Cloud storage is often thought of as a giant repository in the sky, where old backups and archives go, seldom to be seen again. But this may not always be the best cloud use case for data centers. Storing a large amount of data for a long period of time in the cloud may be too costly compared to storing that data on-premises. Sometimes it makes more economic sense to use the cloud for short-term services, like providing compute resources for the rapid processing of data, than as a more permanent resource like long-term storage.
There are essentially three cloud use cases that data centers should now consider. The first is using the cloud for long-term storage with a caching solution for local performance of active data. The second can be a caching solution for increased performance of the cloud compute engine itself. The third use case is as a caching solution both in the cloud and on premises so that large data sets can be stored in the data center after the cloud has finished its processing. This third use case is the newest and may be the most ideal balance of the temporal advantages of the cloud and the permanence of the on-premises data center. We discuss the strategy in detail in our upcoming white paper “Reverse Your Cloud Strategy”.
Avere’s Edge Filer technology extends the caching needed for each use case by providing a high performance filer (virtual or physical) at the right inflexion point. The edge filer also leverages a global file system to make movement between cloud and on-premises storage seamless. For the first use case Avere’s Physical FXT Edge Filer allows data to be cached locally and then automatically aged out to cloud storage, essentially automating the giant repository use case described above. For the second and third use cases the Edge Filers are virtualized, allowing data to be cached in the cloud for improved compute performance. The third use case combines a virtual filer in the cloud and a physical edge filer on-site, enabling data to be stored cost-effectively on-premises after the cloud is done processing it.
Cloud NAS + Google
Google is going after the enterprise with its compute and storage platforms but, of course, enterprises need a method to get their data into the Google cloud. According to Avere Systems, their new Physical and Virtual FXT products are good ways to do this. The Physical FXT is ideal for the first cloud use case, leveraging cloud storage as a giant repository. Avere now supports moving data to Google and/or to Amazon based on access pattern or other user policies. The Virtual FXT solution allows for data to be cached within the Google Compute Engine’s RAM or SSD tiers for faster computational performance. Then, if the organization chooses, it can be cached back to the data center for long-term storage.
Google Storage and Compute
Google cloud storage provides limitless scale and user customizable high durability. The data can be made globally available to ensure an always-online state. It has two tiers of storage, both low cost object storage and lower cost cold archive. Avere supports the use of both tiers. Finally, it provides a seamless integration with Google Compute Engine.
The Google Compute Engine (GCE) provides global, high performance networking, enabling large compute clusters with consistent cross-machine bandwidth. It claims to offer extremely high performance per virtual machine and, of potentially more importance, predictable high performance per virtual machine. Probably the most unique aspect of GCE is the true ability to pay for what you actually use, with per-minute metering. This makes it ideal for the second and third use cases described above.
StorageSwiss Take
The use cases for the cloud are now coming into clear focus. There are two hybrid use cases. This first is for cloud storage as a giant repository and the second is where the cloud is leveraged for compute, but the data center is the repository. In both caching data either to or from the cloud makes the interaction seamless. In between these two is use case number three, the 100% cloud. Virtual caching in the cloud provides value by accelerating the storage I/O component of cloud compute.
We believe that data centers will eventually leverage all three use cases for different types of data and compute needs. Avere, thanks to both its physical and virtual FXT implementation, allows an organization to use all three use cases, while having a single solution to facilitate movement to and from the cloud.
Click To Register
George Crump is the Chief Marketing Officer at VergeIO, the leader in Ultraconverged Infrastructure. Prior to VergeIO he was Chief Product Strategist at StorONE. Before assuming roles with innovative technology vendors, George spent almost 14 years as the founder and lead analyst at Storage Switzerland. In his spare time, he continues to write blogs on Storage Switzerland to educate IT professionals on all aspects of data center storage. He is the primary contributor to Storage Switzerland and is a heavily sought-after public speaker. With over 30 years of experience designing storage solutions for data centers across the US, he has seen the birth of such technologies as RAID, NAS, SAN, Virtualization, Cloud, and Enterprise Flash. Before founding Storage Switzerland, he was CTO at one of the nation's largest storage integrators, where he was in charge of technology testing, integration, and product selection.
StorageSwiss.com - The Home of Storage Switzerland 