1 WHITE PAPER IP Network Building Tomorrow s Data Center Network Today offers data center network solutions that provide open choice and high efficiency at a low total cost of ownership, bridging current technology with the technology of tomorrow. The state-of-the-art, fifthgeneration, network-processor-based architecture enables you to simplify network design and implement and operate a data center network that supports business demands for flexibility, scalability, and agility.
2 Introduction Organizations today face ever-increasing requirements for uptime, application flexibility in rapid application provisioning and later decommissioning/redeployment of resources--and data mobility while data growth and network traffic increase at an exponential rate. The most significant technologies developed to meet these changing business requirements are server virtualization, unified communication, cloud computing, and network convergence. At the same time, network awareness for End-to-End (E2E) data center solutions must be both manageable and provide the foundation for an increasingly agile business application infrastructure. Business agility demands increased network I/O, uptime, which in turn means constantly improving network resiliency, driving toward near-zero downtime and sub-second failover and convergence time. The network becomes more complex, for example, needing awareness of Virtual Machine (VM) mobility, application prioritization, and Quality of Service (QoS) to meet application Service-Level Agreements (SLAs). What other factors are driving decisions about network management in the data center? Doing more with less is a constant, as is the focus on minimizing environmental impact, which drives green design solutions to reduce overall energy consumption related to power and cooling. The One network architecture, described below, addresses the many and diverse challenges for data center networks. One Solution The One architecture framework, shown at a high level in Figure 1, provides convergence-ready fabrics with an emphasis on plug-and-play Virtual Machine awareness, intelligent, scalable Layer 2 architecture for increased application mobility, and E2E management for network, compute, and storage. Guiding design principles include no oversubscription, ongoing investment protection with built-in scalability for maximum resiliency and application uptime, while decreasing power consumption. Figure 1. One data center components. DCX Backbone VCS Data Center 1 Remote extension ServerIron ADX VCS Data Center 2 DCX Backbone PRIVATE CLOUD 2
3 The The Series delivers industry-leading scalability with up to: Terabits per second (Tbps) system capacity 4.8 Billion packets per second (Bpps) throughput on 256 wire-speed 10 Gigabit Ethernet (GbE) ports 32 wire-speed 100 GbE ports 1,536 wire-speed 1 GbE ports This multi-terabit performance (available in a single footprint) provides convergence-ready networks, including advanced network services, without a negative impact on performance. The future-proof architecture with 10 GbE network scaling to 100 GbE, addresses continuous operation, infrastructure simplicity, and scalability in the virtualized data center deploying the One solution framework. Multi-Chassis Trunking Series delivers Multi-Chassis Trunking (MCT), which is an enhancements to Link Aggregation Groups (s)---ieee 802.3ad (802/1ax)--providing nodal redundancy and physical and modular redundancy with sub-second recovery in case of a failed link in the group. Based on the concept of MCT peers, two Series switches act as one logical switch with access switches or servers (called MCT clients ) connected via s divided across the MCT pair. With the MCT pair appearing as a single logical switch, as shown in Figure 2, a loop-free topology is achieved; in which all links are active/active with subsecond recovery, eliminating the need for spanning tree protocols. Support for third-party devices Figure 2. Multi-Chassis Trunking (MCT). MCT logical switch Appearing as a single logical switch IEEE 802.3ad Since the MCT protocol is based on standard, the edge node (access switch or server) is transparent to the MCT protocol, which supports any third party vendor device that supports. By deploying the Series with MCT, network simplicity and higher resiliency is achieved together with full use of all network links. 3
4 Solution Examples The One architecture enables you to scale out a 10 GbE fabric and an efficient and collapsed 1 GbE architecture based on a range of options to meet your current and future needs. In the following section, the advantages of each deployment are explained in greater detail. WAN Figure 3. Data center network design portfolio. Aggregation ICL MCT VCS 1 GbE ToR FCX/CES 10 GbE ToR TI24X/ GbE ToR /48-T 10 GbE ToR VDX SAN Simplified, Scalable, and Resilient 10 GbE Fabric Architecture With the data center computing model increasingly virtualized with consolidation of VMs on hypervisors, hypervisor engines require increased network access bandwidth and resiliency to meet the application SLAs. In addition, they need CPU, memory, and disk capacity. To take full advantage of a virtualized server infrastructure the network must facilitate VM mobility from one physical server to another, handled by the hypervisor without compromising resiliency, simplicity, and scalability. Over 30 Tbps switching capacity Figure 4. Data center network design based on the Series and Virtual Cluster Switching. VCS ICL MCT Layer 3 Layer 2 Standby Active 4
5 With servers connected to the Virtual Cluster Switching (VCS) 10 GbE fabric, a large-scale multipath Layer 2 domain is formed and VM-aware architecture is achieved. All links are fully utilized within the VCS fabric and core achieving high performance, a simplified and resilient network, and enhanced application mobility. The fabric network architecture includes: Sub-200 millisecond link and node failover A network that is easy to manage and configure, no xstp configuration Plug-and-play network scalability and deployment Greater than 30 Tbps switching capacity in the core The architecture is future proof and convergence ready for Data Center Bridging (DCB)/Fibre Channel over Ethernet (FCoE) deployment and can be scaled for future growth, providing optimal investment protection and network flexibility. Efficient and Collapsed 1 GbE vtor Architecture Building out servers with 1 GbE connectivity on a massive scale combined with server virtualization and often rapid deployment can be extremely challenging for IT personnel. The virtual Top of Rack (vtor) solution is based on the Series and the 48-T module, which ensures a simple and resilient architecture. Each 48-T module supports 8 x MRJ-21 connectors (consolidates 6 x RJ45 connectors for each MRJ-21) equivalent to 48 access ports (10/100/1000) per module. Passive patch panels (meaning no power consumption) are placed in the top of each rack where the MRJ-21 cables connect and the servers are then cabled to the patch panels. with 48-T-A blade MCT2 Active-active Layer 2/3 topology with VRRP-E and MCT Over 30 Tbps switching capacity MCT1 with 48-T-A blade Layer 3 Layer 2 Figure 5. Data center network design based on the Series and 48-T-A blade. Patch panel Patch panel Standby Active This data center design with vtor has fewer components to manage, increased reliability, and simplified cabling. In addition, ease of maintenance, hitless software upgrade, and hitless failover ensures continuous network operation. Scaling to more than 1,500 connections per -32 chassis, supporting 2 million MAC addresses enables very large VM environments from the same footprint. This simple deterministic network design, with all links forwarding traffic and sub-200 millisecond link and node failover based on MCT capabilities enables simplified VM communications. The Active-Active L2/L3 topology-- utilizing VRRP-E and MCT from server to core--is easy to manage and configure and supports plug-and-play scalability 5
6 Extending ToR Stack Resiliency to the Data Center ToR solutions, such as deploying the FCX with stacking capabilities, provide several benefits. In this modular design for the data center network, oversubscription (due to fan-in ratios) is handled at the rack level, meaning fewer fiber cables and reduced cabling from access to aggregation/core. In addition, FCX stacking delivers single management of the stack and server-to-server connectivity without traffic traversing the aggregation/ core. The FCX can be horizontally stacked across eight racks or vertically within the rack. The hitless failover within the stack ensures maximum application availability. Figure 6. Data center network design based on the Series and FCX stack. Aggregation MCT2 Over 30 Tbps switching capacity Layer 3 Layer 2 MCT1 FCX IronStack 40 Gbps 1 GbE ToR 1 GbE ToR FCX Standby Active This is an optimal network design with uplink stacking cables connected to the Series devices, all links fully utilized forwarding traffic, and sub-200 millisecond link or node failover. And the same level of network flexibility and resiliency can be extended to third-party devices that support link aggregation or Ether channel. High-Performance 10 GbE Server I/O Consolidation Architecture Figure 7. Data center network design with the Series, TurboIron 24X, and Aggregation MCT2 Over 30 Tbps switching capacity Layer 3 Layer 2 MCT1 10 GbE 10 GbE ToR TurboIron 24X or GbE 10 GbE 10 GbE ToR TurboIron 24X or 8000 Standby Active 6
7 Virtualization is driving server I/O consolidation, which is in turn driven by increased pressure on server I/O channels, cable clutter, and the need to reduce the physical footprint to reduce energy consumption. Consolidating server I/O over a common transport layer helps organizations reduce their initial cost of acquisition (CapEx) by reducing the amount of server adapters, cabling, and switch ports needed for building data centers. These organizations will also be able to realize operating cost (OpEx) savings due to reduced power consumption and cooling. In addition, organizations will realize significant savings because of simplified configurations, reduced potential for human error, and more efficient troubleshooting and problem resolution. The Series 10 GbE wire-speed performance and high availability capabilities enable the highest performance in a resilient network. And with selection of the TurboIron 24X or 8000 for 10 GbE ToR server connectivity, this architecture delivers no xstp from access to core with greater than 30 Tbps switching capacity in aggregation/core, non-blocking 10 GbE, and sub-200 milliseconds link and node failover from access to core. This architecture is ready for I/O consolidation of iscsi devices with the TurboIron 24X or 10 GbE Data Center Bridging (DCB) for iscsi/fc/fcoe with the UNIFIED NETWORK MANAGEMENT Network management today is characterized by discrete tools and disparate lifecycle management processes across wired, MPLS, IP, and SAN networks in the data center. Network Advisor offers network management via a single pane of glass for the One architecture. It provides unmatched simplicity with unified management across wired and wireless IP, SAN, and converged network. Standardizing the user experience in managing different types of networks means minimal re-learning when administrators manage more than one type of network or move from one to the other. An open architecture with industry-standard APIs (SMI-S, Web Services, NETCONF, and SNMP) provides flexibility and investment protection Figure 8. Unified network management with Network Advisor. About provides innovative, end-to-end network solutions that help the world s leading organizations transition smoothly to a virtualized world where applications and information can reside anywhere. These solutions deliver the unique capabilities for a more flexible IT infrastructure with unmatched simplicity, non-stop networking, optimized applications, and investment protection. As a result, organizations in a wide range of industries can achieve their most critical business objectives with greater simplicity and a faster return on investment. For more information about products and solutions, visit 7
8 WHITE PAPER Corporate Headquarters San Jose, CA USA T: European Headquarters Geneva, Switzerland T: Asia Pacific Headquarters Singapore T: Communications Systems, Inc. All Rights Reserved. 09/10 GA-WP , the B-wing symbol, BigIron, DCFM, DCX, Fabric OS, FastIron, IronView, NetIron, SAN Health, ServerIron, TurboIron, and Wingspan are registered trademarks, and Assurance, NET Health, One, Extraordinary Networks, My, and VCS are trademarks of Communications Systems, Inc., in the United States and/or in other countries. Other brands, products, or service names mentioned are or may be trademarks or service marks of their respective owners. Notice: This document is for informational purposes only and does not set forth any warranty, expressed or implied, concerning any equipment, equipment feature, or service offered or to be offered by. reserves the right to make changes to this document at any time, without notice, and assumes no responsibility for its use. This informational document describes features that may not be currently available. Contact a sales office for information on feature and product availability. Export of technical data contained in this document may require an export license from the United States government.
WHITE PAPER campus networks Enabling the Wireless School The growing demand for Wireless LAN access in K 12 schools. Many K 12 schools are seeking new technological solutions that help optimize learning
White PAPER 10 Gigabit Ethernet Virtual Data Center Architectures Introduction Consolidation of data center resources offers an opportunity for architectural transformation based on the use of scalable,
Technical Brief: Voice over IP (VoIP) Solutions Many services, such as mobile Internet and mobile TV, require high bandwidth, and current backhaul infrastructures are not optimized to handle this traffic.
WHITE PAPER Service provider Brocade sflow for Network Traffic Monitoring Although both sflow and NetFlow enjoy widespread industry adoption, sflow is the better technology for traffic monitoring. Business
DATA SHEET www.brocade.com BROCADE 6510 SWITCH DATA CENTER Flexible, Easy-to-Use Enterprise-Class SAN Switch for Private Cloud Storage HIGHLIGHTS Provides exceptional price/performance value, combining
Whitepaper 10 Things to Know Before Deploying 10 Gigabit Ethernet Table of Contents Introduction... 3 10 Gigabit Ethernet and The Server Edge: Better Efficiency... 3 SAN versus Fibre Channel: Simpler and
September 2014 CAMPUS ENTERPRISE NETWORKS Voice over IP (VoIP) Solutions Workers today have many means for communication they use cell phones, office phones, voice-mail, Voice over IP (VoIP) fax, e-mail,
The Patient-Centered IT Network An Executive Primer The healthcare industry generates 30 percent of the world s data. Health Data Management, 2012 study Networks are at the foundation of health information
The Road to Cloud Computing How to Evolve Your Data Center LAN to Support Virtualization and Cloud Introduction Cloud computing is one of the most important topics in IT. The reason for that importance
ARISTA WHITE PAPER 10 Gigabit Ethernet: Enabling Storage Networking for Big Data Hadoop and other big data frameworks introduce a data storage model which is fundamentally different from the NAS or SAN
COMPLEXITY AND COST COMPARISON: CISCO UCS VS. IBM FLEX SYSTEM (REVISED) Not all IT architectures are created equal. Whether you are updating your existing infrastructure or building from the ground up,
Proven Infrastructure Guide EMC VSPEX PRIVATE CLOUD VMware vsphere 5.5 for up to 1,000 Virtual Machines Enabled by Microsoft Windows Server 2012 R2, EMC VNX Series, and EMC Powered Backup EMC VSPEX Abstract
White Paper How to Build a Better Cloud: Leveraging Unified, Virtualized Storage and Data Center Fabrics By Bob Laliberte, Senior Analyst, and Kerry Dolan, Research Analyst September 2012 This ESG White
Cisco UCS Solution for EMC VSPEX Microsoft Private Cloud Fast Track 4.0 Design Guide Last Updated: October 8, 2014 Building Architectures to Solve Business Problems 2 Cisco Validated Design About the Authors
S T R A T E G I C W H I T E P A P E R Cloud Ready Service Infrastructure for Communications Service Providers The emergence of cloud computing combined with the convergence of information technology (IT),
Hyper-V Live Migration over Distance Reference Architecture Guide By Hitachi Data Systems in collaboration with Microsoft, Brocade and Ciena June 2010 Summary Hitachi Data Systems, Microsoft, Brocade and
FAQ BROCADE DATA CENTER FABRIC MANAGER (DCFM) FREQUENTLY ASKED QUESTIONS Overview Brocade Data Center Fabric Manager (DCFM ) is a new solution for managing enterpriseclass data center environments. Combined
Windows Server 2008 R2 Hyper-V Live Migration Table of Contents Overview of Windows Server 2008 R2 Hyper-V Features... 3 Dynamic VM storage... 3 Enhanced Processor Support... 3 Enhanced Networking Support...
ericsson White paper Uen 284 23-3264 February 2015 Next-generation data center infrastructure MAKING HYPERSCALE AVAILABLE In the Networked Society, enterprises will need 10 times their current IT capacity
SAN Design and Best Practices Version 2.3 A high-level guide focusing on Fibre Channel Storage Area Network (SAN) design and best practices, covering planning, topologies, device sharing in routed topologies,
Virtual Machine Mobility with VMware VMotion and Cisco Data Center Interconnect Technologies What You Will Learn VMware has been the industry leader in virtualization technologies for the past decade and
Best Practices for Virtualizing and Managing SQL Server v1.0 May 2013 Best Practices for Virtualizing and Managing SQL Server 2012 1 1 Copyright Information 2013 Microsoft Corporation. All rights reserved.