文档结构  
翻译进度:56%     翻译赏金:0 元 (?)    ¥ 我要打赏

2017年已经到来,大数据似乎也在突飞猛进地成长。 无论是物联网的发展还是云计算更复杂的方面,企业技术都处在上升期,促进了巨大的变革。

许多公司都把大数据作为最新的时尚,将其作为这个竞争激烈的时代的主要优势。 在这篇文章中,我们将讨论Oracle关于大数据及其未来的一些预测

1.拥抱机器学习的时代

机器学习以前只限于数据科学家,但在2017年它将公开化。 无论是谷歌的最新排名算法还是卓越的电子产品,机器学习将找到一个立足点。 大数据在2016年占有率相当大,预计在机器学习的帮助下今年将有更大的增长空间。

第 1 段(可获 1.86 积分)

无论是业务分析师还是从后端获益的一系列工具,机器学习将进军大数据的一些原本单调的领域。 这将改变政府和企业在物理和虚拟服务器之间处理数据集的方式。 前瞻性的变革领域将包括医疗保健自动化和能源。

云数据内聚

大数据一直被认为能够很好地响应基于云的服务器,但在2017年将扩大其覆盖面。 无论是关于云的使用还是数据主权的隐私问题,都将有望改善。由于更大的数据集,大多数企业可能会因为与重定位相关的歧义而转移到虚拟服务器。

第 2 段(可获 1.39 积分)

比起将数据迁移到云,将云带入数据是在2017年发生的前瞻性的变化。 特定于数据需求的云策略将是至关重要的。

3.数据驱动的应用程序

大数据技术以前因信息技术领域的影响而闻名。 然而,最近的趋势保证了大量分析甚至企业应用的更高采用率。 无论是大量的AI驱动的应用程序或流媒体客户端(如Megabox),每个企业都将很快实现大数据转变 - 以及未来的应用。

 

第 3 段(可获 1.19 积分)

4.物联网及其集成

物联网由于接二连三的荒谬的产品设计而受到了很多批评。 就如同我们在物联网中缺乏创新一样,基于高端的直觉,大数据可能会使其复苏。 无论是以移动为中心的应用还是家用小工具,物联网与大数据的配对预计将是2017年的一个革命性的一步。

物联网应用程序开发将会更加简单,并且即使在远处也会感觉到其影响(或者说,涟漪)。 我们正在寻找智能城市,甚至更聪明的全国性项目。

第 4 段(可获 1.2 积分)

5. 数据虚拟化: 虚拟现实

随着企业的扩张,数据的剧增变得寻常。使用类似NoSQL, Spark 甚至Hadoop的技术将使数据在2017肯定会得到剧增。 必须知晓的是暗数据通常难以获取类似的完美资源。因为组织难以识别。统一访问,难以捉摸的实体,将使数据虚拟化在2017得到飞速的飞跃。

因为数据移动不再那么必要,这种方法将提供大数据分析和应用。

第 5 段(可获 1.21 积分)

6. Working With Kafka

Big Data predictions feel incomplete with the mention of Kafka, a technology put forth by Apache. While Kafka is already growing in leaps and bounds, it might just peak by the third quarter of 2017. To be exact, Kafka is expected to be the much-awaited runway for the Big Data technology.

Otherwise a bus-styled technology, in terms of architecture, Kafka can easily handle data structures and even myriad data sets — focusing largely on the data lake and its proliferation and facilitating subscriber access.

7. Boom in Cloud Data Systems (Prepackaged and Integrated)

第 6 段(可获 1.2 积分)

Building a conventional data lab is difficult and that too from the scratch. However, organizations are increasingly becoming reliant on Big Data, facilitating the growth of integrated cloud data systems. These are pre-packaged entities including data science, analytics, data wrangling, and even the complexities of data integration.

2017 will witness a steady growth in the adoption of pre-packaged cloud systems dedicated to Big Data reservoirs.

8. An Alternate to the Hadoop HDFS

Hadoop’s HDFS has long been the most sought-after data accommodation platform, but object stores are expected to trump the same in 2017. The reasons for the same are better data replication, availability, and backup.

第 7 段(可获 1.34 积分)

Moreover, feasibility is a bonus when Object Stores are concerned. These are repositories to Big Data based on the same data-tier technology as the HDFS.

9. Deep Learning Even at the Cloud Level

As mentioned, data virtualization will now be easier sans added layers. This approach will, therefore, boost a host of acceleration technologies including NVMe and even GPUs. In 2017, we will also get to see deep learning joining hands with Big Data metrics. Visible results will include nonblocking, high-capacity, improved I/O, and even better network performances.

第 8 段(可获 1.11 积分)

10. Hadoop Turns Vital

Users and companies looking to leverage Big Data were using Hadoop sparingly but in 2017 we might see multi-level deployment in every possible, Data-centric project. Hadoop security will come across as a non-optional entity and would require possible applications— in every field.

Bottom Line

Big Data is on a rampage and the growth scale is absolutely second to none. However, with the emergence of IoT and even social media, snappier Big Data applications have received overwhelming responses.

In 2017, we will surely be seeing a host of informed predictions, lower costs, and even business-centric gains, courtesy of the global adoption of Big Data and associated technologies.

第 9 段(可获 1.38 积分)

文章评论