Cisco Blogs
Share

Powering the Next Generation Big Data Architecture for Data-in-Motion and Data-at-Rest

- August 1, 2016 - 3 Comments

Every enterprise is becoming a data business. Data is the lifeline that guides intelligent decision making, enabling enterprises to effectively serve their customers. The rise of data has led to the modernization of data infrastructure, with Apache Hadoop as a critical foundational element for data storage and processing. Designed as a multi-workload platform, Apache Hadoop, along with related Apache projects, enables real-time insight, robust interactive analysis, and deep data mining.

In a connected world of Internet of Things (IoT), social networking, and business applications, the capability to capture, monitor, and rapidly process information is becoming essential for modern enterprises. A new model has emerged, the Lambda Architecture, for storing and processing large amounts of data-in-motion and data-at-rest. In many cases, it includes support for complex event processing with applications such as Apache Kafka and Storm, near-real-time analytics with Apache Spark Streaming, interactive SQL with Apache Hive, machine learning with Apache Spark, and data persistence and batch analytics with the Hadoop Distributed File System (HDFS) and MapReduce.

Building a next-generation big data architecture requires simplified and centralized management, high performance, and a linearly scaling infrastructure and software platform. Cisco Unified Computing System™ (Cisco UCS®) for Big Data and Analytics is a proven platform deployed across industry verticals and recognized as a leader in the space by leading analysts.

Cisco and Hortonworks have partnered to create an industry-leading solution with the Cisco UCS and Hortonworks Connected Data Platforms to deliver end-to-end capabilities for data in motion and data at rest. Hortonworks DataFlow (HDF) collects, curates, analyzes, and delivers real-time data from the IoT: sensors, smart devices, clickstreams, log files, and more. Hortonworks Data Platform (HDP), built on Apache Hadoop and Spark, enables the creation of a secure enterprise data lake and delivers the analytics you need to innovate quickly and power real-time business insights. Together, HDF and HDP empower the deployment of modern data applications for data in motion and data at rest within the Lambda Architecture framework. See Figure 1.

UCS LA

Figure 1: Lambda Architecture with Cisco UCS, HDP and HDF

I am delighted to announce the availability of the Cisco® Validated Design for Hortonworks Connected Data Platforms spanning HDP and HDF on Cisco UCS Integrated Infrastructure for Big Data and Analytics. The design represents a close collaboration between Cisco and Hortonworks, providing our joint customers with an industry-leading big data solution.

You can find the Validated Design at http://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/UCS_CVDs/Cisco_UCS_Integrated_Infrastructure_for_Big_Data_and_Analytics_with_Hortonworks_and_HDF.html.

You can find a solution overview at http://www.cisco.com/c/dam/en/us/solutions/collateral/data-center-virtualization/unified-computing/Cisco_UCS_Big_Data_Analytics_Hortonworks.pdf.

 

In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.

3 Comments

    Great article. No doubt how important is data, and the evolution of infrastructure for Big Data and Analytics. It takes a great amount of time, to develop appropriate design solutions, Thanks for sharing this information.

    For those not familiar with Cisco Validated Designs, these represent a significant amount of engineering work that has already been completed by Cisco and the application partners that ensures success. Customers implementing these designs can have confidence that the architectures will work and scale to meet the business needs today and into the future.

    • The Cisco Validated Design documents can be found at the Cisco Design Zone for Big Data at http://www.cisco.com/c/en/us/solutions/enterprise/data-center-designs-cloud-computing/bigdata.html

Share