Avatar

Well, we do not stand still for long…at least our R&D teams don’t!

The breadth of our solutions for Big Data and Analytics continues to grow, as we are pleased to introduce Cisco’s UCS Integrated Infrastructure for Big Data and Analytics with Cloudera and Apache Spark.  Working with one of our key Big Data ISV Partners – Cloudera – our Engineering teams have created this new Cisco Validated Design, downloadable for free here.

Apache Spark is a fast, general-purpose engine for large-scale data processing. With Spark, more enterprises are adopting Hadoop and gaining the capability to process a much wider set of workloads, including streaming and machine learning.

Two common reference architectures for Spark on Cisco UCS are available on Cisco UCS C-Series Rack Servers: one adds Spark processing on Hadoop infrastructure, and the other enables stream processing with Kafka or similar technologies.

Spark2
Figure 1:  Typical use cases for Spark and Spark Streaming. Showing flow of data from various data sources to Fog and Kafka nodes, and then to Spark, and then farther downstream to HDFS, NoSQL, SQL databases, Elastic, Solr, and other systems for additional processing.

 Solution Highlights

  • Comprehensive Integrated Infrastructure for Big Data and In-Memory Analytics
    The Cisco UCS Integrated Infrastructure for Big Data and Analytics offers high performance, capacity, and scalability for Apache Spark with Cloudera Enterprise. It offers proven, high-performance linear scalability and easy scaling of the architecture with single-and multiple-rack deployments.
  • Easy Deployment
    Cisco UCS Manager simplifies infrastructure deployment with an automated, policy-based mechanism that helps reduce configuration errors and system downtime.
  • Simplified Management
    Deploy Cisco UCS Director Express for Big Data quickly and easily for big data infrastructure with one-click provisioning, installation, and configuration. Used in combination with Cloudera Manager, a holistic interface that provides end-to-end system management and detailed and precise visibility and control over every part of an enterprise data hub, the solution makes cluster management simple and straightforward.
  • Flexible Big Data Platform
    Cisco UCS Integrated Infrastructure for Big Data and Analytics on Cloudera Enterprise allows you to deploy Apache Spark in standalone cluster mode or together with Hadoop and leading NoSQL deployments.
  • Real-time Data Processing with Spark Streaming
    By adding Spark Streaming and Apache Kafka on Hadoop deployments to Cisco UCS Integrated Infrastructure for Big Data and Analytics, you enable stream analytics, which ingest data in small batches and perform transformations on them.

To learn more about Cisco’s Big Data & Analytics Data Center program, please visit us at www.cisco.com/go/bigdata. To view our current offerings & architectures for Big Data, please visit the Cisco Design Zone for Big Data.