Managing the Great Wall of Data with the Cloud


[Image courtesy of Dark Room]



By Rick Scurfield, President, NetApp APAC



Earlier this year, I shared my thoughts with Enterprise Innovation on the world's largest annual human migration and the immense amount of data involved in managing China Railway's online booking site.


Every year during the Chinese Lunar New Year, millions of Chinese workers head back to their hometowns for the holidays, in what has become known as the world’s largest annual human migration. Earlier this year, Chinese railways, roads, airlines and waterways carried a total of 2.81 billion passengers during this Spring Festival travel rush. This colossal amount of human traffic puts a great deal of stress on the country’s transportation network, particularly the railways, which saw 295 million passengers in less than two months.


In the past, the state-owned China Railway has been notorious for being unable to keep up with the immense demand of travelers during the season, repeatedly plagued with website crashes on its online booking site which in-turn has resulted in many travelers falling prey to overpriced tickets from ticket scalpers. However, this year mitigated its hosting to Alibaba’s cloud service, Aliyun, with quite incredible results. The system sold 9.56 million tickets in a day at its peak and page views reaching an average of more than 300,000 per second, a record for the site.


The Cloud Challenge


The benefits of cloud are indisputable, especially to an organization like China Railway that has to deal with huge peaks in demand for their service at key points in the year. More broadly, according to market research firm IDC, China’s public cloud market was worth US$ 717 million in 2014 and is expected to grow at 33.2% compound annual growth rate from 2015 to 2018. However, concerns over public cloud remain, especially that of data security as many organisations would still prefer to store and manage sensitive corporate information on a private cloud.  


A further example of this is Peking Hospital, which uses public cloud for unexpected surges in demand but will not use it to store patient’s personal data. This is mirrored across multiple industry sectors, for example manufacturing where Tongda Refractory Technologies Co. Ltd’s CIO, Li Bin recently said: 


“In the manufacturing industry, R&D contains highly confidential information involving the company’s core businesses, which explains why manufacturers are careful with storing related information on public cloud”.


Going Hybrid


This where hybrid cloud comes in. Hybrid cloud allows organizations to store protected or sensitive data on a private cloud that is managed in-house, while having the additional computing power of a public cloud to cope with any spikes in demand for their services. This ensures that critical business information remains secure, with the flexibility to cope with any future workload.


Wait, There’s a Catch


That being said, it is important to get deployment right. An organization’s private cloud infrastructure might differ to that of its public cloud service provider which could cause incompatibility and as a result, it finds itself unable to transfer the needed information freely.


To make hybrid cloud truly borderless, organizations need to put in place an open architecture that allows them to move information seamlessly across their choice of hybrid cloud. This Data Fabric strategy will provides them with a more unified view of enterprise data for them to store, access, protect, share and archive data in consistent and predictable ways across multiple internal and external data centers seamlessly.