If several large projects require the same technical computing resources for a two week period, what should you do? Or what happens if you are experiencing spikes in infrastructure demand at a certain period during the day--do you purchase additional infrastructure to accommodate that increase in demand even though those resources will be underutilized throughout the rest of the day?
Simply adding hardware to your local IT infrastructure to address peak workload requirements, especially if they are short-term, may not be the best option. Expanding your infrastructure can require significant time, expertise and budget. And your data center may already be maxed out on power, cooling and real estate. If this is the case, then a hybrid or public cloud deployment may be a good fit for your organization.
However, the extraordinary performance demands that engineering, scientific, financial, and research workloads place upon technical computing infrastructure has historically meant that these workloads are not suitable for cloud deployments, especially those that use virtualization technology. Depending on details such as network topologies, the type of virtualization technology used, and what else is running on the service, users can experience poor or unpredictable performance. Issues like “noisy neighbor” effects, spikes in network utilization, or inconsistent file server performance can make it difficult to guarantee a consistent level of performance.
In addition, it is challenging to ensure privacy and data security on a shared virtualized infrastructure. While you may have the best firewalls and encryption technology between your own facility and the service provider, if you are sharing infrastructure with other tenants, there is always a risk of data being compromised or accidentally exposed to others.
Fortunately, there is a good solution to address these issues. IBM® Platform Computing™ Cloud Service provides you with ready-to-run clusters in the IBM SoftLayer® cloud. These clusters are optimized for compute-intensive technical computing and analytics applications, and come complete with SoftLayer infrastructure, IBM Platform™ LSF® or IBM Platform Symphony software-as-a-service (SaaS) workload management software and Elastic Storage data management. The entire Platform Computing Cloud Service is fully supported by the Platform Computing Cloud Operations team, eliminating the cost and time associated with managing the environment yourself.
The Cloud Service offers a security-protected, high-performance environment running on bare metal servers (physical machines). Security through physical isolation provided by dedicated physical machines and networking, as well as control of data center locality, ensure that you can safely extend your workload to the cloud. You can also be confident that you will have the performance needed for demanding technical applications. Tests comparing the performance of Platform Computing Cloud Service running on bare metal against competitive offerings showed that IBM delivers the same or better absolute performance in all test categories. For more information, download the IBM white paper, The IBM Platform Computing Cloud Service: Benchmarking performance.
If a hybrid cloud implementation works best for your organization, the Platform Computing Cloud Service enables you to forward workloads from local infrastructure running Platform LSF or Platform Symphony to a ready-to-run cluster in the SoftLayer cloud to quickly accommodate increases in demand. And because the same workload management software, application-centric interface and data management software is being used locally and in the cloud, the user experience is seamless.
If you are experiencing new or emerging resource demands, you might not have the on-site resources required for a hybrid cloud implementation. In this case, a stand-alone cluster running on the SoftLayer cloud provides the compute and storage resources as needed, accessible on a temporary basis with drastically reduced wait times. Instead of the weeks or months it would take to acquire and implement an in-house cluster and the skills required to manage that cluster, you can access a ready-to-run cluster on a temporary basis in as little as a few hours with the Platform Computing Cloud Service.
With powerful cloud-based solutions like these, your organization can easily meet increases in infrastructure demand without compromising on security and performance. The Platform Computing Cloud Service enables organizations with compute-intensive workloads to increase capacity and gain a competitive edge by doing more analysis, research and simulation in less time and with less expense.
To learn more about whether the hybrid cloud is a good fit for your organization, view this webinar. For more information about IBM Platform Computing Cloud Service, please visit: ibm.com/platformcomputing/products/cloudservice