Organizations in various places have big data of various shapes and sizes. These organizations are aware of their importance, opportunities, and the urgent need to pay attention. Obviously, big data will continue to evolve whether you ignore it or not. Organizations that have mastered big data (multi-structured and confusing data stored before their value is known) are improving organizational efficiency, increasing revenue, and developing new business models. How did they do it? The successful methods of these institutions can be summarized into seven recommendations. How to take advantage of big data to achieve success 7 major recommendations to see first 1. Promote long-term considerations with short-term considerations Those who are worried about whether they can keep up with the trend of big data are not just one. Everything is changing fast, so you don't know which tools, platforms, or methods are the best this year or next. Relax your heart. This fast-changing situation can serve you. Every year, suppliers are constantly improving their use of big data. Relationship and online trading systems (OLTP) are becoming more efficient and intelligent, whether they are running internally or in the cloud. The development of technology will ease the relationship between Hadoop and the data warehouse. Moreover, there will always be a product launch that will more precisely meet your specific needs. Therefore, please be relieved. Keep an open mind about adopting new products, as long as these products provide enough value to be justifiably integrated into your existing environment. Maintain a business intelligence platform that can connect directly to multiple formats. You are now ready to respond to market changes. 2, see the wrong choice What do you need, Hadoop or data warehouse? Needless to say, this is really a trap. Not only does Hadoop work well with data warehousing, but organizations can actually benefit from the collaboration between the two. Data warehouses are ideal for compressing important structured data, and can store data where business intelligence tools and dashboards can be easily found. But its weakness is that the analysis process and some types of conversions are weak and slow. This is exactly what Hadoop adds. In addition, although Hadoop is weak in interactive query and data management, it is good at quickly accommodating raw, unstructured and complex data. The two together form a symbiotic relationship. For example, imagine the data that top management uses to predict inventory demand for the next year. The data set can be large, with little time to model, restructure, and have no time to preprocess the data for use in the data warehouse. After the top management is used up, it may be discarded in only one week. This is where Hadoop comes out, refining the data and sending the samples to the data warehouse. “Big data is not a substitute for data warehousing,†Mark Nature, CEO of Third Nature, wrote in his article “The True Meaning of Big Data.†“It’s not a separate island. Big data is part of the new IT environment.†Don't miss the problem of choosing Hadoop or data warehousing. You can and should use both. 3, condense big data, so that one can see After big data visualization, it is concentrated to a level of clarity. A report by the Aberdeen Group found: "In organizations using visual discovery tools, 48% of business intelligence users can find the information they need without the help of IT staff." Without a visual discovery tool, the percentage dropped to 23 %. In addition, according to the study, managers using visual data discovery were 28% more likely to find timely information than managers without visual data. Perhaps most importantly, when it comes to big data, the report finds that visualization can also facilitate interaction with big data. Managers who use visual data are more than twice as likely to interact with data as other managers (33% vs. 15%). In addition, these managers are more likely to be impromptu questions, often inspired by the insights of the previous moment. Exploring the data in a visual way, making the data vivid and allowing the brain to understand instantly. Dana Juber, vice president of strategic planning at Wells Fargo, said: "There will be some ingenious ideas that this spreadsheet can't bring." With visual analysis, you can do two things at any time: • Change the data being viewed because different issues often require different data. • Change the way you view your data because each perspective can answer different questions. With these easy steps, you're in a state called Visual Analytics Loop: Get data, view data, ask questions and answer questions, and then go back and forth. Every time, your inquiry will deepen with your insights. You can search down, search up, or search sideways. You can add new data. As the visualization accelerates and expands your thinking, you create views one after the other. Once you're ready, you can share it. Ask and answer your own questions at the same time to accelerate the insights, actions, and business outcomes of the entire team. View and interact with real-time network dashboards on your desktop, iPad, or Android tablet. 4, give users a deep insight Do you know people who are fascinated by insights? Nothing can stop these people. They are constantly asking new questions and creating new value until they are done, satisfied, or until they need to fill out an IT application. With big data, you will become more fascinated by insights. Then there is enthusiasm and dedication. Now, apart from self-service data analysis, there is no time to do anything else. “Compared to other organizations, organizations with big data have more than 70% more chances to have BI projects driven primarily by the business sector than by IT groups,†Aberdeen Group recently released “turning to big data or bankruptcy? Maximize? Analysis and the value of big data," said. With big data, business users no longer have to endure the old, slower ways of IT, and the way this method publishes data is like a chapter. In The Value of big data, Third Nature research analyst Mark Madsen wrote: “Not only is the business intelligence publishing model old, but the business intelligence model expects the same environment in which information is used. Outdated." It's like reading through old lights or candlelights - the so-called "hard work". Madsen continues to write: "Big data is like electrical lighting, illuminating the dark corners of the past. It provides not only brighter light, but also the ability to take it on demand. It doesn't take months to wait for the data to be cleaned up. And it's ready to use, and now you can use big data technology to detect and discover the value in your data. If the data is valuable, it can be sent through a more rigorous process for use in the data warehouse." Don't force users to “work hard†and give them the ability. 5. Combine small data into larger data At a closer look, the composition of big data is nothing more than: a lot of small data sets. Each data set itself may provide value. When combined, it can provide great value. For example, in the consumer goods industry, high-level management can fully understand consumer behavior only after mixing consumer confidence data and purchasing data. Unilever Cooma, director of analysis at Unilever, said: "With membership cards, you can get rich data. It is necessary to mix all of this data to understand why consumers want to enter the store." Unilever can use this to predict Popular products and new trends. It's important to mix all of this data to understand why consumers want to shop at the store. †– Unilever Reyk Kuma The bulk of value is attributed to organizations that mix relationships, semi-structured, and raw data – minimizing upfront costs without the business users being plagued by technology. This is ok, and it is good enough. Whether the data is in a spreadsheet, database, data warehouse, open source file system like Hadoop, or all of the above, you need to be able to connect and integrate data flexibly and quickly. No matter how big or small your big data is, you can think of it and ask questions. 6, to ensure that big data is free from big trouble Big data is as fun as a sandbox. You can go in, build and shape things, even pick up the sand and put it in the pants of your best friend. right? Yes, but it needs to be monitored by adults. Part of the reason these data are valuable is that they are related to real people. The government requires everyone to pay attention to the use of these data, not to mention ethical conduct. Currently, more than 80 countries have data privacy laws. The EU has set seven “safe harbor privacy principles†to protect the personal data of EU citizens. In Singapore, the Personal Data Protection Act came into effect in January 2013. In the United States, the Sarbanes-Oxley Act requires all public companies to be publicly notified, and the Health Insurance Privacy and Accountability Act (HIPAA) sets the national standard for medical privacy. Therefore, before diving into the ocean of big data, it is important to take seriously the need to comply with governance and privacy standards. Is your institution a medical institution that needs to comply with HIPAA? Still operating in certain parts of the world? Or do you realize that it is wise to take precautions against the key elements of big data? Then, if your organization must be compliant, an obvious solution is master data management, which can tighten data usage across the organization. If you have master data management, then everything is resolved. However, agreeing on definitions and business rules is a slow and painful process for most involved. It may be painful, but it works. Forrester Consulting recommends in the report "Big Data Needs Flexible Information and Integrated Governance" not to skip governance for flexibility and speed. Big data results need governance. Forrester recommends not following a “single set of standards, policies, and practices,†which they found would “kill the value that can be achieved from big data investments and insights.†Conversely, the report recommends governance to establish a governance “zone†where the analytical capabilities are matched to the objectives, taking into account the source and type of data, and testing before formal implementation of the rules. 7, start working The last piece of advice may be the most important: Despite letting go. Just take action and follow the previous six steps. Big data is already at the door even if it has not yet entered the door. Get results now. Once you have something to show, others will pay attention because there is nothing to get attention like the result. Then a virtuous cycle is created, and the results are quickly disseminated throughout the organization. In the end, there will be an executive who is interested, so you make a big contribution. Insulated Terminals,Terminals,High-quality insulated terminals Taixing Longyi Terminals Co.,Ltd. , https://www.txlyterminals.com