Too much information and too little ability to process and understand: the first is a human problem. Computer technology is, of course, automation systems, editing and analysis tools by providing assistance and information has increased in size. But now this kind of websites, mobile devices such as sensors and produce an overwhelming flood of data processing systems and other data sources - or ignore - build strategies for data storage and past frames.
There is talk in circles, or "big data" - a new challenge for computer equipment and data storage, how to deal with this tsunami of information with new tools and techniques to optimize their existing investments in technology for data management.
At least, Forrester Research Inc. analyst Brian Hopkins, a fairly large volume of data requires a rethinking of traditional data warehouses and institutions. For example, Hopkins and the integration of certain types of large volumes of data in a central data warehouse, data marts separate hub and spoke connections will be a challenge, he said. Data warehouses primarily focused on structured data, he said. But what are classified as very large volumes of unstructured or semi-structured.
"In a sense, is to head down the assumptions major data storage," said Hopkins. The traditional answer specific questions focused on data warehousing and business users, business intelligence environments, data is cleaned and to extract, transform and load process, and eventually became Reports and ready for analysis. According to Hopkins an organization's data is generally less than 5% ", sometimes much less" is used.
A new twist on strategies for data storage
Instead, the data focuses on key strategies are often a lot of information. In addition, the version "full reality of the 360-degree compatible with the concept of a large database of heaven evaporates completely," said Hopkins. Directed by the so-called virtual data warehouses and analytical added complexity of managing storage hardware and data and may require heavy processing power, large data environments are common.
"The large data models with the most interesting and anomalies, there must be," Wayne Eckerson, business applications, TechTarget Inc. Director’s research and the media, said the architectural group. "But I have a lot of him because he is much more difficult to analyze large volumes of data we have. Expensive in this area." Many organizations are beyond traditional data storage and is in the open source Hadoop MapReduce so that this type of account, as emerging high-tech evidence, Eckerson said.
Richard Winter, president of Winter Corp., a consulting firm in Cambridge, Mass., data storage, focusing on the data before analysis to obtain a business opportunity would be difficult to identify different offers. It was developed for the treatment of asthma, a new "intelligent" sample showed an inhaler, an inhaler is used, and the wireless technology is a function of the patient's identity and sends the time and location of the base.
"It is, for example, are downloaded to a load of potentially allergenic soy products, soy powder, put in the air may be possible to combine this information with information," said Winter. These correlations can help medical researchers better understand the characteristics of asthma, he said.
Grand Challenges Data: Volume
However, there are potential problems of data management: the devices became widespread in the winter, millions of people worldwide to a substantial improvement, and begins to use the "generation of stored data needs to be said for a long time and is ready for analysis.”
Professionals and storage of data, important information you need to understand not only the volume, Philip Russo, the Data Warehousing Institute (TDWI), the managing director of research data says. "Other features are very important," he said - for example, large volumes of data, including Web browsing, can be very different, in fact, point of sales data and Call Detail Records, text messages, social networks and various types of information.
Large data installations face the challenges of managing, headlights like a deer famous Russo, many organizations, he said: They understand the potential value of knowledge, but over the data was abandoned with the challenges of managing a very traditional process of storing data is not particularly practical.
For example, customer behavior - - to prevent this type of stroke, Russo, investment advice, an area of high returns in an attempt to maintain a strong emphasis on the data to find ways to reduce and then apply knowledge of traditional and new tools to help you take control. "Select a topic to get something useful," he said. "Do not think that the need to keep track of all the questions."
No comments:
Post a Comment