Several value areas where data centers focus on large data

At first glance, the impact of large data on the datacenter is like another high-performance application that requires more processing power, more http://www.aliyun.com/zixun/aggregation/17325.html "> Storage space, And a more high-performance network. In short, this seems to be another burden for the datacenter. In fact, however, large data is not just a tool for business analysis, it can also be a useful tool to help improve data centers.

Better IT Security

IT security has become the daily management project of the General Enterprise, and it has become one of the most concerned issues in IT management. One of the biggest problems with hacker malicious attack threat management is the need to investigate and identify massive amounts of data and track potential intruders into your enterprise's IT infrastructure.

Traditionally, this approach has focused on a single system, preferably with centralized monitoring to identify and correlate different threats. However, given that businesses have dozens of or even thousands of devices and applications, this approach quickly becomes untenable because all devices and applications are potential targets.

This requirement is met by large data that can help close to real-time processing around massive amounts of data. Any supplier, whether an experienced veteran or novice to the market, can quickly enter the field. The benefits are obvious: two or three events in unrelated systems may expose a subtle attack. The use of a system to consolidate, analyze, and determine the nature of these events, and to prevent the occurrence of attacks.

The value of large data

They cannot blindly deny useless data, they also contain value, they should be said to be low value density data. Therefore, the enterprise should also retain this data, but has not found its value for the time being, you can use Low-cost storage server to retain them.

There are some habitual spelling mistakes that people get when searching through search engines. These error data, while seemingly meaningless, can be found in a large number of user habits and patterns by collecting the data.

People are puzzled by the massive numbers because they are not able to master the whole data and just see broken, fragmented, local data. This is like usually "refining the data" and in the process of summing up often lose a lot of detail, but a lot of valuable information is hidden in specific details.

For different value data to be differentiated, the value of the data is not high stored in low-cost environment, but it will not be discarded, because in time, it will still have a certain value will be dug out.

Privacy protection remains to be resolved

You should provide some protection for user privacy, such as through data encryption, so that only people who need to know the data can understand, contact, or access the data. He wants the public to understand that, often through data mining software, the user data or information actually seen is not directed at a specific person.

Many national legislatures or governments want to protect the user's information by enacting privacy laws. However, because of the lack of understanding of data analysis, some regulations may ultimately limit the user's use of information and data.

Capacity management

Companies like ebay have announced successful use of large data to analyze and optimize their data centers. Just as virtualization technology can help to adjust the lack of enterprise capacity, these efforts can achieve the same effect. Again, capacity management brings a near-perfect use of large data: a single hardware may run multiple virtual machines, multiple disk images and countless applications, and it will take months to excavate the full image.

The next step in this evolution is to combine "initiative" with capacity management analysis so that the virtual machines can be provisioned and redistributed in real time based on historical and predictive requirements and other metrics. Imagine that your datacenter capacity intelligence redistribution itself is based on a newly released product or based on seasonal requirements. Although the technology is still in its infancy, it could affect the next generation of data centers.

Better monitoring

Tying these technologies together is a new generation of monitoring tools, powered by large data. Although traditional tools can be identified when a fault occurs, there are many cases where corrective actions are taken automatically, most of which lack extensive predictive capability and rely on user configuration of alert thresholds and metrics.

If you support large data active monitoring, your organization's Data center monitoring tool can predict hardware failures in a database server. The tool will intelligently assign the affected application to another server and notify the person concerned, once the problem has been corrected, to restore the operation.

Even security issues can be integrated into this "smart" data center and then isolate damaged applications or infrastructure, just as most anti-virus software "isolates" the virus-infected files.

Although large data is just beginning to affect the datacenter, it is a technology worth developing because it offers so many opportunities to help protect, fix, and optimize modern data centers.