A short while ago, cloud computing was an abstract concept, difficult to both define and put into action. Now, sixty-six percent of businesses cited a plan to increase cloud-computing investments with big data analytics leading the way in use cases.[1] The concept of quickly spinning up and spinning down compute capacity can be invaluable and provides a competitive advantage in businesses where compute demand continually grows and peaks and valleys disrupt workflows.
But how does the concept of cloud computing match reality? What solutions deliver positive experiences? To qualify promised benefits of flexibility, agility and cost-effectiveness in hybrid cloud computing, ESG deployed an Avere vFXT cluster on Google Cloud Platform and began testing.
Cloud Flexibility: Workflow Control
In cloud adoption, identifying data needed and then moving that data for cloud processing can be both difficult and time consuming. ESG effortlessly moved an on-premises data archive to the cloud without disruption to client activity or the need to make changes to tools or scripts. Avere vFXT Edge filer works with storage and client workloads in on-premises data centers allowing the flexibility to choose where and when data should be made available to cloud computing resources. This flexibility allows companies to make decisions that make sense for their financial and operational goals.
Cloud Agility: The Right Data at the Right Time
Taking advantage of nearly unlimited cloud-based resources without impacting existing workflows and without making functional tradeoffs proved possible with the Avere vFXT. The ESG Lab tested a rendering workflow — running many distributed parallel simulations and scripted events to build a still image for a particular point in time into a video sequence. Testers were able to quickly and easily create a global namespace that made source data located on-premises available to cloud compute instances without having to copy the entire data set. In less than seven minutes, testers rendered 500 HD frames without moving data to the cloud. With the caching ability of the vFXT, 850 MB/sec read throughput performance was delivered to the cloud compute instances. This was a 42x performance improvement compared to retrieving data from the on-premises filer.
Cloud Economics: Big Data Analytics
To demonstrate the economics of the solution, ESG Lab looked at big data analysis, like that commonly performed in the sciences, federal agencies, insurance, and financial companies. Testers sought to find a balance between cost, performance, and interoperability using a cloud-based Hadoop deployment.
Results showed that the Avere vFXT greatly accelerates performance with edge caching and automated tiering. Data is made available to the Hadoop nodes through a standard NFS mount point, and all native applications can easily read and write directly to the data set from any location. Only the Avere instances needed to be powered at all times; the Hadoop nodes only needed power during the 8-hour processing runs, resulting in a 5x reduction in operational costs compared to an HDFS deployment.
Overall, ESG determined that the Avere Hybrid NAS solution offered the best balance of cost, performance and complexity for performing daily Hadoop MapReduce jobs in the cloud.
Learn more about how ESG Lab put cloud computing to the test. Download the complete Lab Validation Report: Optimize Hybrid Cloud Computing Environments with Avere Virtual Edge Filer.
[1] Source: ESG Research Report, 2015 IT Spending Intentions Survey, February 2015