Use Cases: Cisco UCS S3260 Storage Server with MapR Converged Data Platform and Cloudera Enterprise

Cisco UCS S3260 Storage Server, the first product in the Cisco UCS S-Series, is a follow-on from the C3260. It delivers rapid scalability and performance to activate your data and insights in real time. What does it bring to Dense Storage for Big Data today?

In this article we will share the two use cases that tell you how Cisco UCS S3260 Storage Server bring Flexibility and Scalability to Dense Storage for Big Data.

Highlights-Cisco UCS S3260 Storage Server with MapR Converged Data Platform and Cloudera Enterprise

Comprehensive Integrated Infrastructure for Big Data

  • The Cisco UCS S3260 Storage Server offers high performance, dense storage, and scalability for big data systems. The Cisco Unified Computing System™ (Cisco UCS®) platform offers complete integration of compute, network and storage resources with unified management, providing easy, linear scalability of the architecture.

Modular Design to Protect Your Investments

  • Cisco UCS S3260 is built on a fully modular architecture. Compute, network and storage components can be upgraded independently as needed, protecting long-term investments as technology advances.

Flexibility to Handle Both High-Capacity and High-Performance Workloads

  • Configure the Cisco UCS S3260 with one server node when you need more storage capacity, or with two server nodes when you need both high storage capacity and high compute power.

Easy Deployment

  • Cisco UCS Manager simplifies infrastructure provisioning with an automated, policy-based mechanism that helps reduce configuration errors and system downtime.

Exceptional Scalability

  • With the Cisco Application Centric Infrastructure (Cisco ACI™) platform, you can easily scale a cluster to thousands of nodes. Cisco ACI implements an application-aware, policy-based approach that treats the network as a single entity rather than a collection of switches.

Simplified Management

  • Cisco UCS Manager allows for easy provisioning, storage and network configuration with a simple interface.
  • MapR Control System gives Hadoop administrators a single place for configuring, monitoring, and managing clusters.
  • Cisco UCS Manager simplifies infrastructure provisioning with an automated, policy-based mechanism that helps reduce configuration errors and system downtime.
  • Cloudera Manager is a holistic interface that provides end-to-end system management with detailed visibility and control over every part of an enterprise data hub.
Cisco and MapR Evolving the Hadoop-based Enterprise Data Hub
Cisco and Cloudera deliver solutions for powering the Enterprise Data Hub

Data is being generated at an unprecedented scale. More data is being collected more quickly and stored longer. Traditional transactional data is being supplemented with data from high-speed, real-time streaming systems and then stored for long periods of time both for archival and regulatory purposes. Sensors, Internet of Things (IoT) devices, social media, online transactions, and other sources are all generating data that needs to be efficiently captured, processed, and stored.

One of the major challenges of big data systems is managing the rapidly growing data and the corresponding increasing costs. Currently, to save costs, many businesses have to make difficult choices, based on today’s business priorities, about what data to keep and what data to throw away. Frequently, a large amount of raw data is deleted because it is too expensive to preserve it. But later, when business priorities change, to be able to make an informed decision the necessary data may not be readily available to users. This limitation hampers the ability of businesses to respond to market shifts and rapidly address the competitive environment.

The Cisco UCS S3260 Storage Server is specifically designed to address this problem. This next-generation high-density storage system provides up to 600 terabytes (TB) in only four rack units (4RU), providing the best dollar-perterabyte value while delivering superior computing performance and a balanced core-to-spindle ratio. The Cisco UCS S3260 provides superior performance at a lower cost. Fewer servers means less rack space, fewer operating system and software licenses, less networking equipment to purchase and maintain, and lower power and cooling costs.

The modular design of the Cisco UCS S3260 provides unique capabilities to meet the challenges of today’s dynamic business environment. You can easily adapt to changing business requirements by adjusting the configuration. For example, you can add more processing power without having to migrate data.

In addition, the Cisco UCS S3260 protects your long-term technology investment. The compute, network and storage components can be upgraded independently as technology advances.

The Cisco UCS S3260 Storage Server is the latest addition to the highly successful Cisco UCS® reference architecture for big data. It complements Cisco UCS Integrated Infrastructure for Big Data and Analytics, a highly scalable architecture for big data systems that includes compute, network and storage resources fully managed through Cisco UCS Manager and linearly scalable to thousands of nodes using Cisco Nexus® 9000 Series Switches and the Cisco Application Centric Infrastructure (Cisco ACI™) platform.

The MapR Converged Data Platform integrates the power of Apache Hadoop and Spark to optimize your Data Architecture powering the Enterprise Data Hub. The platform is powered by a fast, reliable, secure, and open data infrastructure.

Cisco UCS Integrated Infrastructure for Big Data and Analytics

Organizations today must help ensure that the underlying physical infrastructure can be deployed, scaled, and managed in a way that is agile enough to change as workloads and business requirements change. Cisco UCS has redefined the potential of the data center with its revolutionary approach to integrated infrastructure to meet the business needs of IT innovation and acceleration. Cisco UCS Integrated Infrastructure for Big Data and Analytics provides an end-to-end architecture for processing high volumes of real-time and archived data, both structured and unstructured. At the same time, it transparently integrates relevant complex capabilities to deliver an enterprise-class offering with high performance and scalability to suit the applications demand.

Cisco UCS 6200 and 6300 Series Fabric Interconnects

Cisco UCS 6200 and 6300 Series Fabric Interconnects provide high-bandwidth, low-latency connectivity for servers, with Cisco UCS Manager providing integrated, unified management for all connected devices. The Cisco UCS 6300 Series Fabric Interconnects are a core part of Cisco UCS, providing low-latency, lossless 10 and 40 Gigabit Ethernet, Fibre Channel over Ethernet (FCoE), and Fibre Channel functions with management capabilities for systems deployed in redundant pairs.

Cisco fabric interconnects offer the full active-active redundancy, performance, and exceptional scalability needed to support the large number of nodes that are typical in clusters serving big data applications. Cisco UCS Manager enables rapid and consistent server configuration using service profiles and automates ongoing system maintenance activities such as firmware updates across the entire cluster as a single operation. Cisco UCS Manager also offers advanced monitoring with options to raise alarms and send notifications about the health of the entire cluster.

Cisco UCS S3260 Storage Server

The Cisco UCS S3260 Storage Server is a high-density modular storage server designed to deliver efficient, industry-leading storage for data-intensive workloads. The Cisco UCS S3260 is a modular chassis with dual server nodes (two servers per chassis) and up to 60 large-form-factor (LFF) drives in a 4RU form factor (Figure 1).

The server uses dual Intel® Xeon® processor E5-2600 v4 series CPUs and supports up to 512 GB of main memory and a range of hard-disk-drive (HDD) and solid-state disk (SSD) options and up to two internal SSD drives for boot. The Cisco UCS S3260 chassis has 56 top-load LFF HDDs with a maximum capacity of 10 TB per HDD and can be mixed with up to 28 SSDs with maximum capacity of 3.2 TB per SSD. It comes with a host bus adapter (HBA) controller.

The modular Cisco UCS S3260 chassis offers flexibility with more computing, storage, and PCIe expansion on the second slot in the chassis. This second slot can be used for:

  • An additional server node
  • Four additional LFF HDDs with up to 10 TB capacity per HDD
  • New PCIe expansion tray with up to two x8 half-height, half-width PCIe slots that can use any industry-standard PCIe card including Fibre Channel and Ethernet cards

The Cisco UCS S3260 chassis includes a Cisco UCS Virtual Interface Card (VIC) 1300 platform chip onboard the system I/O controller, offering high-performance bandwidth with dual-port 40 Gigabit Ethernet and FCoE interfaces per system I/O controller

Cisco UCS S3260 Storage Server for Big Data and Analytics

MapR Converged Data Platform

As one of the technology leaders in Hadoop, MapR with the MapR Converged Data Platform, enables enterprise-class big data solutions that organizations can develop quickly and administer with ease. With significant investment in critical technologies, MapR offers one of the industry’s most comprehensive Hadoop platforms, fully optimized for performance and scalability. MapR’s distribution delivers more than a dozen tested and validated Hadoop software modules over a fortified data platform, offering exceptional ease of use, reliability, and performance for big data solutions, as shown in Figure 2.

Main features of the MapR Converged Data Platform include the following:

  • Performance: Use of the MapR file system, designed for high performance and throughput
  • Scalability: Up to a trillion files, with no restrictions on the number of nodes in a cluster
  • Standards-based APIs and tools: Standard Hadoop APIs, Open Database Connectivity (ODBC), Java Database Connectivity (JDBC), Lightweight Directory Access Protocol (LDAP), Linux Pluggable Authentication Modules (PAM), and more
  • MapR Direct Access Network File System (NFS): Random read and write operations, real-time data flows, transparent support for existing non-Java applications
  • Manageability: Advanced management console, rolling upgrades, and representational state transfer (REST) API support
  • Integrated security: Kerberos and non-Kerberos options with wire-level encryption
  • Advanced multi-tenancy: Volumes, data placement control, job placement control, queues, and more
  • Consistent snapshots: Full data protection with point-intime recovery
  • High-availability: Ubiquitous high-availability with no-NameNode architecture, YARN high-availability, and NFS high-availability
  • Disaster recovery: Cross-site replication with mirroring
  • MapR-DB: Integrated enterprise-class NoSQL database
  • MapR Streams: Event streaming on a global scale
Cloudera Enterprise

Cloudera is the leading provider of enterprise-ready, big data software and services. Cloudera provides a scalable, fexible, and integrated platform that enables any enterprise to easily manage rapidly increasing volumes and varieties of data. Industry-leading Cloudera products and solutions enable businesses to deploy and manage Apache Hadoop and related projects, manipulate and analyze data, and keep that data secure and protected (Figure 2).

Cloudera provides the following products and tools:

  • Cloudera Enterprise: Cloudera Enterprise includes the Cloudera distribution of Apache Hadoop and other related open-source projects, including Spark. Cloudera Enterprise also provides security and integration with numerous hardware and software solutions.
  • Apache Spark: An integrated part of Cloudera Enterprise, Spark is an open standard for flexible in-memory data processing for batch, real-time, and advanced analytics. Cloudera is committed to adopting Spark as the default data processing engine for analytics workloads.
  • Cloudera Manager: This sophisticated application is used to deploy, manage, monitor, and diagnose problems with Cloudera deployments. Cloudera Manager provides the Admin Console, a web-based user interface that makes administration of any enterprise data simple and straightforward. It also includes the Cloudera Manager API, which can be used to obtain cluster health information and metrics as well as to confgure Cloudera Manager.
  • Cloudera Navigator: This end-to-end data management tool for the Cloudera Enterprise platform enables administrators, data managers, and analysts to explore the large amounts of data in Hadoop. The robust auditing, data management, lineage management, and lifecycle management capabilities in Cloudera Navigator enable enterprises to meet stringent compliance and regulatory requirements.

Together, Cisco and Cloudera provide organizations with an enterprise-ready data management platform, as well as management integration with an enterprise application ecosystem. They transparently combine to provide a uniquely capable, industry-leading architectural platform for Hadoop-based applications.

Reference Architecture

Cisco UCS Integrated Infrastructure for Big Data and Analytics and Cisco UCS S3260 Storage Server together offer several configurations to meet a variety of computing and storage requirements as shown in the following Table. These configurations support the massive scalability that big data enterprise deployments demand. This architecture can scale to thousands of servers with Cisco Nexus 9000 Series Switches.

In addition, Cisco offers Integrated Infrastructure for Big Data and Analytics with Cisco UCS C240 M4 rack servers. The performance-optimized option supports 24 small-form-factor (SFF) disk drives, and the capacity-optimized option supports 12 LFF disk drives. When using Cisco UCS C240 M4 configuration, use three servers as management nodes.

For a Cisco UCS S3260 Storage Server configuration, management nodes can be three Cisco UCS C240 M4 Rack Servers with two Intel Xeon processor E5-2680 v4 CPUs, 256 GB of memory, a 12-Gbps SAS RAID controller with a 2-GB cache, twelve 1.2-TB 10,000-rpm SFF SAS drives, and Cisco UCS VIC 1387 (two 40 Gigabit Ethernet QSFP interfaces)

Cisco UCS Integrated Infrastructure for Big Data and Analytics Options

The enterprise-class Cisco UCS S3260 Storage Server extends the capabilities of Cisco UCS Integrated Infrastructure for Big Data and Analytics. The modular architecture allows you to configure the system to meet the exact application requirements and to upgrade compute, network and storage resources independent of each other. The Cisco UCS S3260 delivers an optimal combination of high availability, performance, and flexibility while protecting long-term investments and hence lower your Total Cost of Ownership (TCO).

More Related

Cisco’s New Storage Optimized UCS Server-UCS S3260

Cisco’s Next Generation Storage Networking Innovations

Why Choose Cisco Nexus 9000 Series Switches? Top Five Reasons…

What’s the Different Result with Cisco ACI?

Make the Cisco Nexus 9000 Series Your Network Switch Today

Info from https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-s3260-storage-server/ucs-s3260-mapr-sb.pdf

https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-s3260-storage-server/ucs-s-3260-cloudera-sb.pdf

Share This Post

Post Comment