A lire sur: http://www.cio.com/article/749994/Internet_of_Things_Will_Disrupt_Data_Center_Management_Says_Gartner
The Internet of Things will require a major rethinking of data center capacity management to deal with huge data growth, according to analysts Gartner.
Matthew Finnegan, Wed, March 19, 2014
Computerworld UK — The Internet of Things will require a major rethinking of data center capacity management to deal with huge data growth, according to Gartner.
The latest research from the analyst firm indicates that 26 billion sensors, devices and other gadgets will be connected to the internet by 2020, driving a product and services industry worth $300 billion (APS180 billion).
"IoT deployments will generate large quantities of data that need to be processed and analysed in real time," said Fabrizio Biscotti, research director at Gartner. "Processing large quantities of IoT data in real time will increase as a proportion of workloads of data centers, leaving providers facing new security, capacity and analytics challenges."
The IoT will connect remote assets online, and provide a data stream between these assets and centralised management system. These assets can then be integrated into new and existing enterprise system to provide real time information on location, status, functionality and so forth, using data analytics.
According to Gartner the magnitude of network connections and data enterprises will need to deal with as part of the IoT will drive the need for distributed data centre management, reversing trend in recent years where many large businesses have moved to centralise data centre operations.
"IoT threatens to generate massive amounts of input data from sources that are globally distributed. Transferring the entirety of that data to a single location for processing will not be technically and economically viable," said Joe Skorupa, vice president and distinguished analyst at Gartner.
"The recent trend to centralise applications to reduce costs and increase security is incompatible with the IoT. Organisations will be forced to aggregate data in multiple distributed mini data centers where initial processing can occur. Relevant data will then be forwarded to a central site for additional processing."
These new architectures will present data centre operations staff with significant challenges as they attempt manage the influx of data as a homogenous entity across different locations. This will create data governance issues for example, as business will need to take a selective approach to selecting which information will be backed up, with the storage of all raw data likely to be prohibitively expensive.