The adoption of Hadoop by companies looking to improve their big data analytics capabilities isRead More
Cloud misconceptions ‘hinder big data migration’
Although many companies will be finding it a challenge to put in place the necessary resources to support their big data analytics plans, a large number of businesses may be overlooking an obvious solution because they are unaware of what it can deliver.
Cloud computing has been around for a number of years now and is in many cases an established way of managing key operations. However, when it comes to big data, enterprises are still frequently reluctant to get on board.
It was noted by eWeek that there are several common misconceptions that prevent businesses from migrating their data analytics operations to the cloud.
The publication observed that the challenges involved with moving large, unstructured data sets into and out of the cloud have acted as a major barrier to the development and adoption of cloud analytics services.
For example, some companies believe that the sheer scale of their data warehousing solutions makes it impractical to move them to the cloud. But while this may be the case if businesses are still using old technologies such as FTP to transfer data, there are now a range of more modern protocols that have been developed specifically to move large quantities of information.
"This large file transfer technology was originally used to move huge video files during the production of film and television, but as demand has grown it has become available as Software-as-a-Service (SaaS) solutions both for person-to-person transfers and large-scale cloud ingest of any kind of large data sets," eWeek stated. "There are no file size limits in this solution, and therefore no data is too large to move."
Bandwidth and cost are other common concerns when it comes to migrating data away from on-premises solutions. Many businesses assume that the only way to ensure they have adequate bandwidth for such operations is to install a dedicated line, which can quickly become expensive.
However, there are managed SaaS services available that can eliminate any upfront infrastructure investments. Such models can often be very cost-effective, as they ensure businesses only pay for what they use.
Enterprises looking to move big data to the cloud also need to remember that bandwidth is not the only factor that can impact on the success of a migration. It was stated by eWeek that in many cases, it is actually latency that determines the speed of the transfer.
"In these situations, buying additional bandwidth won't make things faster," the publication said. "Newer SaaS transfer solutions, on the other hand, have to ability to greatly reduce latency so that you can fully utilise any available bandwidth."