such vast amounts of data . The ‘ lift and shift ’ approach is a temporary fix but just isn ’ t a cost-effective long-term solution ; it underscores the limitations of legacy systems . Instead , data should be flowing directly to the cloud , where it is better placed to take full advantage of the all the benefits the cloud provides .
Another problem is that a great deal of the data being generated is complex and of many types : structured , semistructured , unstructured . Yet many of the systems that manipulate big data are still row-and-column orientated . Therefore , you have this mismatch between data in the format that it was generated in versus the format it needs to be in so that the systems can actually process it , which results in a big challenge for businesses .
69
Is there still a place for handling Big Data on-premise ? If so , why would you ?
As much as I ’ d like to say that the cloud currently has all the answers , I think it ’ d be dishonest of me to say that on-premise is completely redundant . For some highly regulated industries ,
on-premise infrastructure is still incredibly important to meet specific security
www . gigabitmagazine . com