Big data is changing the way you think

Since 2013, EMC has restructured its organizational structure to form a business portfolio of EMC II, VMware, and pivotal troika. Cloud computing and large data have become the most important part of EMC's corporate strategy. Dr. Patricia Florissi, vice president of EMC and global chief technology officer for sales, says cloud computing is changing it consumption patterns, and big data is changing the way people work, live and think.

Big data is not a new bottle of old wine

As early as cloud computing was emerging, the industry was a transformative innovation for cloud computing or a gimmick packaged in new bottles of old wine. Today, with the gradual fall of cloud computing, the understanding of cloud computing is slowly beginning to clear. The advent of the big data age has also been dismissed as a repackaging of business intelligence and data analysis, but a cliché for new bottles of old wine. Obviously, it vendors do not agree with this view.

"It seems to me that cloud computing is changing it consumption patterns, which represents the consumption of IT technology," says Dr Patricia Florissi, vice president of EMC and global chief technology officer of sales. The big data is changing the way we work, live and think, not new bottles of old wine. Big data is actually more than just big data, it represents three trends: first, massive amounts of data are the way we look at things, different insights about the whole thing, and secondly, because of cloud computing, we have more storage and processing power to analyze large data; People have reached the level of analysis for large data. ”

Dr. Patricia Florissi that any technology requires sufficient conditions and time to work: "If we don't have enough storage capacity, we can only delete the data, if we have enough storage capacity, but no data analysis capabilities, Then the data has no value, just like gold ore in the immediate, but not the ability to refine into gold; if we have the ability to store and process data, but do not have the ability to interpret and analyze data, then we need to wait a while to meet the arrival of the big data age, It's like the invention of printing in the past has been a real business for generations. These three conditions are ripe for big data. ”

Amazon and Google's big data services are not necessarily better

The debate about the merits of a public cloud versus a private cloud is quite lively at the very beginning of the concept of cloud computing. We can also see that the public cloud services provided by Amazon, Google and Rackspace are now being used by more and more users. In the big Data age, Amazon and Google also launched a similar public cloud of large data Analysis Services. This has left users with a similar problem with the cloud computing era: building a large data analysis environment, or capturing analytics through Amazon or Google? In this regard, Dr. Patricia Florissi said that users need to consider options in terms of ease of use, cost, and use of purposes.

"First, we need to understand how easy it is for Amazon or Google to provide large data services, and how it differs from an enterprise-level environment," Dr. Patricia-Florissi. If it is a small and medium sized company, it will be easier to use Amazon's API to store data on Amazon. One of EMC's current trends is to provide users with the same ease-of-use, programmability, and Easy-to-use interface they had previously acquired with Amazon and Google. In this context, EMC is helping enterprise users make their data center virtualization, that is, through software-defined data centers, software-defined storage that gives users the same ease of use and programmability as Amazon and Google, and second, Using Amazon's or Google's Big Data services can be a gain, and Amazon and Google services are easier and more user-friendly, but using them means you'll lose control, while using your own data center to control, but not in terms of usability, EMC wants to achieve a balance between ease of use, programmability and control in an enterprise environment; Finally, from a cost perspective, if you have an economic assessment of the frequency of data processing, you will find that if the data processing frequency is relatively high, the data processing capacity and capacity requirements are relatively large, When data usage is over 50%, there is an economic tipping point and turning point, and using your own data center is cheaper than Amazon or Google. Of course, Amazon and Google are also easier to handle with data spikes and unforeseen problems. ”

According to Dr. Patricia Florissi, the core of EMC II is focused on large data management, and pivotal is about large data analysis.

Data science is a discipline of a team

With the rise of big data, some occupations related to big data are also starting to pop up, including Chief Data Officer (CDO) and a variety of big data analysis talent. Dr. Patricia Florissi, vice president of EMC and global chief technology officer for sales, said that large data derived from data science should be a team-work discipline.

"Data science actually includes different professionals, mainly large data scientists, who mainly build different data models, visualize the abstract data, and need the ability to analyze the interrelationships between different groups of variables," says Dr Patricia Florissi. The second category is the business scientist, This kind of talent for the industry characteristics and business links are very understanding; The third category is computer scientists, who specialize in designing infrastructure, managing and operating large data models designed by large data scientists on an infrastructure basis, as required by business scientists. Chief Data Officer CDO is actually the establishment and operation of data analysis in this sector, CDO is a large data expert and computer expert interface, because the CDO needs to manage large data experts, large data experts need to constantly interact with computer experts and cooperation, the middle of the coordination work is done by the CDO. ”

Software definition: Disrupting the traditional IT market

Software Definition Data center, is considered to be another major impact on the direction of IT development, derived from the software definition of storage, software definition network, software definition calculation, etc. are gradually affecting the IT industry. This trend is also very good understanding, in the IT hardware gradually towards homogenization of today, software is the key to differentiated competition.

"Software-defined data centers will disrupt the concept of existing data centers, including storage, networking, and computing, and this will be a huge it change," says Dr Patricia Florissi, vice president of EMC and global chief technology officer of sales. In fact, the essence of the software definition is the intelligent from the hardware to the software, through the software to achieve the different needs of the data center, which means that the software definition allows you to have a very flexible, smart way to manage your hardware. In the storage area, we see that software-defined storage divides control and data management, and hardware becomes less important in terms of control, and in data management, because specialized data requires very fast speed to store data, the need for hardware specialization will become increasingly high. ”

Now, with open source software projects like Hadoop, Openstack, and Cloudstack, open source has become an indispensable force in software-defined data centers. What about EMC's approach to open source software projects? "EMC fully supports and embraces Open-source technologies like OpenStack, Cloudstack, and integrates all open source technologies with EMC products," says Dr Patricia Florissi. What EMC pursues is that if an open source technology is adopted, the performance on the EMC platform will be the best. ”

Of course, Dr. Patricia Florissi still insists that enterprise-class IT vendors have the inherent advantage of not open source software platforms, and that the non-open source software platform of enterprise-class IT vendors is not comparable to other open source platforms in terms of function and time leadership. "It is the responsibility of enterprise-class vendors to bring innovation and newer versions to market faster than OpenStack, and to lead open source projects in powerful and time," says Dr Patricia Florissi. ”

Finally, Dr. Patricia Florissi a joke to illustrate her view of open source: "There was an old man and a boy in Africa, both of whom saw a tiger, and the boy quickly wore his running shoes, and the old man said why you wore running shoes, and you couldn't outrun the tiger in your running shoes. But the boy said, I don't care if I can run through the tiger, I only care if I can run through you. ”

(Responsible editor: Lu Guang)