Boosting a university’s computing performance with an Intel-based hybrid cloud server

Nanyang Technological University (NTU) is one of the world’s fastest rising universities, ranked 39th globally and 1st among the world’s best young universities. 3,800 teaching and research staff educate about 33,000 undergraduate and postgraduate, local and international students.

Students and faculty members of its College of Science, Engineering and Earth Observatory department handle projects that often rely heavily on compute resources. The projects were getting increasingly intensive and required computing resources that were challenging for NTU to accommodate.

NTU needed improved computing resources to address the constant increase of large-scale research demands

With more students doing increasingly complex research, NTU needed to boost its compute capacity and provide improved performance. There was a need for a secure, open solution that allowed NTU to have computing resources with unlimited capacity and platform abstraction. This solution not only had to be able to handle the growing computer requirements of the faculty and students, it had to do so in the most cost efficient way possible.

Improving and maximizing current computing resources for unexpected loads
Staff and students worked on projects that required complex computation tasks, and the intensive processes created a spike in compute demands that were putting a strain on NTU’s resources. Instead of purchasing new resources to keep up with the research-heavy workload, NTU needed a cost-effective way to improve its current existing ones. This solution also needed to be flexible and scalable, enabling it to accommodate these demands as and when they occurred.

Creating a means for remote disaster recovery
Because of all the research-intensive projects, NTU needed an infrastructure that allowed users to set up remote disaster recovery to prevent data loss in case of any disruptions. This infrastructure needed to be affordable, yet had performance and latency close to physical machines.

An Intel-based flexible hybrid cloud solution as a powerful answer

NTU worked with Red Hat, a software company providing open-source software products, to create a scalable, open-source hybrid cloud to tackle this multi-faceted challenge.The stability and high performance of this cloud gave it the power to cope with the increase in compute demands, while its quick scalability made it agile and dynamic enough for a rapid response to sudden surges in resource needs1.

Utilizing this solution allowed us to showcase ingenuity that can positively impact day-to-day research initiatives globally.

Components of NTU’s Intel®-powered hybrid cloud

Red Hat Cloud Infrastructure*
Lets users build and manage an open, private cloud. It comprises of Red Hat Enterprise for traditional scale-up workloads, Red Hat Enterprise Linux Open-Stack Platform for scale-out workloads, and Red Hat CloudForms for unified management across the technologies.

Amazon Web Services (AWS)*
A virtual cloud platform that allows flexible, scalable servers that can be called upon to accommodate large workloads. It provides users with complete control of their computing resources and lets them run on Amazon’s proven computing environment.

NetApp SnapMirror*
Enables NTU to replicate volumes from its private NetApp storage to another NetApp storage linked to AWS via AWS Direct Connect. Not only does this allow data to be transferred at high speed, it provides data recovery so users can back up their research and data in case of emergencies.

Intel® Xeon® processor E5-4650L
The Intel® Xeon® processor E5-4650L is an agile, efficient, high performance 8-core, 16-thread processor that is used predominantly on both the NTU and AWS clouds. As a result, communication between the technologies is virtually seamless.

 

Intel® based hybrid clouds: helping NTU meet the compute needs of its students and faculty

Because of this multi-pronged, yet integrated approach, NTU created an innovative hybrid cloud solution model that maximized seamless flexibility between private & public clouds. This provides elastic computing while maintaining lower overhead costs. Not only did this Intel-based solution enable NTU to have processing power on demand, it also acted as a model that can be adopted by and tailored for other institutions.

If you want to set up a hybrid cloud for your organization:

Cloud Computing


Get the 3-part EDM series on cloud computing for your business.

Product and Performance Information

1All performance tests were performed and are being reported by Nanyang Technological University. Please contact Nanyang Technological University for more information on any performance test reported here. Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information, go to www.intel.com/performance.