Subscribe to our blog
Thanks for subscribing to the blog.
March 15, 2020
Topics: Cloud Tiering Data TieringAdvanced6 minute read
Technology has completely reshaped consumer expectations in the last few decades, creating a world where services of every kind are expected to be available at all times, from any location. Educational institutions are gradually waking up to this new reality, recognizing the importance of digitalization in maintaining their rank and prestige. To this end, many institutions have begun to adopt digital learning models and environments which include online learning, increased connectivity, highly available content and IoT devices.
These rich learning environments dramatically increase the number of workloads, apps and devices, generating an explosive amount of data. This flood of data can be overwhelming for existing architecture, causing data centers to run out of storage space and resulting in degraded performance.
In this blog post we discuss the storage challenges educational institutions face as they step into the digital age and show how NetApp’s Cloud Tiering service for AFF and SSD-backed FAS systems can be used to tackle these issues, taking an in-depth look at the case study of an Ivy League university.
The Pain Point of On-Premises Storage for Educational Institutions
Schools, colleges, e-learning platforms, and universities rely on vast data repositories spanning student enrolment all the way to graduation and alumni. Data is not only necessary for running day-to-day operations but can also be used to generate data-driven decisions to boost ranking and competitive advantage.
The architecture must be able to handle unpredictable amounts of data, both structured and unstructured, of varying types including courses, research, images, video and test results. An additional challenge is that the data stored includes sensitive and personal faculty and student information, including medical and financial records which are subject to stringent regulation. This has led to security and privacy posing the top two IT challenges for higher education in 2020. But even before tackling security and governance challenges, IT departments must contend with an even more basic and urgent issue—running out of storage space.
As they increase their digital offerings, educational institutions suddenly have to deal with data generated at an exponential rate. The data, much of which is used infrequently, accumulates quickly, taking up space on-premises and racking up hardware and operational costs. This can also impede performance and even lead to outages, seriously damaging the institution’s reputation.
To survive and thrive in this reality, IT teams must find a way to handle their infrastructure cost-efficiently and ensure it is highly available and can scale to meet future storage needs. One major way to transcend the limitations of existing infrastructure is to leverage cloud storage.
The cloud is already the destination for education. The use of the cloud in the sector is increasing, and its value is anticipated to top over $25 billion by next year. This growth also comes from the increased adoption rate attributed to the enterprise-grade solutions offered by the cloud vendors. Proof of this for example is that the email systems of nearly 70% of North American higher educational institutions have been transferred or are in the process of being transferred to the cloud.
While migrating to the cloud comes with its own set of challenges and rewards for the education sector, NetApp’s Cloud Tiering provides an immediate and simple way for institutions to adopt a hybrid cloud model and maintain their high performing on-premises architecture and at the same time instantly enhance their storage capacity via the cloud.
Cloud Tiering: What It Is and How It Can Optimize Performance
Cloud Tiering extends your on-premises high-performant storage to the cloud. Using NetApp’s FabricPool technology, Cloud Tiering automatically transfers inactive data from flash-based AFF and SSD-backed FAS NetApp systems to lower-cost object storage in the cloud (Amazon S3 in AWS, Google Cloud Storage on GC, and Azure Blob Storage in Azure), all the while preserving existing ONTAP efficiencies. When the data is needed again, Cloud Tiering automatically “reheats” it, transferring it back to the on-premises storage. Thus, only active data is stored on high-performance SSDs.
Cloud Tiering operates silently behind the scenes in the block layer, with zero disruption to the application layer and no effect on performance. By easily extending cold data blocks to the cloud, educational institutions gain an immediate solution for their ever-expanding data repositories, and reap numerous benefits:
- Storage space is instantly freed up reducing on-premises data footprint.
- Performant storage is dedicated solely to mission-critical high-demand workloads.
- Hardware, space, and storage costs are significantly reduced (by an average of 80%) by offloading cold data to lower-cost cloud storage.
- Storage can scale up and down as needed and customers only pay for the data that is tiered.
- And many other important use cases.
To get a better grasp on how Cloud Tiering can ease IT operations in the education sector, let’s take a look at a specific use case.
Major Ivy League University Case Study with Cloud Tiering
With more than one billion dollars in annual funding, this private East Coast Ivy League university is the fifth oldest higher educational institution in the United States. The university boasts diverse undergraduate and graduate programs, as well as over 200 research centers and institutes.
Having adopted a cloud-first strategy, the university’s IT team prepared to migrate the majority of their workloads to AWS. The team quickly discovered, however, that the nature of the on-premises operations made migration extremely difficult. As a result, the team found itself under mounting pressure to advance the cloud-first agenda and at the same time met with resistance to purchase much-needed extra on-premises hardware.
To solve this problem, the university turned to NetApp to help them tier 120 TB of data to the cloud. This was done via Cloud Tiering, which automatically identifies data that is infrequently used and seamlessly transfers it to AWS S3 object storage. When the cold data is needed once again, Cloud Tiering automatically moves the data back to the high-performance tier on NetApp AFF storage. By using Cloud Tiering the university gained multiple benefits:
Tiering Snapshots of Active VMDK Files
Using Cloud Tiering’s snapshot-only policy, the university’s cold snapshot data is currently transferred automatically from on-premises AFF storage to AWS S3 after a brief cooling period. Because operations mostly rely on the latest active snapshots data, older snapshots can be tiered to AWS S3 without impeding performance.
By tiering the snapshots’ cold data from a volume containing VMDK files, the IT team was able to transfer 20 TB of snapshots data, reducing storage costs dramatically.
Tiering Decommissioned VMs
The storage team used to retain decommissioned VMs on-premises in lower-cost storage for six months before deleting them. With Cloud Tiering, the team can now tier whole volumes directly to AWS S3. In addition, all other active data that has not been accessed for a defined period of time is tiered by default to Amazon S3.
By tiering these decommissioned VMs, the university managed to clear 100 TB of data from its in-house storage, leading to significant savings on space, electricity and hardware.
Conclusion
As they expand their digital presence, educational establishments around the world must find a way to handle a stream of data that is getting stronger by the day. As a result, storage teams find themselves under mounting pressure to provision more and more in-house storage. The data, most of which is infrequently used, piles up quickly and takes a toll on overall performance.
With NetApp’s Cloud Tiering, storage teams can offload inactive data to low-cost object storage in the cloud. This frees up on-premises storage to do what it should do best—ensure that mission-critical and latency-sensitive workloads have the performant storage they need.