scalar brocade toronto roadshow 2013

Eric Ng VP, Sales | scalar

Scott McFarland Director | brocade

Kyle Turner Director | brocade

Simplify virtualization with

Brocade Ethernet fabrics April 24 2013

Is This Your Datacentre ?

© 2013 Scalar Decisions Inc. - Not for distribution outside intended audience.

The BIG Problems Today


Inflexible

Infrastructure

Technology

Complexity

Social & Mobility

Challenges

“Always On”

Global Marketplace

Scalar Decisions Toronto (HQ) · Vancouver · Calgary · Ottawa · London

• National Systems Integrator

• Deliver enterprise class, data centre solutions

• Strong technical focus in people

• Deep vendor technical experience

• $110M+ revenue

• 115 employees across Canada


Kyle Turner

Director, Principal Systems Engineering – Americas

April 24, 2013

Simplifying Virtual Infrastructures: Ethernet Fabrics & IP Storage

Legal Disclaimer

All or some of the products detailed in this presentation may still be under development and certain specifications, including but not limited to, release dates, prices, and product features, may change. The products may not function as intended and a production version of the products may never be released. Even if a production version is released, it may be materially different from the pre-release version discussed in this presentation.

NOTHING IN THIS PRESENTATION SHALL BE DEEMED TO CREATE A WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, STATUTORY OR OTHERWISE, INCLUDING BUT NOT LIMITED TO, ANY IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NONINFRINGEMENT OF THIRD-PARTY RIGHTS WITH RESPECT TO ANY PRODUCTS AND SERVICES REFERENCED HEREIN.

Brocade, the B-wing symbol, BigIron, DCX, Fabric OS, FastIron, IronView, NetIron, SAN Health, ServerIron, and TurboIron are registered trademarks, and Brocade Assurance, DCFM, Extraordinary Networks, and Brocade NET Health are trademarks of Brocade 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.

© 2012 Brocade Communications Systems, Inc. CONFIDENTIAL

Simplifying Virtual Infrastructures

Developing Data Center Trends

Technology Alignment

Why Ethernet Fabrics? Why Now?

NAS Optimization

VM Integration

DISCUSSION TOPICS


ENTERPRISE SCALABILITY, PERFORMANCE, AND RELIABILITY

The Evolution of Network Attached Storage (NAS)

15

144

1

384,000

9,000+

- Petabytes, Proven Enterprise Scale

- File System

- Nodes in an EMC Isilon Cluster

- VMs in a single VCS Device

- 10 GbE Ports in a VCS Cluster


Disruptive Data Center Trends BROCADE’S ETHERNET FABRIC SOLUTION FOCUS

SOFTWARE- DEFINED

NETWORKING

UNSTRUCTURED DATA GROWTH

1GE-TO-10GE TRANSITION

VM PROLIFERATION

SCALE-OUT / NAS FLASH STORAGE

File-based storage

growing at record pace.

Network virtualization

and tunneling.

VM mobility and MAC

addressing.

Driving storage growth &

higher network speeds.

Networks evolving to

higher speeds.


THE ECOSYSTEM

Common Data Center Strategies

Many devices managed as one

Master-less distributed control

plane

Separation of physical & logical

infrastructure

Machine-to-machine

communication


Brocade Virtual Cluster Switching (VCS)

Mature Ethernet Fabric with

>1,200 customers

Topology & protocol

independence

L1, 2 & 3 Multipathing

Auto-healing, non-disruptive

Lossless, low latency NAS iSCSI FCoE

ETHERNET

FABRIC

DISTRIBUTED

INTELLIGENCE

STORAGE

OPTIMIZATION

LOGICAL

CHASSIS


L3: Fabric load balancing across multiple L3 gateways

FEATURE

Multi-pathing at Multiple Network Layers PROVIDES PREDICTABLE PERFORMANCE FOR NAS

L1: Trunking with frame striping

33

%

33

%

33

% L2: Equal Cost Multi-

Pathing (ECMP)

Improved scalability and resiliency

BENEFIT

Near-perfect load balancing across all links in a trunk group

All links utilized with flow-based load balancing

Layer 3 Core

Brocade VCS Fabric

NAS


SIMPLE 10-NODE FABRIC

Ethernet Fabric Set-up

Brocade VCS, 10 x VDX8770: Step 1: Turn switches on

Step 2: Configure RBridge IDs

Step 3: Connect cables

Step 4: Next Project

Cisco FabricPath, 10 x Nexus 7K: Step 1: Create portchannels between pair Step 2: Activate vPC between pair Step 3: Activate FabricPath on the vPC pair

Step 4: Activate vPC+ in the pair (~7 commands per switch) […] Step 19: Config MAC learning for FP VLANs Step 20: Activate FabricPath in portchannel Step 21: Repeat previous 20 steps […] Step 101: Configure portchannels between edge vPC

pairs and core vPC pair […] Step 106: Activate FabricPath on 8 defined

portchannels […] Step 114: On each Nexus 7K in the core vPC, define 8

portchannels […]


BANDWIDTH, FAILOVER TIME, PLUG & PLAY

How is an Ethernet Fabric Different?

NAS_2 NAS_1

Scalability Svc Ability

Scale Out NAS

2

1

3

2

3

1

1. Half the compute bandwidth via NIC Teaming

2. Multi-second failover, half the bandwidth, manual config

3. Segregated storage via scale-up NAS

1. Twice the compute bandwidth via LAG

2. Sub-second failover, dynamic bandwidth, self-forming

3. Universal access to scale-out NAS



Master-less distributed control &

forwarding plane

API-to-ESX for profile auto-config

Network-wide knowledge of all

members, devices, and VMs

No need for compute licensing,

Fabric-enabled Automatic

Migration of Port Profiles (AMPP)

ETHERNET

FABRIC

DISTRIBUTED

INTELLIGENCE

STORAGE

OPTIMIZATION LOGICAL CHASSIS

NAS iSCSI FCoE © 2012 Brocade Communications Systems, Inc. CONFIDENTIAL

BELT & SUSPENDERS STRATEGY – COMPLIMENTARY

Investment Protection with Overlay Networking

Hypervisor Based Networking - VXLAN, NVGRE, STT

Visibility

– Compliance, Congestion Avoidance

Accounting

– Billing, Capacity Planning

Termination

– Gateway to Non-virtualized Devices



Aligned with Scale-out NAS

Hitless Bandwidth Allocation

Automated QoS Configuration

Head-of-line Blocking Mitigation

Traffic Engineering via SDN

ETHERNET

FABRIC

DISTRIBUTED

INTELLIGENCE

STORAGE

OPTIMIZATION LOGICAL CHASSIS


ITS SIMPLE AND AUTOMATIC

Why QoS is Imperative to NAS?

VM Proliferation: Density increasing, 4 to 128 per server

Performance: Compute-based networking, 10GbE per adapter

Cost: High density, high speed interfaces are cost prohibitive

Congestion: East-to-West Traffic is growing

NAS Growth: NAS is growing @ 7.5% CAGR over the next 5 years

Source: 2012 IDC WW Enterprise Storage Systems Forecast 2012-2016



Management Cluster

– Single Point IP Provisioning

– Multi-Point Granular Accounting

Growth & Serviceability

Self Provisioning via APIs

SDN Enabled, Fabric Aligned

ETHERNET

FABRIC

DISTRIBUTED

INTELLIGENCE

STORAGE

OPTIMIZATION

LOGICAL

CHASSIS


Brocade & EMC Isilon Scale-Out NAS Architecture ETHERNET FABRIC OPTIMIZED

OneFS Operating Environment

Intra-cluster Communication Layer

Servers

Client/Application Layer Ethernet Fabric

Servers

Servers

Sin

gle

FS

/Vo

lum

e

CIFS NFS

FTP HTTP

HDFS

for

Hadoop


BETTER TOGETHER

Scale-Out Innovation

Simple to Scale Manage 15+ PBs like 1 TB drive

Manage 24+ switches as logical chassis

Predictable Performance Grows linearly

Add capacity non-disruptively

Efficient to Operate Max. utilization 80%+, Automated tiering

All links active, adv. L1/2 load balancing

Available Always Survives multiple failures

Self-healing, always on

24 x 10G

--------------

1 Node

360 x 10G

--------------

10 Nodes

2304 x 10G

--------------

50 Nodes


Requirement #1 High Availability

Solution Active-Active 10GE

Requirement #2 Storage Optimized

Solution Frame-based Trunking

Requirement #3 Ease of Use

Solution Simple Plug & Play

HPC SUPER COMPUTING CLUSTER

One of the World’s Most Important Research Centers

“We couldn’t believe how easy

it was to configure the fabric, it

was a piece of cake.”

12 – Brocade VDX 6730s

4 – EMC Isilon Arrays

536 – HP Servers w/ 10G Mezz Cards


INNOVATION FEEDBACK

The Future of NAS

Are you converging your IP and storage networks into one shared infrastructure?

How important is QoS for your IP storage traffic?

What challenges exist in multi-tenant environments with a single shared NAS?

Are storage node failovers a problem with your NAS? In some environments, front-end network congestion can result in node failovers. The failovers can take >30 seconds to detect and often result in a node reboot.

Are you looking at any alternative storage like SSD, flash, etc…?

Are you using OpenStack for management?

Are you using VXLAN for Network Virtualization?

Are you using OpenFlow for Programmatic Control?


Thank You!

“Brocade should be considered for the shortlists of all data center network infrastructures and large cloud providers.” Gartner