首页 > 分布式系统 > Hadoop如何形成SAP Hana的大数据平台

[悬赏]Hadoop如何形成SAP Hana的大数据平台 (已翻译60%)

查看 (123次)
英文原文:How Hadoop Tools Shape SAP Hana’s Big Data Platform
标签: Hadoop
admin 发布于 2017-07-26 14:11:29 (共 5 段, 本文赏金: 12元)
参与翻译(2人): greenflute cyt5969858 默认 | 原文

【已悬赏】 赏金: 1元

自2008年起,SAP的HANA已经日渐成为领先的数据库系统之一。不仅仅使因为它能比其他数据库管理方案更有效地处理数据,更因为能够使用Hadoop提供的高端工具。

没有Hadoop,大多数的SAP Hana系统的用处将变得相对有限。很多数据将较难获得,尤其是当存储的是原始数据的时候。

greenflute
翻译于 2017-11-16 21:48:45
 

参与本段翻译用户:
greenflute

显示原文内容

【已悬赏】 赏金: 3元

为什么Hadoop是SAP Hana的支柱

Michael Cox 和 David Ellsworth 在他们1997年的文章《应用程序控制的需求期望核外可视化》  中提出了大数据这个概念。但是知道最近大数据才真正变得可以接触和理解。

问题并不在存储能力上,云计算的发展以指数级地扩展我们的数据存储能力。但是存储之后如何访问数据就是另外一个故事了。大多数数据提取工具可以从存储了若干TB的磁盘阵列中提取数据。根据数据科学中心,数据的可用性提高了109%。

大量的数据是以非格式化存储的,这样提取就比较困难了。Hadoop 就是为了解决这个问题而诞生的。

一些SAP Hana 解决方案能存储高达1.6TB的数据,但是这些数据通常使用不同的文件类型存储的,这使得用一个统一的格式来访问和管理变得很困难。有了 Hadoop 就简单多了。

greenflute
翻译于 2017-11-16 21:59:17
 

参与本段翻译用户:
greenflute

显示原文内容

【已悬赏】 赏金: 2元

SAP Hana 是如何与 Hadoop 集成的

SAP Hana 与 Hadoop 集成后可以更方便地访问远程数据集群,但是设置过程比较耗时。第一步使设置和安装集群,有如下几种方式:

  • 按需集群, 比较适合应对少于50个节点,位于特定地点的项目。
  • 基于云的集群,适合需要跨地区调度超过50个节点的情况。

确定了合适的集群类型,就需要建立一个测试环境,Cloudera Director是个比较好的工具。

经过几次模拟测试以后,就可以使用 Hadoop 去访问 SAP Hana 的智能数据了。

greenflute
翻译于 2017-11-17 00:39:40
 

参与本段翻译用户:
greenflute

显示原文内容

【待悬赏】 赏金: 5元

What Are the Benefits of Using Hadoop With SAP Hana

There are numerous reasons that SAP Hana administrators use Hadoop. Many people choose to use SAPUI5 on HANA, because it has an exceptional Hadoop infrastructure.

Cost-Effectiveness

According to Dell EMC, cost-effectiveness is one of the top reasons to integrate Hadoop and SAP Hana. The cost savings depend on the volume of data stored, regardless of whether the data is structured, unstructured of semi-structured.

“A VMAX All Flash array typically consists of a variety of storage groups, SAP HANA production and nonproduction databases, and non-SAP HANA workloads, each with its own CR. The overall system CR is therefore a mix of the various underlying storage group ratios. With a normal mix of workloads, you can expect to see an approximately 2:1 system CR. This ratio could be higher or lower depending on the workload mix. When inline compression is combined with other VMAX All Flash space-saving capabilities (such as virtual provisioning, zero space reclaim, and space-efficient snapshots), an overall efficiency rate of 4:1 is achievable.”

Fast Response Times

There is trade-off between response time, scalability and reliability. Hadoop prioritizes fast response times, so it is ideal for applications where administrators need to urgently access data. For applications where scalability is more of a concern, Hadoop may not preferable.

You will need to outline your priorities first. However, since most expediency is the priority of most SAP Hana users, Hadoop is usually their go-to solution.

Batch Processing and Mining Raw Data

Accessing raw data is difficult with more primitive big data extraction tools. Hadoop makes it much easier, which is one of the main reasons it is widely used in SAP Hana applications.



【待悬赏】 赏金: 1元

A Solid Hadoop Framework is Crucial for SAP Hana Applications

When you are setting up an SAP Hana data environment, you will almost always need to integrate it with Hadoop. Otherwise, it would be very difficult to access unstructured data sets.

共1人翻译此段 (待审批1人)


参与本段翻译用户:
cyt5969858

GMT+8, 2018-1-23 22:11 , Processed in 0.036747 second(s), 11 queries .