grow. Fortunately, HPC is a way to effectively harness new scientific opportunities. A few examples of how HPC is impacting the world around us:
The MIT Lincoln Laboratory Supercomputing Center is focused on accelerating U.S. research in machine learning, device physics and autonomous systems (Dell).
NASA Center for Climate Simulation executes on research science initiatives using HPC, giving scientists the ability to access to large data sets and powerful compute resources (Dell).
The University of Pisa built an HPC infrastructure designed to deliver the performance, storage capacity and compute power required for advanced research in numerous competitive fields (HPCwire).
HPC requires highly specialized IT infrastructure and data center designs because the performance demands are staggeringly high. Georgia Tech, for example, recently built a new HPC system that supports data-driven research in astrophysics, computational biology, health sciences, computational chemistry, materials and manufacturing, and numerous other projects. It will also be used for research that improves the energy efficiency and performance of the HPC systems themselves. This necessitates specific needs for proper heating, cooling, space and managing massive workloads. Such an endeavor is best undertaken with the support and partnership of an advanced data center provider well-versed in the IT challenges of HPC. Last fall, DataBank built a high-density solution for Georgia Tech to solve for this very issue.
Supercomputers have the potential to improve human lives through discoveries and advancements in science and research. This is why it’s so critical for universities to have access to a data center and infrastructure provider with experience in helping higher education institutions create roadmaps for their current and future demands, from configurations to reliability.
Planning for research initiatives that require super computing capabilities, like genomics or deep space analysis, are major drivers for attendance and acceptance rates, as well as attracting new professors that bring value to the institution. As such, these initiatives must be factored into planning around compute. It’s important to remember that technology is a direct tie-back to the value proposition a university can offer to its constituents.
What to Consider When You’re Developing an HPC Platform
If your university is planning to develop an HPC platform, there are several considerations that need to be made with respect to IT infrastructure and data center planning. When we partner with customers in higher education, such as Georgia Tech, we start by walking technology decision makers through existing infrastructure and evaluating a number of factors to determine the changes that need to be made.
- Existing assets
- Potential to repurpose
- Power density needs
- Reliability needs
- Computing demand profile
“We help universities break through requirements and estimate their demand profile so they’re able to make a decision in terms of how much capacity they need and what type of design they should seek out. It’s a collaborative process with DataBank sharing much of its expertise.”
-Justin Puccio, EVP, Corporate Development, DataBank
Adopting A New Model for Enabling HPC in Higher Education
Universities commonly operate data centers that have been built over time in an ad hoc fashion. Often, these facilities run legacy infrastructure powered by antiquated gear that supports important prospective programs. It’s time for a new model for enabling HPC in higher education, wherein universities upgrade outdated infrastructure and future proof for the next ten years. A campus data center should be an opportunity to showcase the importance a school places on science and technology.
The new computing paradigm for education institutions means abandoning aging infrastructure and embracing an entirely new approach that will drive the goals of super computing, and by extension, the university. To accomplish this, it’s necessary to solve for the challenges of heating, cooling, space, and managing massive workloads. Consider forming a partnership with an advanced data center provider well-versed in the IT challenges of higher education, HPC in particular.
Building a high-performance computing center is an expensive undertaking, requiring specialized talent in construction and operation as well as significant capital to build and operate a cutting-edge facility. By partnering with DataBank, GeorgiaTech was able to reduce the upfront costs while still achieving its goals of a world-class facility.
Working with a data center partner is also valuable in the context of higher-ed HPC initiatives because the right partner will work with you on your security and compliance concerns. Whether you handle data subject to regulatory entities or you’re trying to secure your environment to protect valuable information, it’s possible to gain the proper power, cooling, and space for your environment and secure it in one fell swoop.
Backups and disaster recovery are important considerations, too. Research moves fast, and backing it up regularly and properly will protect your findings as they evolve.
Benefits of an HPC Partnership
- No need to expend capital dollars to maximize super computing capabilities
- Eliminate the burden of maintaining infrastructure, redirect focus to core research activities, and extend budget as a result
- Universities focused on medical or government agency research can gain access to an HPC expert with security and compliance expertise, from FedRAMP to HIPAA and HITECH
Take it from us: it’s not a small commitment to entrust critical infrastructure components that underpin research efforts, and it calls for an expert partner you can trust. If your technology team is looking to pursue high performance computing initiatives, but is struggling to figure out support and financing, consulting with a specialized data center provider is a pertinent next step.
Next Steps to Formalizing an HPC Strategy at Your University
We live in an incredible time during which scientists, engineers, and scholars around the globe continue to make progress on discoveries that are shaping our future. An HPC solution fosters research on these discoveries, and enables the ability for universities to solve complex problems. Further, many organizations outside of higher education institutions are equally invested in participating and benefiting from ground-breaking research. A new HPC model can attract and build upon these relationships.
Not to mention, it’s important to empower your research team to execute on their search for knowledge, but it’s equally important to protect the fruits of their labor. You can solve this problem by taking advantage of a data center that will give you the space, power and cooling you really need to use HPC effectively.
Fully upgrading a legacy IT infrastructure and creating a complete HPC strategy are decade-plus investment decisions, which is why the right expertise is so critical. DataBank possesses the expertise, skill, and financial strength to be able to meet the unique needs of university clients. DataBank also specializes in providing such environments and the managed services that complement them so that your university doesn’t have to lay out excessive capital for IT initiatives that would be better spent elsewhere.
If you’re seeking to identify the best way to protect your HPC technology and support your researchers, get in touch with DataBank today, or call us at 1.800.840.7533 and speak to an expert.
UPCOMING EVENT
If you would like to learn more about how DataBank can assist universities with the technology, operations and business model for an HPC facility, sign up for our webinar co-hosted with EYP Critical Facilities on Sept 26 at noon ET