NETSCOUT S-Blade Pro & TestStream V4 - Spirent...LAN to manage the NETSCOUT device as well as to...

© 2016 Tolly Enterprises, LLC of 6Tolly.com

#216149November 2016

Commissioned byNETSCOUT Systems, Inc.

Executive SummaryToday’s enterprises, service providers, and government agencies are

being tasked to improve the efficiency of their network, application, and

security test labs. Sophisticated tools are required for various test

scenarios; however these tools are expensive and often in short supply in

large organizations where assets are shared.

To help optimize tools and infrastructure, the nGenius® 3900 Series Packet

Flow Switch from NETSCOUT® is a modular, chassis-based switch

architecture, that supports high-density 1GbE, 10GbE, and 40GbE.

NETSCOUT's Test Optimization product line provides specialized hardware

and software to meet requirements that are unique to test labs. NETSCOUT

S-Blade Pro is a low latency blade that works in conjunction with NETSCOUT

TestStream Management Software to provide an easy-to-use, multi-function

environment for Layer 1 switching. Additionally, S-Blade Pro provides

bandwidth-based utilization statistics for all ports.

NETSCOUT commissioned Tolly to evaluate the Layer 1 functionality of the

NETSCOUT S-Blade Pro in both 10GbE and 40GbE network topologies

evaluating the latency across a full range of RFC 2544 frame sizes as well as

with jumbo frames and utilization statistics.The S-Blade Pro delivered ultra-

low latency at every frame size tested. See Figure 1.

Test Highlights

NETSCOUT S-Blade Pro & TestStream V4Layer 1 Switching (1/10/40GbE) with Intelligent Port Utilization Statistics Evaluation

3

Ultra-low & deterministic Layer 1 latency1

4 Intuitive GUI for operational ease and rapid and repeatable topology reconfiguration

NETSCOUT S-Blade Pro & TestStream V4 provide:

Figure 1

S-Blade Pro & TestStream V4: 10GbE and 40GbE LatencyPort to Port Latency at 100% Line Rate for Various Frame Sizes

as Reported by Spirent TestCenter v4

Source: Tolly, September 2016

0

5

10

15

20

25

30

10GbE 40GbE

64 256 512 1518 9600 64 256 512 1518 9600

24

28

23

28

23

28

24

28

23

28

Aver

age

Late

ncy

(nan

osec

onds

)

2 Error-free operation & minimal jitter at all packet sizes

Utilization metrics for visibility into test tool and system-under-test port usage

Note: Traffic generated by Spirent TestCenter. Spirent resolution 10ns at GbE and 2.5ns at 40GbE. Jitter under 10ns for all tests. Tests are port to port on different blades, Layer 1 to Layer 1 switching.

Comparison of Layer 1 vs Layer 2

Latency for 10GbE

(Lower is better)

0

100

200

300

400400

28

Aver

age

Late

ncy

(nan

osec

onds

)

Layer 1NETSCOUT

Layer 2 Arista

Note: Recent Tolly results of the Arista Networks 7150 switch as compared to NETSCOUT S Blade Pro.

Figure 1a

IntroductionThe nGenius 3900 Series Packet Flow Switch

(PFS) with integrated S-Blade Pro from

NETSCOUT delivers an intelligent Layer 1

switch solution. Available in 1 and 3 slot

chassis options, the 3900 with S-Blade Pro

offers 24 QSFP ports per blade with advanced

Layer 1 switching capabilities. The S-Blade Pro

provides utilization statistics, which enable

visibility into actual usage of test tools and

infrastructure in the test lab.

NETSCOUT's TestStream Management

Software provides a simple software interface

that serves as a secure access portal, enabling

connections and connection changes

between devices to be made confidently via

its user interface or an automation system,

ensuring those connections are accurate. No

more running to the lab to troubleshoot

cabling faults or failed test sets due to ports

already in use.

Tests were run in an nGenius 3903 PFS 3-slot

chassis, configured with two S-Blade Pro

blades. The traditional Layer 1 engine

encompasses ports 1-72, while the

remaining ports (73-96), are powered by

NETSCOUT’s Layer 1 Smart Engine. These

ports allow for additional functionality, such

as monitoring utilization, or defining taps/

mirrors on both traditional Layer 1 ports and

Smart Ports.

Tests were run in both “standard” Layer 1-to-

Layer 1 port mapping as well as making use

of the “Smart” features enabled by the FPGA.

This latter part of the test benchmarked the

latency when traffic was being either

transmitted to a monitoring/utilization port

or replicated to another port. Tests were run

where both source and destination ports

were on the same blade and again when

those ports were on different blades.

LatencyLatency was measured in both 10GbE and

40GbE configurations and at frame sizes

ranging from 64-bytes to 9600-bytes.

NETSCOUT S-Blade Pro & TestStream V4 #216149


NETSCOUT SYSTEMS, INC.

S-Blade Pro & TestStream V4

Latency Performance & Replication/Utilization Functionality

Tested September

2016

Figure 2Source: Tolly, September 2016

S-Blade Pro & TestStream V4: Port Utilization Measurement Example

The most-challenging, “worst case” scenario is

when traffic leaves one blade, crosses the

backplane, and exits a port on a different

blade. Passing traffic with minimal and

consistent latency is important in test labs

where test results must be accurate and

repeatable. Consistent latency, i.e., low jitter, is

also important.

10GbE

For 10GbE, the measured latency was a mere

28ns across all frame sizes from smallest to

largest. To provide some context, Tolly

reviewed recent Tolly internal test results for

the Arista Networks 7150 switch. Those tests,

run at 10GbE only, showed cut-through

latency results for the same range of frame

sizes as 400ns, nearly 15X longer than the S-

Blade Pro. For all latency results, see Figures 1

and 1a.

40GbE

Tolly engineers ran the same set of tests using

40GbE ports. S-Blade Pro latency was 23 or

24ns.

UtilizationAs noted earlier, the presence of the “Smart”

FPGA provides additional functionality

beyond Layer 1 switching. All ports can

monitor utilization in real time. Additionally,

resources from Smart Ports can be utilized by

standard Layer 1 ports to extend utilization

measurements to any port on the blade.

Tolly engineers evaluated this capability as

well. An example of the NETSCOUT

TestStream network utilization output can be

seen in Figure 2.

ReplicationThere are instances when the traffic on a

single input port needs to be processed, in

real time, by multiple test tool ports. Tolly

engineers tested NETSCOUT’s port replication

feature where traffic was replicated to a port

on a different blade. Tests showed that at

10GbE the latency averaged 24.4ns. Tests at

40GbE showed average latency of 23.4ns.

Error-Free Operation & JitterAcross the range of tests, Tolly engineers

noted 100% error-free operation.

Additionally, jitter (i.e. variation across results)

was negligible. As noted earlier, this provides

consistent latency and is an important

consideration.

Ease-of-Use & GUIWhile not a formal part of the evaluation,

Tolly engineers noted that the NETSCOUT



S-Blade Pro & TestStream V4: Capabilities

Source: NETSCOUT, September 2016

user interface was intuitive and easy-to-use.

The interface made it easy to change the

Layer 1 port switching topology.

Test Setup &

Methodology

Test Bed SetupEngineers evaluated the NETSCOUT S-Blade

Pro blade, each containing 24 physical

40GbE ports (96 logical 10GbE ports). See

Figures 3 and 4.

The blades were housed in a NETSCOUT

nGenius 3903 Packet Flow Switch. The S-

Blade Pro is physically divided into two

modules. The traditional Layer 1 engine

encompasses ports 1-72, while the

remaining ports (73-96), are powered by

NETSCOUT’s Layer 1 Smart Engine. All ports

allow for additional functionality, such as

monitoring utilization, or defining taps/

mirrors.

All tests were configured from NETSCOUT

TestStream Management Software

v4.00.200.030.

Traffic profiles were configured and

executed using Spirent TestCenter

v4.66.8916. The Spirent chassis contained 1x

FX2-40G-Q3 and 1x DX2-100G-P4 blades,

which were used for all 10GbE and 40GbE

tests. The DX2 line card was configured with

CFP2 ports, and a breakout cable was used

to provide 10x 10GbE lanes.

All devices were connected to a LAN for

management purposes. A PC running

Microsoft Windows 7 was connected to the



NETSCOUT nGenius 3900 Series Packet Flow Switch with S-Blade Pro Blades

Figure 3Source: NETSCOUT, September 2016

Test Bed Topology

Figure 4Note: Configuration tested consisted of a 3 slot nGenius 3903 chassis with two S Blade Pro blades installed

Traffic Generator: Spirent

N11U Chassis

Source: Tolly, September 2016

System Under Test:

nGenius 3900 Series Packet Flow Switch

outfitted with S-Blade Pro blades

LAN to manage the NETSCOUT device as

well as to configure the traffic generator. See

Figure 4.

Test Permutations

Engineers designed tests to exhibit the use

cases of the S-Blade Pro and to demonstrate

its low latency across multiple test scenarios.

As such, several permutations were made to

the base test case, to allow for these

comparisons. See Table 1.

Test Methodology

Latency

Stream blocks were defined for each of the

physical connections, and the standard

RFC2544 test was used to collect

measurements. In order to accurately

identify the latencies imposed by the

NETSCOUT solution, engineers configured a

latency offset on the Spirent ports under

test, equal to the time on wire for the test

equipment cabling.

For 10GbE testing, 16 meter cables were

used to make both connections, resulting in

80ns of cable latency which was factored

out of all test results at the port level.

Likewise, the 40GbE tests were run over 5

meter QSFP cables, resulting in a 25ns cable

latency.

All tests were run for thirty seconds at each

frame size, and across 25, 50, and 100% load

levels.

Normal Mode Operation

The typical use case of the S-Blade Pro is to

manage physical Layer 1 connectivity via

logical mappings. Engineers configured

logical maps between 10GbE and 40GbE

port pairs, in “On-blade” and “Off-blade”

scenarios, generating traffic profiles to stress

the physical connection made.

These scenarios were also tested with one or

both links connected to Layer 1 Smart Ports.

Utilization Mode Operation

One feature of the Layer 1 Smart Ports is the

live monitoring of port utilization and

statistics via Port Scanners. Engineers

enabled this functionality, and configured a

Port Scanner to monitor utilization on four

ports under test, then re-ran the Layer 1 <-->

Layer 1 test, noting that the latency did not

increase, and measurements were within

Spirent’s own tolerances for latency

resolution. Spirent notes: “The test modules

have a resolution of 10ns at 10Gbps and

2.5ns at 40Gbps and 100Gbps. To yield the

best results when measuring latency

between interfaces of different speeds, the

test signature has a resolution of 2.5ns on all

modules.” 1

Additionally engineers observed no increase

in average port-to-port latency measured

over a period of 16 hours, and no packet

drops were observed.



1 https://www.spirent.com/Blogs/Networks/2012/May/2012-05-10_Measuring_Latency_in_High-Speed_Low-Latency_Networks

Test Variables

Table 1

Test Variables Frame Sizes (bytes)

Frame Sizes

S-Blade Pro Operating Mode

Network Topology

Traffic Destination

S-Blade Pro Port Type

64, 256, 512, 1518, 9600

Normal (Standard), Utilization

10GbE, 40GbE

On-blade/Off-blade

Layer 1, Smart Port, Replication

https://www.spirent.com/Blogs/Networks/2012/May/2012-05-10_Measuring_Latency_in_High-Speed_Low-Latency_Networks



Terms of UsageThis document is provided, free-of-charge, to help you understand whether a given product, technology or service merits additional investigation for your particular needs. Any decision to purchase a product must be based on your own assessment of suitability based on your needs. The document should never be used as a substitute for advice from a qualified IT or business professional. This evaluation was focused on illustrating specific features and/or performance of the product(s) and was conducted under controlled, laboratory conditions. Certain tests may have been tailored to reflect performance under ideal conditions; performance may vary under real-world conditions. Users should run tests based on their own real-world scenarios to validate performance for their own networks.

Reasonable efforts were made to ensure the accuracy of the data contained herein but errors and/or oversights can occur. The test/audit documented herein may also rely on various test tools the accuracy of which is beyond our control. Furthermore, the document relies on certain representations by the sponsor that are beyond our control to verify. Among these is that the software/hardware tested is production or production track and is, or will be, available in equivalent or better form to commercial customers. Accordingly, this document is provided "as is", and Tolly Enterprises, LLC (Tolly) gives no warranty, representation or undertaking, whether express or implied, and accepts no legal responsibility, whether direct or indirect, for the accuracy, completeness, usefulness or suitability of any information contained herein. By reviewing this document, you agree that your use of any information contained herein is at your own risk, and you accept all risks and responsibility for losses, damages, costs and other consequences resulting directly or indirectly from any information or material available on it. Tolly is not responsible for, and you agree to hold Tolly and its related affiliates harmless from any loss, harm, injury or damage resulting from or arising out of your use of or reliance on any of the information provided herein.

Tolly makes no claim as to whether any product or company described herein is suitable for investment. You should obtain your own independent professional advice, whether legal, accounting or otherwise, before proceeding with any investment or project related to any information, products or companies described herein. When foreign translations exist, the English document is considered authoritative. To assure accuracy, only use documents downloaded directly from Tolly.com. No part of any document may be reproduced, in whole or in part, without the specific written permission of Tolly. All trademarks used in the document are owned by their respective owners. You agree not to use any trademark in or as the whole or part of your own trademarks in connection with any activities, products or services which are not ours, or in a manner which may be confusing, misleading or deceptive or in a manner that disparages us or our information, projects or developments.

About TollyThe Tolly companies have been

delivering world-class IT services for

more than 25 years. Tolly is a leading

global provider of third-party

validation services for vendors of IT

products, components and services.

You can reach the company by E-mail

at [email protected], or by telephone at

+1 561.391.5610.

Visit Tolly on the Internet at:

http://www.tolly.com

216149-wt-2016-11-10-VerJ

For more information contact:

NETSCOUT Systems, Inc.800-309-4804

www.netscout.com

http://www.netscout.com

mailto:[email protected]

http://www.tolly.com

NETSCOUT S-Blade Pro & TestStream V4 - Spirent...LAN to manage the NETSCOUT device as well as to...

Documents

Transcript of NETSCOUT S-Blade Pro & TestStream V4 - Spirent...LAN to manage the NETSCOUT device as well as to...