Big data brings it big challenge security consolidation is the key

As social networks rise, mobile interconnection develops rapidly, data sensor applications and cloud computing bring about a new round of data concentration, and there has been an explosion of "data" in this era. According to statistics, The number of Internet users in 2011 has reached 350 million, which is expected to break 500 million this year; there are about 1 trillion networked devices around the world; over the past decade, global servers have increased twice times, global servers have reached 31.9 million, and 2012 digital data has reached 2.7. ZB, 48% more than 2011 ... Today, the digital universe is not only large, but also diverse, fast growth, the "Big Data" era has come.

"There's gold in the data."

After cloud computing, big Data has become the forefront of the era of the topic. More and more companies are starting to study large data, and traditional it vendors are beginning to focus on the big data field, trying to occupy the opportunity in this new chance. So what is big data and why is it so attractive? Although there is no standard and textbook definition for large data in the industry at present. But the general description of large data is roughly as follows: Massive data, a lot of growth in transaction data, data that needs to be stored and supervised, new data sources for explosions, and so on. For large data in the "big" understanding, there are mainly two aspects, the first is a large number of rapid growth of data, the second is the data contained in the great value and insight.

The 2nd general perception of Big data has made big data hot, with "gold in the data", which makes big data tempting. Making full use of large data can help global personal positioning service providers add $100 billion trillion in revenue and help Europe's public sector manage to raise the value of $250 billion a year, the McKinsey Global Research Institute said in a May 2011 issue of "Big Data: innovation, competition and productivity" in the next frontier area. Helping America's healthcare industry boost its output by $300 billion a year, and help us retail industry gain more than 60% of its net profit growth. McKinsey argues that data is becoming an important factor in production, and the use of massive amounts of data will herald a new wave of productivity growth and a wave of consumer surpluses.

Digging out useful information from large data can help companies make more accurate decisions, analysts believe that corporate decision-making should be more dependent on data than experience, although large data unstructured data to account for the majority, but it contains a large number of customer behavior and business operations and other important information, can create huge business value. The first interest in big data was the financial sector, then, retail, telecommunications, real estate and other industries have also released the use of large numbers of successful cases, such as an international retail giant through consumer buying behavior analysis, successful increase in sales of goods, as well as the real estate industry, from the search engine feedback consumers on the housing market search results, Can even predict the sale of real estate.

Big data is a double-edged sword

Everything has two sides, big data is no exception.

While large data offers tremendous opportunities for businesses and it vendors, it also poses unprecedented challenges to the enterprise, particularly for the IT infrastructure. The rapid growth of data makes the enterprise IT infrastructure larger and harder to manage, with information difficult to consolidate and share, and soaring energy consumption. According to statistics, the number of servers has increased by twice times over the past decade, the number of virtual machines has increased by 42%, server energy consumption has increased by twice times in the past 5 years, and since 2000, security vulnerabilities have increased by 8 times times. Therefore, before enjoying the cookies brought by large data, enterprises must first deal with the problem of IT infrastructure, otherwise there will be obstacles in the way of exploring big data.

While there is great value in large data, companies want to benefit from large data, first data to be "big", in other words, at present, the big data is not suitable for every enterprise and each industry, it is more suitable for those who have a large number of fast-growing data enterprises. Like in the sand found in gold, the sand is big, the chance to find gold is even greater, if deliberately to look for gold from the sand, gold may not have, even if found, the cost of human resources may be greater than the value of gold. More typically, companies with large amounts of data, such as finance, retailing, telecoms, real estate, and large e-commerce companies, may benefit from large data earlier, and the challenges of these enterprise IT infrastructures emerge earlier.

For example, in the telecommunications industry, with the rapid development of mobile Internet, Telecom needs to rapidly expand IT infrastructure to support new business and increasing load, which leads to the spread of servers, the rapid reduction of room space, the cost of energy consumption even to achieve half of the operating costs, as well as data Center server architecture, configuration Operating system diversity, operational difficulties continue to improve. In the health industry, the IT infrastructure supports the storage and querying of text images such as electronic medical records and health files, as the electronic case is to track the patient's life, the underlying data grows at an annual rate of 30%, and more and more cross-regional referral consultations require a large concentration of data to provide a unified view of the data.

Data explosion has brought about a huge IT infrastructure, a shortage of room space, soaring energy consumption, complex management and soaring operational costs, and consolidation of IT infrastructure is a good way to reduce it costs. Integration can improve server utilization, reduce energy consumption and management complexity, easier to achieve the unified deployment of resources, more efficient implementation of large data storage, classification, analysis and mining work, so it can be said that integration is a basis for applying large data. But there are a variety of business applications, and it infrastructures are different, and we often see more cases as good benefits of consolidation, but consolidation is not as simple as it seems.

Consolidation allows data to become concentrated from decentralization, making the data "big", also means more risk than distributed, which involves two considerations: first, at the IT infrastructure level, consolidation makes servers, storage, and other devices less, and a single point of failure would be much more serious than a distributed deployment. Therefore, the integration of the choice of it facilities in the security, reliability is much higher than distributed; second, in the data security perspective, although the data to focus on the protection of a simpler, but also become more attractive, once the data is invaded, the loss suffered much more, so in the data integration, The security of the infrastructure is one of the primary considerations.

Consolidation of the pain point security is the key

Today, the effective way to achieve IT infrastructure consolidation is to implement virtualization, which has changed the application mode of "single Machine Alone", can realize application centralization, improve the utilization of system resources, effectively reduce the number and space occupancy of servers, and reduce operating costs such as energy consumption and refrigeration. But there are hundreds of applications running in the room, how to realize virtualization is also a difficult problem for users with complex system platforms; In addition, the virtualization of this "put all the eggs into a basket" approach has a natural security risks, but also to many operators running a large number of data users worry.

The head of informatization Department of Telecommunication Branch of China Telecom has introduced its data center in an interview: "Large and small applications are hundreds, distributed on dozens of servers; These servers purchase time is different, platform and configuration is not the same, some are minicomputer, some are x86 server; Some Unix, some Linux, and some servers have been virtualized, how to use these different platforms, different operating systems to integrate the application is a difficult problem. "In the traditional application model, each application is a separate purchase server, because the application of different, resulting in different platforms, which caused a lot of" chimney-type "information islands, resources difficult to share, even with the use of virtualization integration, to be so many different platforms to integrate to a few servers, the hardware system, Virtualization systems have high capabilities and security requirements.

The person in charge said that there are some more critical applications in these applications, if the implementation of virtualization, to ensure the continuity of application, the reliability and security requirements of the virtualization platform, for the virtualization of this will "put eggs into a basket" application method, has a great challenge.

Retail and catering industry is also a typical business involving a large number of data, according to consumer behavior analysis, can analyze the consumer's purchase or consumption preferences, so as to help enterprises to make more correct business strategy. Wal-Mart, for example, was one of the first retail companies to make large data analyses, successfully boosting sales, and Tesco, which had a large data analysis, drove supermarkets to the subway wall, and Coca-Cola used large data analysis to discover consumer preferences to develop new products. Retail and catering have some common features: with the rapid growth of business data and the large number of stores, the traditional information mode is often a distributed "silo" structure, which is difficult to share and manage; If centralized IT deployment mode is adopted, it can greatly reduce the difficulty of information management, realize resource sharing and reduce operating cost, And more conducive to the use of large data.

I have interviewed two well-known domestic food and beverage and retail industry in charge of ISVs, the two companies in the catering and retail industry has many years of information implementation experience. The two principals have a more consistent view: Although centralized it deployments reduce management difficulties and help customers reduce the cost of informatization, the IT infrastructure for consolidation is very high in reliability and security due to the business involved in multiple stores, if business is interrupted, or data is lost or corrupted, will have a direct impact on the consumer experience, which will be a huge loss for customers. Therefore, their choice of integrated platform, security and reliability in the first place, then the performance and scalability.

An information person responsible for a provincial government tax is also confronted with the difficulties of infrastructure integration, the director said that in the current tax industry informatization advancement, although each individual system business function has been very perfect, but because the application system information does not connect with each other, in addition some application system also has the degree different function overlapping and so on, This brings many difficulties and inconveniences to the management work. At the same time, with the comprehensive development of the national Provincial data, it also puts forward higher standards and requirements for the information management and the IT infrastructure integration of the national tax and local revenue units. Therefore, the key to building the future IT infrastructure in the tax industry is consolidation and simplification.

"Our requirement for consolidation is that the device is highly secure, to minimize the risk of data storage", the same information in charge of integration concerns or security, for tax such users, integrated business system can not be interrupted, at the same time the large concentration of data to face more security risks, This is an important point to consider. The official said that the land tax information integration required IT infrastructure, security and reliability is the first factor to consider.

To sum up, "Big data" brings, there are opportunities and challenges. Indeed, there is gold in the data, the effective mining analysis of large data, can help enterprises gain more insight, make more correct decisions, and thus occupy the initiative, which is the largest data contains the greatest charm. At the same time, large data brings with it a huge challenge to the IT infrastructure, which leads to a more pressing need for consolidation, reducing the cost of IT operations and making it more productive, innovative, and not operational. On the other hand, consolidation also poses a security risk to the data, so that in the large data age, a more secure and reliable IT infrastructure can be more favored by customers, which is one of the ways it vendors need to work together.

(Responsible editor: The good of the Legacy)