DTI Custom Publication, 2009

From a Leading Provider in
Embedded Computing and Switch Products
for the Government and Military Markets
Diversified Technology, Inc.

> COTS Gotchas
What to Look for When Selecting a COTS Provider

> 10G Ethernet - AdvancedTCA Switching
An AdvancedTCA Switching and Protocols Primer

> COM-onomics
The Economics and Use of COM Express in Embedded Applications

> Mobile Power for Mobile Forces
On-Board Vehicle Power

> High Availability Load Balancing
With HAProxy on AdvancedTCA Hardware

Embedded Computing Solutions
Diversified Technology, Inc. (DTI) designs and manufactures embedded
platforms, communication servers, and subsystems for the mission critical and
non-critical demands of the government computing market. DTI Solutions,
delivered on a variety of form factors, offer long life support for multiple
government agencies involved with protecting global and domestic interests.
Embedded computing in the military will continue to drive DTI to new COTS and
proprietary solutions to support the government's defense initiatives.

> Find out more at www.dtims.com

AdvancedTCA • VME • CompactPCI • PCI • ETX • COM Express • Custom Design and Manufacturing

High Performing
Platform Systems

• Communication Blade Servers based on Open
Architecture Standards such as CompactPCI®,
VME and AdvancedTCA®, and featuring the
latest processing and switched fabric technologies.

• PCI/ISA CPU Boards and Rackmount Systems
utilizing high-slot count backplanes for
third-party cards in the end solution.

• Embedded Computer Modules based on ETX
and COM Express provide computing power
in reduced physical dimensions.

• Outsourcing and Customization Options
Available through DTI’s In-House Engineering.

Three Decades of Embedded Solutions
CPU Boards - Switch Blades - Rackmount Systems - Modular Platforms
Computer Modules - System Integration - Custom Designs - Outsourcing Capabilities

1.800.443.2667 • sales@dtims.com • http://www.dtims.com

All trademarks and tradenames are the property of their respective owners.

Providing a Cohesive Approach to Embedded Solutions

476 Highland Colony Parkway
Ridgeland, MS 39157
(800) 443-2667
Inside This Issue
marketing@dtims.com
www.DiversifiedTechnology.com COTS Gotchas..........................................................05
Key Areas to Consider when Selecting A COTS Provider
Pat Busby
President 10G Ethernet ............................................................15
Barrie McArthur
Vice Chairman - Special Projects

Steve Craven
COM-onomics ..........................................................22
Vice President, Program Management The Economics and Use of COM Express in Embedded Applications

Donald Germany
Vice President, Systems Engineering Mobile Power for Mobile Forces ............................26
On-Board Vehicle Power (OBVP)
Ken Martin
Vice President, Finance
The Cohesive Approach
Joe McDevitt
Chief Technology Officer How to Achieve High Availability............................32
Load Balancing with HAProxy on AdvancedTCA Hardware
Gary Smith
Vice President, Operations
Message from the President ..................................04
Keith Varner
Vice President, Solutions Group VME Solutions..........................................................12
Patrick Welzien
Vice President, Engineering Custom Designs ......................................................13
Tom Witt Employee Spotlight ................................................13
Vice President, Business Development
About Diversified Technology, Inc. ........................45
About Diversified Technology, Inc. Joe Speak ................................................................45
Diversified Technology, Inc. (DTI), an Ergon
Company, was founded in 1971 and focuses DTI Tradeshow Schedule ........................................46
on primary markets of Communications,
Government / Military, Commercial, Power
Electronics, On-Board Vehicle Power and
Simultaneous Field Radiation Technology. As No part of this publication may be reproduced, stored in any retrieval system, or transmitted,
in any form or by any means, electronics, mechanical photocopying, recording or otherwise,
an embedded hardware, software and without the prior permission of Diversified Technology, Inc.
systems company, DTI’s strength lies in a
DTI Wired is available free of charge to qualified subscribers, by emailing marketing@dtims.com
cohesive approach to assisting customers.
This cohesive approach means DTI works To be removed from our mailing list, send a removal request via email to marketing@dtims.com

hand-in-hand with companies to ensure they Other product and/or company names mentioned in this publication may be trademarks or
are getting the best performance, highest registered trademarks of their respective companies and are the sole property of their respective owners.

reliability, shortest time-to-market and the Copyright © 2009 Diversified Technology, Inc. - All Rights Reserved
most efficient use of computing hardware for
their program's embedded application. DTI
has a history of design and manufacturing
experience with standardized form factors
such as PCI, ETX, COM Express, VME, Page 3
CompactPCI and AdvancedTCA.
DTI Wired : Get Connected with Diversified Technology, Inc.

Message from the President
Pat Busby, President of Diversified Technology, Inc.

Given the wide range of business and economic challenges in today’s markets, it’s become vitally
important to have experience and staying power. Since 1945, the average recession lasted 10 months,
but from 1919 to 1945, the average dragged on for 18 months. The last two recessions - from March
to November 2001 and from July 1990 through March 1991 - were shorter than average, lasting only
eight months each. This is the 5th recession during DTI’s 38 year history (1971), and the 8th for
parent company Ergon (1954). Downturns are a normal part of an economic cycle, but they also
provide opportunities for those seasoned and primed to take advantage. Yes, some businesses will go
under, but others will emerge stronger. Those desiring to be among the latter need to be careful about
many factors, including supplier choices.

DTI is the supplier of choice to many customers for design, manufacturing and delivery of rugged
embedded computers for their solutions. DTI also provides complete integrated solutions for telecom,
medical and military customers. Our CompactPCI and AdvancedTCA portfolio is transitioning well into
military applications. Along with our new technologies such as VME, rugged mobile power, RF
technology, and energy inversion products, DTI will continue to invest in R&D to innovate and deliver
products and value as a long term supplier.

In this issue of DTI Wired we offer direction and guidance on compute, switch fabric, rugged mobile
power, and wireless technologies integrated into products which improve our customer’s financial return.
With the complexity of disparate environments, high performance processing, and balanced resources
DTI provides solutions for long term expandability, economic value and operational maturity. I hope you
find this publication helps you in navigating through the technology and product acquisition in an
ever-changing environment. We look forward to serving you for the next 38 years, too.

A. Patrick Busby

Page 4


COTS Gotchas
What to Look for When Selecting a COTS Provider

Introduction

The growth in use of COTS (Commercial There are, however, issues that can arise when
Off-the-Shelf) hardware and software across moving to a COTS-based platform for your
numerous markets, most notably in the organization’s application. We will call these
telecommunications, government/ military, and issues “Gotchas” in this text, and explore not
commercial sectors has led to discussions over only the financial impact, but also other nega-
the benefits a COTS-based platform provides tive results on your organization’s development
over a proprietary or completely customized and deployment of your application that will
solution. These benefits revolve around occur by you neglecting to investigate these
reduced deployment costs, easier maintenance COTS Gotchas.
and support, the removal of vendor lock-in, and
an increased ease of integration and scalability. COTS Gotchas come in two forms:
These aforementioned virtues of COTS Feature Gotchas and Supplier Gotchas.
hardware have been well documented and
proven to be true for all but the most severe of Feature Gotchas refer to claims made about
circumstances. the specifications of a product as it compares to
the standard itself and as it will be used within
AdvancedTCA has seen an up swing in usage an application, essentially forcing you, as a
in classical military COTS recently. As a PICMG potential COTS purchaser, “Does this product
specification (3.x), AdvancedTCA (ATCA) actually do what it claims to do?” or “Does this
focuses on providing high speed interconnect product have all of the things the datasheet says
technologies and improved reliability, it does?”.
serviceability, and availability to telecom
applications; many military applications demand Supplier Gotchas, however, are more nefari-
the exact same needs. Developed as an ‘Open’ ous and insidious in their undermining of your
Standard (meaning PICMG member organiza- organization’s success. They typically do not
tions all have access to the specs to show themselves until late in the game, typically
develop their own boards, chassis, and well after a purchase order has been placed!
add-ons), ATCA seeks to provide a true COTS Supplier Gotchas can be summarized as the
environment in the traditionally proprietary end results of what a supplier promises versus
military industry, and it is seeing growing what the supplier actually delivers. Supplier
success in this mission due to many of the Gotchas can be applied to nearly any COTS en-
advantages mentioned above that a COTS vironment and are not solely the property of
based environment brings to a market. ATCA suppliers. Any organization looking to

Page 5

move to any COTS platform should keep these Before we move into explanations of the above
in the front of its mind. For this reason, this Gotchas and discussing the impact they can
article will focus primarily on Supplier Gotchas. have on your organization, we must first
understand that what follows is by no means an
To note: This is by no means an attack on COTS implication of COTS products as a negative.
methodology or use. Commercial Off-the-Shelf I fully support the use of COTS products
hardware is certainly preferable in nearly every wherever it makes sense to do so and firmly
way to proprietary deployments, and I do not believe that the touted benefits of COTS are very
intend this as an attack on COTS as a whole. much real and tangible.
The following is simply a discussion of key areas
of interest that must be planned for and looked at These Gotchas are issues you must be aware of
closely when selecting your COTS supplier. when selecting a COTS provider.

Overview of the COTS Gotchas Validation and Compliance
In discussing COTS Supplier Gotchas, we must
aim to stay relevant. Therefore our COTS Validation
Supplier Gotchas will revolve around some of Validation refers to a solution being ‘validated
the most critical supplier services organizations and tested’ with your application and in a field-
desire when looking at a COTS Solution : like environment. One of the main benefits of
proprietary solutions is that due to a close
• Validation and Compliance relationship with the provider, your organization
can be assured that the solution will be validated
• Support and Maintenance
to integrate with and work in your specific
• Deployment Costs and Life Cycle Management application. The concern is that when dealing
- Obsolescence/EOL Planning with a COTS product, it has been only generally
- Asynchronous Upgrades validated within the scope of similar usage or,
even, not validated within any ‘like’ application.
• Certification and Training This has been called Black Box Testing,
meaning boards are released once basic
• Custom Hardware Design
application equivalency is demonstrated within
- Differentiation
the scope of the Supplier’s testing methods.
• No Vendor Lock-in
(Continued Open Standardization) If a board passes testing, such as environmen-
tal, or demonstrates like-application equivalency
These six COTS Gotchas are by no means the within the COTS Supplier’s testing facility but
only considerations you must make when fails within your deployment, will or can the
looking into a COTS Supplier, however they do COTS Supplier provide the support you need to
represent some of the more important issues overcome your unique issues, in say a barge
that can arise. test?

Page 6


I will put forth and state plainly the notion that and is intimately familiar with the process and
some COTS Suppliers can provide this sort of testing procedures. Tying in with Validation, be
validation and testing, but you must choose sure the supplier will be willing to work very, very
wisely. Generally speaking, a more focused closely with you to help you pass your testing.
COTS Supplier will be more willing to demon-
strate exact application usability with their Support and Maintenance
product rather than simple equivalence testing,
and you can be assured that this COTS product Simple Support
will not only work within your application, but that Once you have deployed your solution with
the COTS Supplier will indeed be willing, ready, COTS products, you will be required to maintain
and capable of helping you overcome any the application throughout its lifecycle. Will your
unique problems that may arise. COTS provider be willing to give you in-depth
support and maintenance help?
COTS Gotcha Validation:
Select a COTS Supplier who can and will devote Many COTS providers will claim global support
the time necessary to solve the unique issues in and most of them will charge you for providing
your deployment. it. But what is it that these COTS providers are
offering? Part-timers following a flowchart with
Compliance scripted replies, elevating a problem once critical
Many organizations in the United States are status has been reached, your application has
trying to achieve compliance to regulatory been down for a week (leaving you losing not
standards and testing or to comply with non- only revenue, but the respect of your end user),
American standards to supply their product all the while your interaction with the Level 1
overseas . The most common standards being support guys for that time has left you reduced to
looked for are MilSpec 810F. begging and pleading to speak to someone who
is familiar enough with the product and your
Many organizations will claim a product meets application to actually offer you some modicum
compliancy. Very rarely, however, does a of actual, real help.
supplier actually take their product through the
testing. COTS products, when they claim are Realistically the above will be the least of your
typically “designed to meet” rather than have worries should you find yourself involved in
actually passed testing. For your use, if your “Waiting on Feedback” limbo. Even after Level
product needs to meet specific compliancy, you 1 phone support elevates your issue, will it be
will have to take, and pay for, the qualifications resolved? Many times, the next level of support
yourself. does not involve the COTS provider attempting
to solve your issue; rather it involves the COTS
COTS Gotcha Compliancy: supplier looking for any slight variation between
When selecting a product designed to meet intended and actual board use in your
specific compliants, be sure that the supplier has deployment in order to label your issue as ‘not
a history of taking products through that testing my problem’.

Page 7

COTS Gotcha Simple Support: leading technologies, solid COTS products
Try to talk to other customers of the proposed presently available, and assess technology
COTS supplier to get a feel for how they handle trends to determine a plan for your product
support requests. Also, evaluate just how close refresh and upgraded COTS insertion. Included
of a relationship will be forged between you and in this would be a declaration of expected EOL of
your COTS provider, similarly to how you should a product from your preferred COTS vendors.
to make sure their Validation procedures will
match your needs. You should focus on All products will someday be declared EOL by
selecting a COTS supplier who will be dedicated their manufacturer. Reasons may be that their
to solving your unique issues swiftly, rather than vendors have stopped producing the necessary
simply focused on reading a script. materials, or an upgraded version of the current
product is now offered and the manufacturer will
not produce more of the older product due to the
Deployment Costs & Life Cycle Management costs. By having accurate and thorough EOL
One of the key benefits of utilizing COTS planning, your organization will be prepared to
products is a lower deployment cost for your deal with last-time buys of a dated technology
application. While it is absolutely true that initial and be more ready and able to deploy a next-
deployment costs within a COTS platform are generation iteration of your application, while
cheaper than a proprietary solution , there are avoiding many pitfalls and headaches.
associated Gotchas when looking at the overall
lifecycle of a platform, from initial deployment to By thorough obsolescence and EOL planning,
product upgrades. your organization can have actual, real life cycle
management of your application when using a
Obsolescence/End of Life Planning COTS platform. It is not an impossible task – far
One of the biggest concerns with using COTS from it. A more frustrating problem exists, and
hardware is that of obsolescence and EOL this problem is not generally planned for: Life
Planning . For our purposes, we can refer to the cycle asymmetry.
Department of Defense’s definitions of obsoles- Every system has a life cycle. A system is
cence and End of Life. Obsolescence is merely a set of parts, and each of these
diminished manufacturing and material individual parts has its own life cycle, which
availability of a COTS product, and End of Life probably will not coincide with one another.
(EOL) is a manufacturer’s notice to customers of Upgrades will allow an extension of overall
a last time buy of a product. system life time by replacing obsolete or EOL
parts of the system. But, given the nature of
Obsolescence and End of Life occurrences are development cycles and part availability
definite events that must be planned for with schedules, component upgrades can become
care and exactitude when developing a COTS out of sync with each other.
application. Ideally your organization will
conduct market research on its own to survey

Page 8

RUGGED COMPUTING
The PlexSys-RP from DTI is a self-contained,
rugged CompactPCI based portable computer.
This MIL-STD chassis is equipped with 2 handles
for easy transport making it a completely portable
battlefield workstation.

Multiple CPU Blade and Switch configurations are
available including Intel’s Core 2 Duo, Dual-Core
Xeon, and Low Voltage Pentium M.

Call NOW
for Immediate
Deployment
1-800-443-2667

Even in a simple hypothetical two-board COTS comes to upgrades. Sometimes the vendor’s
system, this can be demonstrated. If one board development cycle happens to slip due to un-
has a life cycle of 18 months while the 2nd is on foreseen circumstances. Ether way, your appli-
a 24 month schedule, the next-generation cation suffers and you have been hit by yet
release dates will only align once every six another COTS Gotcha.
years. For every board you add on a different
schedule, your release alignment probability
decreases exponentially. This does not even COTS Gotcha Lifecycle Management:
include patches, updates, firmware revisions, You must fully implement an obsolescence and
and other changes. lifecycle management program focusing on
acquisition instead of problem resolution after
Asynchronous life cycles are likely to cause the deployment. Without this in place, a COTS
greatest headaches and problems of your COTS system will cause more headaches down the
system deployments. line than it will solve. To this, add in the problem
of Asynchronous Upgrades. You must plan for
Your organization may find that although you the life cycles of the individual boards within your
have accounted for your applications life cycle COTS system and plan for implementing the
and part EOL situations, your vendor’s develop- future technologies in your application without
ment lifecycle can cause you headaches when it loss of uptime.

Page 9

Training/Skill Updating So, when evaluating your potential COTS
When developing your new COTS application, it suppliers, you must ask yourself, “How will my
is likely that your engineers will be charting new engineers be trained and gain the skills
waters. They may be unfamiliar with the form necessary to effectively implement this solution
factor, fabric, or any number of other into our application?”
components that will be used in your new build.
In any case, they are going to have to become COTS Gotcha Training:
intimately familiar with the technology in order to Try to find a COTS Supplier that is willing to form
successfully develop and deploy your next- a cohesive bond to essentially become an
generation COTS-based application. extension of your engineering team to help in the
training and skill development of your engineers.
There are two ways in which your Engineers can Doing so will give your engineers the ability to
learn about this new technology. First, they can fully implement and maintain your COTS
read the product documentation (manual) and solution.
use their search engine of choice to find
discussions, papers, and information relating to
the technology. I think it should be fairly obvious Differentiation through Hardware?
that this would be a time consuming and You have probably been told that in a COTS
wasteful method. To truly understand your new marketplace, your solution differentiation comes
COTS deployment, your Engineers will need via software. Let’s extrapolate in case you
true training and learning experiences that focus haven’t:
on your application.
Can software provide you with double or triple
Some COTS companies, if they are truly serious (or more) performance increases between two
about their technology and YOUR success, will generations of your product? Very unlikely.
be willing to invest the time needed to provide These kinds of gains are only obtained through
you with evaluation and test units, give online hardware upgrades. If you and your competitors
seminars, answer specific questions via are using the same COTS hardware you will not
telephone, and a rare few will be willing to hold differentiate yourself from them in terms of
workshops at your lab or location to teach your performance – you and your competition will
engineers the technology on a near one-to-one have become aligned, attempting to outdo each
basis. They will do everything in their power to other via marketing instead of by out perform-
train your engineers in the effective and efficient ance.
use of their technology and instruct you on how
best to implement their COTS solution within Some COTS suppliers can and are willing to
your application – not for their own gain but provide not only minor “look and feel” changes
rather for your success. but also fully customized features, peripherals,
and even boards for your COTS deployment.
Sometimes non-recurring efforts are needed to
assure future success through differentiation.

Page 10


A good COTS provider will be able to give your within their market, you should encounter very
application the hardware performance few interoperability problems. But what about
differentiation that is needed to ensure your times in which a COTS vendor chooses to take
success in the marketplace. After all, to a good a proprietary route of some form while claiming
COTS supplier, your success is their success. full adherence to an open marketplace and the
interoperability standards therein – something
COTS Gotcha Differentiation: like a proprietary switch stack?
While we’ve heard time and again that with
COTS deployments your differentiation must Using the proprietary switch stack example,
occur via software, you should determine the interoperability can become a problem. Want to
value of a COTS supplier who will also provide use Switch 1 from Vendor A with a proprietary
you with the hardware differentiation you need stack and Switch 2 from Vendor B with a
and desire. If you fail to do this, the performance standard stack in the same application? You
alignments among you and your competitors can very likely will run into severe interoperability
make it difficult to justify your end-solution to problems that will bring your development and
prospective customers. deployment schedule to a standstill until the
problem is solved.
Vendor Lock-In
One could make an argument that the COTS Gotcha Vendor Lock-In:
abandonment of Vendor Lock In is a ‘Holy Grail’ Just because an organization claims to be a
justification for the use of COTS in your COTS supplier does not necessarily mean that
deployment – which is a wonderful thing. When their components will be fully interoperable
using COTS, gone are the days of jacked-up within your specific application. You should, at
hardware, integration, and support prices solely the very least, test the solution before selecting
for the reason that the provider can charge what- your COTS supplier. If you have already chosen
ever they feel they can. That’s not to say that your supplier and run into the aforementioned
hardware, integration, and support prices can’t problem, I hope you have chosen one of the few
be exorbitant, but with a COTS ecosystem the suppliers who will be ready, willing, and able to
market dictates pricing structures, rather than a devote the time and resources to overcoming
single company. COTS vendors have to com- your issues.
pete not only on feature set, lead times, support,
and capability, but also on price.
Key Fact for DTI:
Oversight organizations within the COTS One in every five AdvancedTCA Switches being
marketplace help ensure interoperability among used today were designed and manufactured
manufacturers to maintain an open marketplace by Diversified Technology, Inc.
so users of COTS hardware are not forced to
purchase from one vendor. If COTS suppliers Ref: VDC Report, 12/3/06
maintain adherence to standards and practices

Page 11

The above six items represent key areas that If you incorporate heavy planning in all of these
you must be aware of when developing and areas to your COTS development plan, you can
deploying a COTS based system. If your COTS not only be sure of real success, but also of a
supplier is lacking in any one of the aforemen- headache-free deployment leading to a much
tioned areas, you can be sure that you will have faster time to revenue for your organization.
to overcome costly delays and headaches.
Article Written by:
When planning your COTS system, you must
account for all areas of Lifecycle Costs, from
initial deployment to planning for obsolescence.
Your COTS provider must be willing to work for
you to overcome your unique issues, essentially
making themselves an extension of your
engineering team. You must be aware of
vendor lock-in traps while selecting a COTS Joe McDevitt
supplier that will be capable and willing to give Chief Technology Officer
you the differentiation you need via hardware –
not just passing the buck of true differentiation
to your software developers. Also, your COTS
provider should be willing to train your engineers
to get the most out of the hardware, be it via
email, telephone, web-based training, or most
importantly, on-site training.

VME Solutions
Due to our increased focus on Military and Governmental applications, Diversified Technology, Inc. is
considering bringing our computing and leading switch design expertise into the VME/VPX marketplace.
If you have VME/VPX needs, or would like to discuss future plans therein with DTI, please contact Joe.

Joe McDevitt
Chief Technology Officer
joe.mcdevitt@dtims.com
1.800.443.2667

Page 12


Custom Designs
Application-Specific Computing Solutions
Diversified Technology, Inc. realizes that standard products do not meet the needs of every application.
To satisfy these design-specified requirements, DTI routinely develops custom solutions. These designs
can be as simple as adding or removing one feature to an existing catalog product, or they may be as
complex as a full platform design implementing next generation technology. In either case, the
development cycle is simplified by DTI's in-house tools, capabilities, and manufacturing lines. Advanced
PCB and mechanical design software combined with libraries of function blocks streamline the initial de-
velopment. BIOS source code engineers ensure the functional configuration is optimized to customer
needs. Rapid prototyping is achieved on DTI's state-of-the-art SMT assembly lines and in DTI's metal
shop. Finally, DTI's verification and validation labs guaranty that the product is fully operational and
meets industry standards and can pass certification testing.

www.DiversifiedTechnology.com/custom

The Employees are DTI’s Most Important Asset
Employee Spotlight - Elmer Durrell III
Elmer currently serves as the Design Group Manager for DTI and is
responsible for the overseeing and coordinating of all hardware design
engineers, the mitigation of program roadblocks, and assuring that project
milestones and objectives are met. Elmer graduated from Mississippi State
University with a BS in Electrical Engineering and a minor in Mathematics.
He first joined Diversified Technology, Inc as a Hardware Engineer eight
years ago. In his role he enjoys debug efforts, solving tough problems, and
growing others to be their best.

Elmer and his wife, Wendy, are the proud parents of two children; daughter, Arianna age 5 and Son,
Elmer IV age 3. He and his family are active members of Triumph Church in Vicksburg, MS where he
servers as a Church Elder, Board of Trustees Member, Teacher, and Former Sunday School Director.
Elmer loves spending quality time with his family, playing golf and basketball.

Page 13

BOTTLENECKS SUCK

ELIMINATE CONGESTION WITH
DTI’S 10 GIGABIT ATCA SWITCH
www.diversifiedtechnology.com

ATS1936
Flexible 10 Gigabit
AdvancedTCA Switch


Diversified Technology, Inc. (DTI)
1.800.443.2667 • sales@dtims.com • www.diversifiedtechnology.com/atca
Visit us on the web to find out more about DTI’s AdvancedTCA products.

All trademarks and tradenames are the property of their respective owners.


10G Ethernet

processing level but rather at the switch level,
where the system cannot keep up with the band-
width demanded. With ATCA, the switch drives
Traditionally in embedded hardware, specifically not only the overall functionality of a given
in blade systems, the host or processing node chassis, but also the overall performance of the
slot has been the focal point of chassis application.
functionality. The architecture itself lent the CPU
node to being the control, managing everything
from IP traffic to I/O connectivity. In Ad- Hesitation of ATCA Adoption
vancedTCA, the core of the system is no longer Bandwidth Increases from 10GbE
the CPU blade. Not only is the IP traffic
processed at the switch level with Ad- As the implementation of bandwidth-intensive
vancedTCA, but also most of the I/O connectiv- applications continues to grow within the
ity will be processed over IP, creating a situation Communications industry, the need for a drastic
in which the switch drives the overall chassis increase in throughput becomes not only
functionality. obvious, but absolutely necessary in order to
provide the kinds of feature-rich and robust
battlefield applications that the military requires
to remain successful in the evolution of the
“connected battlefield.” There is no denying that
the AdvancedTCA architecture is gaining
tremendous ground within this sector, namely
because of the increased processing power and
bandwidth it provides.

The issue is, however, that although
AdvancedTCA offers these increases in
performance, the rapid growth of bandwidth-
intensive applications means that even this
soundness of architecture begins to choke and
the available processing power and bandwidth
With the development of the next-generation becomes over-saturated. Here is where the
battlefield and network-centric warfare, many in Communications industry must look forward
the military are finding that bottlenecks of an ap- within the AdvancedTCA offerings to 10G
plication’s functionality are not at the traditional Ethernet fabrics.

Page 15

applications, its evolved form of 10G will be the
default go-to for the Communications industry.
Familiarity with the ‘core’ is key.

Ethernet is also a “best effort” fabric. Traffic is
processed as best a switch can manage, but,
unlike FedEx or UPS, there is no guaranteed
“overnight” service. When a “best effort” fabric is
in use, the best effort an application architect can
make is to over-design their AdvancedTCA
platform.

DTI’s AdvancedTCA Manufacturing As stated, Ethernet is inexpensive to implement.
If an organization were to over-design their
Some target engineers have hesitated on the planned 1G solution to 10G, they would find the
adoption of AdvancedTCA in the development additional expenditure to do so is merely around
and deployment of applications, mainly due to 4% of the overall chassis cost. Yes, four percent.
the lack of bandwidth that 1Gb or 2x1Gb This means that for a nearly indistinguishable
Ethernet provided in comparison to their increase in cost, a solution can be over-designed
proprietary solutions. That argument is now to use 10G Ethernet, providing more assured
gone. AdvancedTCA board and system bandwidth within the platform.
manufacturers are poised to offer the necessary
bandwidth and reduced latency that telecommu-
nications applications require (with room for
overhead), but can also offer the lower cost of
moving to an open-standard architecture with
well-designed 10G solutions.

Ten-Gigabit Ethernet is the current leader
(in terms of actual throughput capabilities) of the
Ethernet standards, providing ten times the data
transfer rate of the conventional (and most-used)
1GbE solutions provided on most AdvancedTCA
processing nodes and switches. Ethernet is a
familiar standard, and 10G is the natural
evolution, providing tremendously increased
throughput and performance over current 1G
solutions. Due to the low cost of implementation DTI’s AdvancedTCA Testing
and corresponding astronomical use of Ethernet
in nearly all data and communication

Page 16


ATS1936 : DTI’s AdvancedTCA 10G Switch Blade with 3 AMCs

DTI’s ATS1936 is a flagship 10G ATCA Switch The ATS1936 fully supports IPv4 and IPv6
that provides the bandwidth required on today’s routing on both the Base and Fabric networks.
networked battlefield, as well as offering the The ATS1936 features 3 AMC mid-size sites, all
flexibility to allow for direct customization through of which support telecom linecards for telecom
AMC sites. clocking options. The ATS1936 also supports an
RTM with 10G Ethernet ingress/egress ports.
The ATS1936 is a true ATCA 3.1 Option 9 ATCA Using a fully standardized design approach, DTI
Switch, featuring 20-port 10G, wirespeed has enabled the ATS1936 to provide the highest
non-blocking fabric powered by Broadcom’s performance at the lowest possible cost in an
XGS3 silicon. With the same silicon, the base AdvancedTCA switch for use in telecommunica-
switch features 3 10G port, 23 auto-negotiating tion applications such as VoIP, IMS, IPTV,
1G ports, and one 10/100 management port. Wireless, Security, and other solutions.

Ethernet management on the ATS1936 is A Closer Look
provided through a third party with emphasis on Key Advanced Ethernet Features
Enterprise support. DTI utilizes the Full Source
Model with the management software, which The advanced Ethernet features of the ATS1936
runs directly out of user memory—negating any make it a truly robust and resilient solution for
license exposure issues that are sometimes any military application, and provide the tools
found on other ATCA Switches. The Ethernet necessary to satisfy the aggressive need for
management functionality fully supports IPv6 bandwidth in today’s military applications.
routing, and offers superior performance in
certain critical protocols, such as Spanning Tree. Redundancy
In utilizing the management software in this The key theme in any AdvancedTCA system
manner, Diversified Technology has been able to and, truly, of the ATCA specification as a whole is
focus solely on providing superior ATCA redundancy. Within an Ethernet network, two
functionality with the ATS1936. key protocols come into play when supporting

Page 17

redundancy: Multiple Spanning Tree Protocol In the presence of a Link Aggregation Group
(an extension of Rapid Spanning Tree) and Link (LAG), network designers can combine multiple
Aggregation Group. links between many network devices and group
them as a single link, providing increased
Multiple Spanning Tree Protocol is an extension throughput and redundancy.
of Rapid Spanning Tree and Spanning Tree. By combining MSTP and LAG on the ATS1936,
Spanning Tree Protocol (STP) allows loops and DTI has allowed for N-way active (all N active)
redundant links to exist on a single network by redundancy for higher throughput and
allowing only one active path between any two availability between systems and networks.
network devices. Rapid STP works to make the
network adapt to changes much more quickly Layer 3 Routing & Redundancy
than it would without it. Multiple STP is an When Layer 3 routing is used, Virtual Router
extension of RSTP that gives VLAN support to Redundancy Protocol (VRRP) works wonders if
STP. With hardware that supports STP, RSTP, the switch is acting as a gateway. With VRRP,
and MSTP, a network is provided 1+M (1 active, multiple routers can be set to appear as a single
M Standbys) redundancy, because only one path router to nodes and other routers, which creates
is allowed to be active at a given time. highly available gateways for Layer 3 networks.

Page 18


If either switch goes down, the nodes will Multicasting with Layer 2 Networks
dynamically failover to use the other switch as Internet Gateway Management Protocol
the network gateway. VRRP often does not add Snooping and GARP Multicast Registration
much cost to a switch. VRRP should be Protocol bring the power of multicasting to Layer
considered a requirement if Layer 3 Routing is 2 networks. Though multicast is not often used,
to be used in a network, because it can be so it is critical to some networks. Without multicast,
useful. data is ether sent either to a single port or to all
ports of a switch.
Network Segmentation
Virtual Local Area Networks are a simple, but Multicasting is a Layer 3/Layer 4 technology that
powerful tool that increases security while allows a router to easily send one set of data to
decreasing broadcast traffic via segmenting the multiple recipients. Both IGMP Snooping and
network. While nearly every managed switch GMRP extend this technology to Layer 2
supports basic VLANs, there are several networks. IGMP Snooping does so by
important extensions to VLANs that are not “snooping” the higher level multicast packets and
always supported, including GARP VLAN forwarding the packets only to the ports that
Registration Protocol (GVRP) and protocol have members of the multicast group in that
based VLANs. GVRP can greatly ease the packet, whereas GMRP works by propagating
hassles of the configuration of larger networks multicast settings though a network. Both GMRP
by propagating VLANs automatically though the and IGMP Snooping reduce broadcast traffic on
network. Protocol-based VLANs allow for networks that use multicast.
network segmentation by protocols rather than
by physical ports. This is useful for networks that Often unused, port mirroring can be a valuable
need to support diverse protocols. tool for addressing network problems. With port
mirroring all the data going out of a port, or ports,
Data Control can be copied to another port for monitoring. Port
With high network traffic, important data should mirroring can be used in traffic analysis, problem
be prioritized through what is generically called debugging, and even security monitoring.
Quality of Service. QoS can be implemented in
many different ways, but two of the more useful Remote Management
methods in ATCA are Access Control Lists For large networks Remote Management is a
(ACLs) and Differentiated Services (DiffServ), definite requirement. For AdvancedTCA, SNMP
both of which are available on the ATS1936. (Simple Network Management Protocol)
provides the framework and is the key feature for
ACLs are simple rules that define how to remote management. CLI (Command Line
prioritize traffic and are based on things such as Interface) does provide a management interface,
IP addressing or a Layer-4 port number. but it is difficult to use programmatically. SNMP
DiffServ, on the other hand, allows for much provides remote management and monitoring
more powerful rules, and in turn, better control that is relatively easy to use in a programmatic
of data flow. way. Tools built around SNMP allow for constant

Page 19

monitoring, statistics collection, dynamic Cost-Benefit of Advanced Ethernet
reconfiguration, firmware updates, configuration A Visual Cost-Benefit Analysis of Key Features
backups, and more on the entire network.
SNMPv3 is of particular interest because of the Every AdvancedTCA system has different
broad security features it brings to SNMP. The requirements of the Ethernet switches used.
end result of SNMP support is less downtime Often the cost-benefit analysis of a feature will
because problems can be identified and solved show that the feature is needed, even when the
more quickly. cost is high.

Problem Analysis Below are examples of features that are very
Often unused, Port Mirroring can be a valuable important to certain networks while not used in
tool for addressing network problems. With port others. The chart will show that many of the key
mirroring all the data going out of a port, or ports, features discussed above that are present on the
can be copied to another port for monitoring. Port ATS1936 are of the utmost importance, but are
mirroring can be used in traffic analysis, problem only negligible in their impact on the overall cost
debugging, and even security monitoring. of the system.

GO AHEAD AND JUMP IN Simplify with COTS
Small Form Factor
Building Blocks

Small Form Factors
ETX and COM Express CPU Modules

Call Today to Get your System
“Application Customized”
1-800-443-2667


2 MHz to 2 GHz
Bringing it Together
What it all means

As stated earlier, with AdvancedTCA, the core of
the system is no longer the host or system slot
housing a processing blade, but rather falls to
the capabilities of the switch used in a system to
provide the bandwidth and features needed to
adequately support an organization’s key
applications. The rapid growth of the network-
centric battlefield means that even this
soundness of the architecture of AdvancedTCA
begins to choke and the available processing
power and bandwidth becomes over-saturated.

Ten Gigabit Ethernet offerings are what the
SFRT Antenna Development
government and military must look to in order to
solve these issues with their next-generation DTI also focuses on high
applications, and the ATS1936 10G ATCA Switch performance RF antennas for
provides not only the necessary bandwidth but customer-specific applications:
also the key features needed on an advanced
Ethernet network. Even in solutions that only
“require” 1G bandwidth, over-engineering to 10G
- Enhanced performance for
is an affordable and best effort design when specific bands
implementing a fabric that takes a best effort
approach to bandwidth. - Reduced coverage gaps

- Extended range

- Options for power limited
platforms

Article written by: - Reduced Physical and
JP Landry Radar Signature
Networking Program Manager

www.DTIRFSolutions.com

Page 21

COM-onomics
The Economics and Use of COM Express in Embedded Applications

Many embedded applications today require not and power management, and power and ground
only significant processing performance, but also interfaces, while the second provides SDVO and
increased I/O function in a compact form factor: legacy IDE and PCI signals with additional PCI
Enter COM Express. In utilizing the COM Express, LAN, and power and ground signals.
method of a module containing all of the
components for a host computer (offered as an The maximum total I/O support in COM Express
COTS component) attached to a carrier board becomes:
(COTS-developed or custom per application), 32 PCI Express lanes (80Gbps aggregate)
the COM Express form factor is poised to x16 PCI Express graphics
become the de facto architecture for balancing 2 LVDS Channels
performance, size, I/O and thermals in 2 Serial DVO Channels
embedded computing markets. By hardening 4 SATA-150 links (600MBps aggregate)
through extending operational temperature 3 10/100/1000 Ethernet ports (10G provisions
ranges and adding physical strengthening in the future)
characteristics, COM Express is also making 8 USB 2.0 ports
direct inroads to governmental and military/
aerospace applications. COM Express has also been designed to
support future generations of PCI Express and
In physical applications, COM Express can be SATA interfaces.
utilized as a standalone embedded solution or
as a processor mezzanine. To see this, we must The physical dimension requirements of the
look closer at the COM Express design standard COM Express CPU modules allows for two
and how its specifications allow it to be utilized different sizes: Basic and Extended, measuring
within an application setting. 125mm x 95mm and 155mm x 110mm,
respectively. The Extended Form Factor has a
Function and Design larger power budget, doubles memory capacity,
The standard for a COM Express carrier board is utilizes dual-channel RAM configurations, and
CPU-agnostic. The architecture allows the allows the use of larger processors and chipsets.
embedded industry to adopt LVDS interfaces
while maintaining legacy support. The Basic and Extended Form Factors,
however, share the same board-to-board
The board-to-board connectors in COM Express connections, signaling definitions, and mechan-
consist of two rows, in which one row provides ical assemblies. This means that any carrier
pins for PCI Express, LPC Bus, SATA, LVDS board can support either a Basic or an Extended
LCD Channel, VGA and TV-out, LAN, system CPU Module.

Page 22


As can be seen, COM Express is a very flexible As a standalone embedded board, the COM
architecture that can be used in a diverse range Express methodology creates a solution in which
of applications, due to the significant I/O there is significant processing and I/O bandwidth
capabilities, CPU-agnostic design, and the in a small physical footprint. Many applications
ability to interchange host CPU modules. So, in the aforementioned markets can take
how do these capabilities contribute to promote advantage of these characteristics by utilizing a
the use of COM Express as either a standalone standalone embedded board: Interactive kiosks,
embedded board or as a processor mezzanine? Self-Checkouts, training simulators, electronic
billboards and such can take advantage of the
features of COM Express by utilizing a stand-
COM Express – The Little Board that Could alone embedded board within their applications.
Standards such as AdvancedTCA are focused
on a much defined market segment (Telco and Within the same markets and applications,
Data) because of three main reasons: First, designers of embedded systems and applica-
significant bandwidth processing power. tions may also partition COM Express CPU
Second, there is no need, in most cases, for Modules from their own platform base, including
significant amounts of I/O within ATCA specialty I/Os and FPGAs, utilizing COM
applications. Third is the fact that, in the Express as a processing mezzanine. In doing
applications that utilize AdvancedTCA, size is so, these designers can focus on their applica-
not a very important issue. It makes sense, in tion-specific I/O on a custom baseboard, without
these cases, to utilize a separate processing the need to address high-speed switched fab-
node, separate switch nodes, and to create rics, signal reflections and impedance controls,
expansion via AMC modules. These three routing, termination placement, etc.
points can also be applied to MicroTCA, with
lesser implications on necessary size. The COM Express methodology creates an
AdvancedTCA and MicroTCA are high-band- architecture that can satisfy the needs of most
width, high-power, low I/O solutions mainly for (if not all) applications that require a balance of
Telco and Data applications. They suit their performance and I/O in a compact form factor,
market very, very well. and the flexibility of the specification allows for
the utilization of COM Express as either a stand-
COM Express, however, is more flexible. Any alone embedded board or as a processing
application in any market that requires high mezzanine within the proprietary building blocks
bandwidth, high I/O, and small size can utilize of other applications, allowing designers to focus
COM Express. Military/Aerospace, Industrial on their own core engineering competencies.
Automation, Test & Measurement, Security,
Transportation, and Medical markets all require While not an inherently rugged form factor,
the processing power, I/O capabilities, and small efforts are being made across vendors to offer
form factor that is present in COM Express. extended temperature variations of their
offerings, as well as adding options such as

Page 23

physical hardening characteristics for use in respond to demand fluctuations, competitive
embedded military applications, such as forces, and new technologies by quickly and
vehicular controls, robotics, and simulator efficiently modifying existing designs, inexpen-
control mechanisms. sively broadening their product portfolios, and
plug-and-play implementation of new COM
All of this is well and good, but the real questions boards.
are “How will the adoption COM Express affect
the overall bottom-line of an organization?” COM Express represents a safeguard to R&D
What is the total cost of ownership, the product investments and a lowering of total ownership
life cycle costs, time-to-market, transitional cost, because it allows designers to separate
metrics, etc.?” commodity COM Express modules from their
proprietary building blocks. Investments into
R&D for specific, value-added application
Dollars and Cents elements will not have to be reengineered for
First and foremost, COM Express represents a successive technology generations.
significant reduction in life cycle costs due to the
scalability and interoperability of modules. Once It is possible to achieve some of the form-fit-
a specific carrier board has been implemented function of COM Express via mini-ITX/nano-ITX
to an application, future hardware upgrades can motherboards and the like at a lower cost.
be implemented by simply changing the COM However, when dealing with these competing
Express Module. This significantly reduces the form factors, an organization loses significantly
hardware costs of upgrading, which represent more than it gains, mainly in the realm of life
the bulk of expense when a new generation of cycle availability and revision control. COM
CPU and RAM technology becomes available. Express manufacturers strive to provide at least
five years of availability and revision control of
In hand with the quick implementation of their products, allowing critical applications to be
upgraded hardware comes the faster time-to- provided with the necessary hardware through-
market for deployment and time-to-revenue out their lifespan.
after. There is a direct relationship between how
fast an organization can deploy and implement
hardware upgrades and how quickly said Key Fact for DTI:
organization can begin to see ROI. With COM All of the Products designed and manufactured
Express, the essential plug-and-play architec- by Diversified Technology, Inc. are done so in
ture of the standard means that next-generation the United States of America. We do not
processor and RAM technology can be outsource overseas in an effort to cut costs.
deployed, implemented, and working much
faster than previously, quickly generating a
revenue stream. Simply stated, the COM
methodology allows an organization to rapidly

Page 24


COM Express improves form-fit-function in ability to harden COM Express by extending
specific applications, and helps to reduce design operational temperature ranges and adding
risks by cutting development time and costs, physical strengthening characteristics is also
and, at the same time, providing the scalability extending COM Express into the realms of
and flexibility needed over a multi-year life cycle government and military/aerospace applications.
of an embedded application.
By reducing life cycle costs, protecting R&D
Finally, being a PICMG embedded standard, investments, and reducing time-to-market and
COM Express has no ‘proprietary’ builds. time-to-revenue metrics for any organization,
Standards-based COTS products have already COM Express makes a very compelling
been shown to reduce initial investment costs, argument for its economy within the embedded
as well as life cycle maintenance costs, due to industry.
the inherent competition among suppliers to
produce cost-effective solutions. Article written by:

As of now, COM Express represents the highest
performance available in small form factor
modules. The scalability, flexibility, throughput,
support for legacy and non-legacy standards,
and low cost of ownership make COM Express
a very viable choice for a host of embedded Doug Mays, Field Application Engineer
applications across many different markets. The

CMXLX45 Module

Baseboard

CMXLX45 : DTI’s COM Express Module with Custom Baseboard

Page 25

Mobile Power for Mobile Forces
On-Board Vehicle Power

Current threats on our military forces have So what's changed? The truth is the U.S. has
created a tremendous requirement for mobility come through a long dormant period, nearly 30
as it relates to mission specifications. The years following the end of the cold war, where
objectives of OBVP are to develop, demonstrate we as a nation got complacent while the events
and transition electronic technologies that in the world were changing rapidly around us. In
enable lighter, physically smaller, more efficient defense of our military, the need to change was
and more reliable electronics, which are easier recognized much sooner by them, than we as a
to operate and maintain. This in turn will enable nation through our elected officials. An apathetic
current and emerging electronic systems to view of our vulnerability developed that has
increase performance, have improved physical resulted in our current state of military readiness.
characteristics, be more reliable and require The events of 9/11 and in Iraq were a shock and
significantly less lifecycle resources. Further an unfortunate wake-up-call. Since that time
development of these programs will also there has been a major shift in military strategy.
maximize the advantages of increased DC The primary emphasis is no longer on watching
supply voltage inherent in the HMMWV, FMTV our borders for approaching missiles or
and HEMTT for delivery of a power solution for bombers. We now are faced with a new enemy,
multiple military platforms in excess of today’s terrorists, who scatter like roaches, hide in the
available 10KW. dark and pop up anywhere and everywhere on a
global basis.
Sometimes OBVP is referred to as "Under the
Hood Power", but whatever the vernacular, it Our military is now in a rapid transformation to
simply refers to providing an alternative source become a lighter, more agile, mobile and an
of AC power on the battlefield that is generated extremely more lethal force. To accomplish this
from the vehicle itself, as opposed to hauling a many technical and logistic problems must be
large, heavy, diesel generator around on a trailer. solved; however, simply throwing money at a
What's the big deal with this concept you may problem doesn't always produce results. Even
ask. Well the military has been hauling today the old engineering adage, "the
generators around for decades and quite probability of obtaining an answer to an
successfully. When the enemy was confronted undefined problem, is zero" still applies.
at well defined lines and occupied a specific
theater of operation, that may have worked; but Within the realm of this "military transformation"
even then it was just barely adequate enough to problem definition there are opposing, or
accomplish the mission. divergent forces, i.e., to become more lethal and
better networked involves new weapons

Page 26


systems and more sophisticated electronic HMMWV's in operation by the joint military
systems, both of which demand more DC and services. The large majority of the rolling stock
AC power. How do you get more power?...ah of HMMWV's today contain a 28vdc electrical
yes...larger, heavier generators. That makes the system.
"lighter" and "more agile" components of the
"transformation" equation harder to achieve. True, there is a move to possibly increase this to
42vdc, 100+ vdc and beyond in new-to-market
The demand for fuel on the battlefield has models currently being designed and tested.
doubled in the past ten years and the demand However, a nice, clean, simple retrofit could
for tactical power has quadrupled during that result in an upgrade of the existing fleet to
same time period. include On-Board Vehicle Power in the range of
8 to 10 KW. While if we had our druthers we
Should that be a surprise to us? Well no. Isn't would want 30KW, it has been estimated that
it logical that one might conclude that the between 70 and 80 percent of the diesel
demands of the war effort have resulted in generators in the military inventory today are in
advanced technology, in the form of sophisti- the 10KW power range. So addressing this
cated electronic systems, improved weapons segment of the HMMWV installed base provides
systems, state-of-the-art field medical equipment an interim step on the way to the more powerful
and elaborate networked Tactical Operations systems planned for tomorrow.
Centers (TOC's). These all put a strain on the
power systems currently deployable to the By providing the HMMWV with a 7 to 8 KW
battlefield. In looking forward at what might be On-Board-Vehicle-Power system supplying
another 20 to 40 years of anti-terrorist, military continuous AC power, the need to trailer a diesel
engagement, its easy to say that the fuel and generator is reduced significantly. In addition,
power requirements on the battlefield will now the HMMWV can negotiate terrain that it
experience the same levels of increase over the could not previously engage while towing a
next ten years. generator thus leading back to the more agile
concept.
So how does this monumental problem get
solved? The best way is to break it up into What kind of AC power is required? Since the
smaller pieces. Identify the chunks that can be missions vary between the branches of service
addressed with today's technology and begin to and even within the same branch of service, the
deploy, not all, but some portion to the battlefield. AC power requirements vary considerably. In
Then as technology allows, focus on tomorrow’s many cases a simple conversion from the
solutions. vehicle's 28vdc source to single phase 120vac,
60hz is adequate. In that regard, a number of
The first thing that can be done to lead to companies are providing this inverter
tomorrow’s solutions is to focus current efforts technology; however, the demands for power on
on the largest installed base of wheeled the battlefield, more often than not, exceed this
vehicles, that being the different variants of level of sophistication. Therefore, the challenge

Page 27

is designing an On-Board Vehicle Power system As one reviews the specifications for an OBVP
that meets multiple mission demands, system, to be installed on 28 vdc vehicles, there
regardless of the branch of service. This is by no are certain performance minimums that become
means a minor task. There are simply times evident, as it relates to the Output Power, Output
when the little ole lady from the television Voltage, Input Voltage, Total Harmonic Distortion
commercials says, "Where's the beef?", i.e., and Output Frequency. Physical characteristics
120/208, 3 phase VAC. and mechanical dimensions tend to vary from
vendor to vendor such as length, width, height,
weight and efficiency (heat dissipation) of the
chassis.

Listing of parameter requirements:
Power Level
10kw peak,7 Kw continuous
Output Voltage
120/208 vac ( single phase and 3 phase )
Input Voltage
24 - 28 vdc
THD
Less than 5%
Extreme Power Requirements in the Military Output Frequency
50,60 or 400 hz (selectable)
Dimensions
Now, regardless of whether the soldier is Small as possible
needing single phase 120 vac or 3 phase, 208
Weight
vac, in Europe where 50hz is prevalent, or if he's Light as possible
on the flight line, or a radar installation where
power at 400hz is the frequency of choice, the Efficiency
OBVP system should be able to provide all of the High as possible
frequency options. With sophisticated commu-
nications, computers and other specialty Among the major challenges in developing a 10
electronic systems, Total Harmonic Distortion KW OBVP system, especially a 208 vac, 3
(THD) becomes important since harmonic phase version, from a vehicles 28 vdc electrical
distortion emitted by the AC power source can system, is the internal power dissipation of the
render this type of equipment inoperable. converter itself, without creating a heat problem
Therefore, a low THD is essential to completing within the enclosure. Therefore, cooling is a
the desired operation. major challenge, especially if the goal is to
conductive cool the system so as not to require
forced air through fans.

Page 28


To obtain the necessary 28 vdc, 400 amps the solutions, the optimum solution to the 28 vdc to
HMMWV must either already be equipped with a high ac voltage, multi-phase requirements of
400 amp alternator or be retrofitted with one. today's OBVP system involves a variant of these
While most HMMWV's today are equipped with existing technologies.
200 amp alternators, this is easily resolved since
the Army has 400 amp, retrofittable, alternators Through a series of events, that spanned a
in their inventory. Once the 28 vdc, 400 amp number of years of research in the field of
alternator power source is installed in On-Board Vehicle Power, Diversified Technology
conjunction with the 10 KW OBVP Power has introduced an OBVP system concept that
Converter system, there is the problem of meets the specifications addressed in this
keeping the alternator running at an rpm that is article. The system includes a power conversion
adequate to supply the current to the Power electronics chassis that is shock-mounted inside
Converter system, which in turn will satisfy the a ruggedized, waterproof enclosure.
AC electrical load, as well as, provide a continu-
ous positive charge to the array of batteries. To
accomplish this, an engine speed control system
must become an integral part of the overall
OBVP solution by increasing the engine rpm's
as the output load demand increases. Likewise,
as the load is removed the engine rpm's must
return to idle.

Once the alternator, battery package and speed
control have been defined the actual design
concept or method of converting the 28 vdc to
120/208 vac (single phase or 3 phase), 50, 60,
or 400 hz selectable power can be addressed.
Some methods of solving the dc to ac conver- Power Conversion Electronics Unit, VPS-10K
sion involve dc-link or high-frequency ac-link
technology. The most common form of switch- Description of OBVP Application
mode dc to ac conversion is characterized by a The original system definition for DTI’s OBVP
dc link; however, the use of the dc-link has program targeted the creation of compact,
limitations imposed by the dc-link itself. Another mobile power generation to be used for the
method involves high-frequency ac-links that Sentinel Radar System in stationary mode
directly convert a higher frequency ac voltage to applications. As our development continued,
a lower frequency ac voltage. This involves the additional applications were realized due to the
rectification and impulse excitation of a low-pass system’s functionality and multiple mounting
filter with integral half-cycle sinusoidal or quasi- platforms. Mobile power became even more of
sinusoidal, voltages originating from a source of a desire therefore the ability to produce power
high-frequency ac voltage. While both represent on-the-move was incorporated in to the unit.

Page 29

More urgent applications arose for OBVP with now can operate under preferred mission
the recent aftermaths of hurricane Katrina and requirements without having to modify its
2008’s Gustav. These include the capability of mission due to limitations of equipment deploy-
supplying power to ravished areas in dire need ment. This provides the soldier with more
as it relates to first responders, health care options in accomplishing objectives without
facilities, water purification centers, communica- having to jeopardize safety based on limits
tions centers, gas / fueling centers, etc. Using imposed by the previous solutions. Flexibility,
the capabilities of the OBVP system, instant mobility, portability and durability were the main
power could be realized in any area where a points of design during development. With the
HMMWV could maneuver. National Guard first VPS-10K from DTI, those objectives are
responders would have the tools necessary to achieved.
provide instant relief to those needing it most.
The military hierarchy is moving to become a
More and more programs require additional lighter, more agile, mobile, networked, lethal
power generation capabilities. Future develop- force through a modernization effort within the
ment of the OBVP program will target higher Pentagon, known as "transformation". The
power generation capabilities. Using the 10KW profile of our enemy has changed from the more
system as the framework, higher power levels traditional adversaries to those threats imposed
can be achieved by utilizing additional DC by the far more elusive terrorist movement that is
generating techniques for tactical wheeled global in nature.
vehicles. The focus is to make this technology
available for already established vehicles This article is an effort to divide the technological
currently in use as well as for new programs challenges of providing tactical On-Board
under development as part of the vehicle Vehicle Power to the battlefield into levels of
integration package to power more advanced difficulty. Thereby resulting in a "real-time
technology for immediate requirements in the solution" for the 10kw and under systems, while
war against terrorism. acknowledging that there is intense research
being conducted to attain the higher power
The development work for DTI’s program demands of the future.
focused all its efforts in providing a new solution,
never before seen by the armed forces, and
supplying a modular approach of power genera-
tion in a smaller, smarter and durable concept.
This method of power generation provides the
capability of achieving power levels only seen in
the past by the utilization of a trailer towed
approach. The compact nature of this design
allows for installation that does not affect cargo Article written by:
space needed for mission requirements. Being Barrie McArthur
waterproof and not trailer towed, the HMMWV Vice Chairman Special Projects

Page 30

MOBILE POWER
For Mobile Forces On-Board Vehicle Power (OBVP)

What is OBVP? Three Phase and Single Phase Operation Available
OBVP provides 3-Phase AC Power for Rugged Environments for Mobile or Stationary Power

OBVP offers Electronic Power that is: •••••
• Solid State Electronics (No Moving Parts)
Hassle Free, Trailer Free
• Physically Lighter
7KW Continuous, 10KW Peak Power
• Smaller Dimensions
• Easily Mountable •••••
• More Efficient and More Reliable
• Easier to Operate and Maintain 50Hz, 60Hz and 400Hz Selectable
from One Power Inverter
•••••
- DTI’s VPS-10K is Now Being Deployed -
Minimizes Logistic Space Requirements
Aboard Aircraft or Watercraft
•••••
Simple Retro-fit for Vehicle Integration
and Multiple Mounting Options

•••••
Provides Power to Teams, Patrols, Convoys,
During Unexpected Delays

•••••
Offers Simplified Fuel Logistics for
Mission Planning and Deployment

•••••
Back-up Power for Mission-Critical Equipment

www.DTIRuggedPower.com

The Cohesive Approach
How to Achieve High Availability Load Balancing with HAProxy on ATCA

Defining Load Balancing Simply
Load Balancing is a fairly simple concept: work that would normally be performed by one system is dis-
tributed amongst many systems based on some predetermined criteria in a transparent fashion such that
the client or end user is unaware of the separation of services. This is not what is traditionally thought
of as Clustering, although Load Balancing is usually a feature of clusters. Typically load balancing is
used by Internet service provides to distribute the computational and bandwidth loads of applications
across many servers in a server farm.

In this type of setup, the load balancer distributes requests to the backend servers, who reply to the load
balancer. The load balancer is the “point of contact” to the client. The client or user cannot contact and
is unaware of the presence of the backend servers and the separation of the services amongst them:

Basic Load Balancing Diagram

Layer 4 and Load Balancing
When the term “Load Balancing” is used, what is typically meant is Layer 4/7 Load Balancing. This
refers to the fourth and seventh layers of the OSI Model of networking abstraction, called the Transport
Layer and Application Layer, respectively. In the OSI Model, the Transport Layer is where we find TCP
and UDP protocols, segmentation of the data stream, error correction, and flow control. TCP (Trans-
mission Control Protocol) provides, among other things, reliable end-to-end communication (is “con-
nection-oriented”) via virtual circuits, error detection and congestion avoidance at Layer 4, whereas
UDP (User/Universal Datagram Protocol) does not, and therefore has a much lower overhead than TCP.
UDP is used at Layer 4 for application services that do not require guaranteed delivery of datagrams.

Page 32


TCP and UDP use port numbers to keep track of multiple application connections to the same location,
essentially keeping track of connections to higher level protocols. What are called “well-known” ports
are reserved for 1 – 1023. For example, TCP uses port 80 when initializing HTTP connections to a
server, port(s) 20/21 for FTP connections, and UDP uses port 69 for TFTP (Trivial FTP) connections.
Some protocols may use both TCP and UDP to form a connection, but will only use one port, such as
DNS, which uses both TCP and UDP to transmit data but only uses port 53.

The Application Layer (Layer 7) interfaces directly with and defines application processes and services,
such as FTP, TFTP, HTTP, DNS, DHCP, SSH, SIP, and TELNET. The Application layer should not be
confused with desktop or user applications – these applications, such as a web browser, use the
processes and services of Layer 7, but they do not reside in this layer themselves.

Layer 4/7 load balancing involves making distribution decisions based on the port number of an in-
coming connection, which indicates the application level process or service that is using the transport
level protocol. For example, one may choose to load balance incoming connections to a web server.
In this case, the load balancer would be told something akin to “incoming connections to port 21, or
FTP connections, should be distributed among Servers 1, 2, and 3.” The load balancer then does just
that, transparently balancing the bandwidth and computational load of the user connection.

Basic Load Balancer Distribution to Backend Application Servers

There are various methods of connection scheduling within a load balancing implementation. The most
basic, and most common, is called Round Robin Load Balancing (which is what is pictured above). If
a fifth connection were to be made above, that connection would go to Server 1, a sixth would go to
Server 2, and so on. The connections are distributed in a round-robin fashion among the four applica-
tion servers. Some load balancing solutions may use destination or source hashing, shortest expected
delay scheduling, least-connection scheduling, or any of a variety of other methods for scheduling the
distribution of connections.

Page 33

AdvancedTCA

What is AdvancedTCA?
For high availability and high bandwidth needs in telecommunications, commercial and industrial sec-
tors, and in government/military applications, there is little doubt that AdvancedTCA (ATCA) presents one
of the most robust and powerful architectures available. Combining the performance and high availability
with the ease of scalability and upgrades, ATCA should be a strong contender for most applications in
such sectors.

For the uninitiated, AdvancedTCA is the largest specification effort in the history of the PCI Industrial
Computer Manufacturer’s Group (PICMG) and is targeted to requirements for next-generation com-
munications equipment, and AdvancedTCA is finding homes in military, commercial, and industrial ap-
plications. This series of specifications incorporates the latest trends in high speed interconnect
technologies, next-generation processors, and improved reliability, high availability, and ease of serv-
iceability and focuses on key benefits of a COTS (Commercial Off The Shelf) methodology.

ATCA was created with six key goals in mind:
• Enable interoperability among manufacturers’ components
• High Availability (.99999 Uptime)
• Reduce Time to Market/Revenue
• Reduce development costs
• Reduce the length of development cycles
• Create an open, standards-based scalable, flexible platform architecture

As of today, ATCA has achieved these goals. In using the ATCA architecture, organizations will achieve
not only a demonstrably reduced Time to Revenue, but lower overall development and deployment
costs compared to traditional 1U or 2U deployments and proprietary solutions. Interoperability is a hall-
mark of ATCA, allowing the choice of best-in-breed components for any application – preventing lock-
in with one vendor whose proprietary solution may not meet all of an application’s unique needs.

As a bladed architecture, ATCA offers many benefits over traditional 1U and 2U solutions, namely in re-
duced integration and cabling costs, as well as being extraordinarily scalable and flexible. A shelf can
be adapted to fulfill nearly any need via AMC expansion sites on most blades (or on a carrier blade) or
through specialized full-sized blades fitted for one specific purpose.

The achieved interoperability allows the selection of the best components to fit the needs of an appli-
cation. One vendor’s CPU node will work just fine with another’s switch in a third vendor’s chassis.
ATCA offers the freedom of choice, the freedom of flexibility, and the freedom from vendor-enforced up-
grades.

Page 34


AdvancedTCA Switching & Networking Basics
An ATCA system is designed around a redundant dual star topology of communication among blades
in a chassis. A typical 5-Slot chassis will house two switches and three computing nodes, as well as a
Shelf Manager (ShMM) for system-level management. The blades may have AMC or RTM capabilities
for board-level I/O expansion and to allow for customization of board and system capabilities.

A 5-Slot chassis fully loaded resembles the following:

5 Slot ATCA Chassis with 2 ShMMs,
2 Switches, and 3 CPU Nodes

As stated, the intra-chassis communication links are routed through the chassis backplane among the
boards in a redundant, dual-star topology. Since each blade has a “base” and a “fabric” interconnect,
each CPU node has two connections to each switch – one base interface and one fabric interface. The
Base interconnect is treated as a “control plane”, while the Fabric interconnect is called the “data plane.”
Some ATCA backplanes support a full mesh architecture, in which every node (any endpoint) is con-
nected to every other node. It is still possible to implement a simple dual star, but the expanded con-
nectivity is available.

Redundant Dual Star Topology
Page 35

In the above diagram, the red 1GbE connections are the base interconnects, while the blue 10G con-
nections represent the fabric interconnects. All of the shown connectivity is intra-chassis, meaning it is
routed through the backplane of the chassis itself; there is no external cabling needed to connect the
nodes and switches to one another. As one may guess, the above diagram, when implemented “naked,”
would result in a whole host of networking problems like loops, congestion, and broadcast storms. How-
ever, ATCA switches support a great many protocols to solve these issues, such as M/R/STP, Flow
Control, VLANs, QoS/CoS, etc.

Representation of Full Mesh Topology

Most ATCA switches are Layer 2 switches and support hundreds of protocols. This allows for clean, ef-
ficient, and effective networking at the Data Link layer of the OSI model and aides in developing flat net-
work design. Some ATCA switches, like DTI’s ATS1936, support many Layer 3 (Network Layer) routing
functions like RIP/v2, OSPF/v2, static routing, VLAN routing, VRRP, etc. While we still refer to these as
Layer 2 switches, they do have greatly expanded and robust routing functionality.

Even fewer switches support the aforementioned Layer 4 functionality that Load Balancing requires.
That is not to say they do not exist – one could find a very few ATCA switches that have Layer 4 Load
Balancing functionality integrated. However, these switches will typically break the bank when it comes
to affordability!

Page 36


The Set Up

Choosing the Basic Solution
Assuming an existing application utilizing AdvancedTCA with a switch that does not inherently support
Layer 4 load balancing, how could one take advantage of the nature of ATCA and the COTS method-
ology to deploy a Load Balancing solution?

There are three basic methods (in terms of hardware) that could be implemented. First, buy an exter-
nal load balancing appliance and force it as the connection between the external and ATCA networks.
Secondly, since most ATCA system resources are under-utilized, if there exists a blade that isn’t doing
much, or an extra slot in the chassis where a new blade could be inserted, we can develop a Linux-
based, High-Availability Load Balancing proxy server. Lastly, we could replace our switch with one that
supports Layer 4 Load Balancing, which seems to be an awful lot of money and an awful lot of time spent
all for just one feature!

Let’s assume that we have a 5 Slot chassis with Diversified Technology’s ATS1160 switch and two DTI
MPCBL0030 CPU nodes. The two CPU nodes are LAMPs – one is a live web server and the other is
our backup. We would like to implement a Layer 4 Load Balancing solution that would make both ex-
isting blades live servers that split the cost of bandwidth and usage, as our website has seen a drastic
increase in bandwidth and computational usage, maybe because our marketing and advertising de-
partment has launched a new campaign focused on driving traffic to our website, and it has been wildly
successful.

Since we can get another MPCBL0030 for about one-half to one-third of the cost of a Layer 4 switch,
we decide that the second method mentioned above (developing a high availability load balancing proxy
server) is the best method, in terms of a stop-gap, until we have the time and the budget to fully evalu-
ate and deploy a true, hardware-based load balancing solution. Once our board arrives, we put it in our
chassis and prepare to develop our load balancing solution:

Targa-5 with 3 MPCBL0030 Nodes and 1 ATS1160 Switch

Page 37

The Load Balancing Software (HAProxy)
For our load balancer, we’re going to be setting up HAProxy (http://haproxy.1wt.eu/), which focuses on
high availability load balancing and automates failover on TCP and HTTP-based applications.

The documented featureset is as follows:
• Able to route HTTP requests depending on statically assigned cookies
• Spread the load among several servers while assuring server persistence through the use of
HTTP cookies
• Switch to backup servers in the event a main one fails
• Accept connections to special ports dedicated to service monitoring
• Stop accepting connections without breaking existing ones
• Add/modify/delete HTTP headers both ways
• Block requests matching a particular pattern
• Hold clients to the right application server depending on application cookies
• Report detailed status as HTML pages to authenticated users from an URI intercepted
from the application.

It seems HAProxy will meet our needs quite nicely, and offers a few very beneficial features, such as
reporting, Layer 7 persistence, and automated failover and recovery.

The Operating System
First we must take stock of where we stand with our current deployment. Web Server 1 (WS1) is our
live server, Web Server 2 (WS2) is our backup, and the new board will be our High Availability Load Bal-
ancer (HALB).

HALB needs an OS, and in this case we’re using Linux. I like CentOS 5.2, so that’s what we’re going
to go with. Since CentOS is a derivative of RHEL, the methods of deployment on RHx or Fedora should
be very similar. Other distributions can do what we want, but we’re going for enterprise-level high avail-
ability, so CentOS it is.

Serial Redirection and OS Install
We will be running and operating HALB via serial console, so we need to install CentOS 5.2 via serial
console, as well. A key word of advice: if you are going to be interacting with the board via serial redi-
rection, install your OS via serial redirection – it will make your life much easier. In order to begin we
need an RJ45 cable and a rolled-null serial adapter, which we will connect at one end to the console port
of HALB (the one marked IOIO), and the serial end to our laptop’s serial port. Open your favorite ter-
minal program (I use TeraTerm), and set it to connect via COMx (whichever COM is the one with the se-
rial adapter connected). We need to set our baud rate to 115200, data to 8 bits, no parity, 1 stop bit,
no flow control (115200, 8n1).

Page 38


Baud Rate: 115200
Data: 8 bits
Parity: None
Stop: 1 bit
Flow Control: None

Now, we could boot/install via network if we desired, but let’s just say (to make things simple) that we
have an external USB-based DVD-ROM drive and a CentOS 5.2 DVD. We connect this to the USB port
of HALB, insert our CentOS DVD, and engage HALB with the chassis (shutting the Hot Swap handle
completely). Once the Shelf Manager gives the ‘OK’, HALB will boot and we’ll get output through the
console into our terminal program.

If no text is present past POST, it may be necessary to enter the BIOS (press DEL on POST) and en-
able console/serial redirection.

At the CentOS install screen, there are installation options and the ability to input boot parameters.
Since we’re installing via serial console, we must do a text install, and there is a very important boot pa-
rameter we must use: console=ttyS0,<baud rate>. This must be used as a boot parameter, and must
be placed before the word “text” when installing in text mode:

**********************************************
* Press Enter to install graphically
* Type linux text to install in text mode
* Press the Function Keys for more options
* Enter boot parameters below
**********************************************
boot: linux console=ttyS0,115200 text

Installation is fairly straightforward from here out. There are a few choices for packages to install, and I
recommend not installing the Virtualization or Clustering packages unless you really, really need them.
The Virtualization package is simply Linux Virtual Server and Xen, which have been integrated into the
kernel, and the Clustering package is OpenMosix and a few other applications. It is always possible to
install these at a later time should they be necessary.

When asked to set up networking, eth0 should be the default active port, assuming your connection to
the external network is through the base interface of your switch, rather than through a front panel Eth-
ernet port on the MPCBL0030. If your external connection is through the fabric channel of the switch,
eth3 should be the default. Set your IP address and other parameters as required.

Page 39

Once the OS has installed and you have rebooted (it will force reboot in order to complete setup), I like
to immediately update and upgrade everything.

To Note: While CentOS is an RPM based distribution, if you’d rather use APT than Yum, the process
to do so is:

1. Download the proper release package for your hardware from:
http://dag.wieers.com/packages/rpmforge-release/

2. Install the package:
[root@HALB ~]# rpm –i rpmforge-release-0.3.6-1.el4.rf.*.rpm

3. Now perform
[root@HALB ~]# yum install apt

4. Now update /etc/apt/sources.list.d/os.list and add the following lines:
***CentOS Entries***
repomd http://mirror.centos.org/ centos/$(VERSION)/os/$(ARCH)
repomd http://mirror.centos.org/ centos/$(VERSION)/updates/$(ARCH)
repomd http://mirror.centos.org/ centos/$(VERSION)/extras/$(ARCH)

5. Now you can use apt-get. Try it out
[root@HALB ~]# apt-get update
[root@HALB ~]# apt-get upgrade

HAProxy Installation and Setup

Remaining House Cleaning
As stated HAProxy is a high availability TCP/HTTP load balancer. Since the concept application is an
HTTP/80 load balancer, we’re going to need a free Ethernet interface. What is meant by this is that we
cannot have another service bound to the main Ethernet interface (such as Apache) or the TCP port
we’re going to use of it’s IP address.

So, there are still a few things to do before tackling installing HAProxy. First, we need to stop Apache
(if it is running) and make sure it will not return on system restart:

Page 40


[root@HALB ~]# service httpd stop
Stopping httpd: [ OK ]
[root@HALB ~]#
[root@HALB ~]# chkconfig httpd off

Apache has been stopped and will not initialize if there is a system reboot. It should be noted that we
want to leave Apache up and running on WS1 and WS2.

Sometimes there are issues with Kudzu, the automatic hardware probing mechanism in RHEL/Cen-
tOS. I generally turn it off at this point, as I won’t be adding any hardware to the board in question.

[root@HALB ~]# chkconfig kudzu off

HAProxy Archive and Installation
We need to download the HAProxy executable:

[root@HALB ~]# cd ../usr/src
[root@HALB ~]# wget http://haproxy.1wt.eu/download/1.2/src/haproxy-1.2.18.tar.gz

--13:50:22--http://haproxy.1wt.eu/download/1.2/src/haproxy-1.2.18.tar.gz
Resolving haproxy.1wt.eu... 88.191.22.98
Connecting to haproxy.1wt.eu|88.191.22.98|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 206219 (201K) [application/x-tar]
Saving to: `haproxy-1.2.18.tar.gz'

100%[=======================================>] 206,219 15.3K/s in 11s

13:50:33 (18.9 KB/s) - `haproxy-1.2.18.tar.gz' saved [206219/206219]

Now we untar, make, and set it as a service to start on boot:

[root@HALB src]# tar -zxvf haproxy-1.2.18.tar.gz
[root@HALB src]# cd haproxy-1.2.18
[root@HALB haproxy-1.2.18]# make target=linux26
[root@HALB haproxy-1.2.18]# cp haproxy usr/sbin/haproxy
[root@HALB haproxy-1.2.18]# wget http://www.dtims.com/haproxy/ haproxy.init -O /etc/init.d/haproxy
[root@HALB haproxy-1.2.18]# chmod +x /etc/init.d/haproxy
[root@HALB haproxy-1.2.18]# chkconfig haproxy on
[root@HALB haproxy-1.2.18]#

Page 41

HAProxy Configuration
HAProxy should be generally ready to go. All that remains is to edit the configuration file to meet the
needs of our concept application and to actually start the service.
The HAProxy configuration file is stored at etc/haproxy.cfg
If the file does not exist, you can download an existing configuration file from Diversified Technology,
Inc.’s website at
http://www.diversifiedtechnology.com/haproxy/haproxy.cfg

Overviews of HAProxy’s settings and configuration parameters are in the configuration documentation:
http://haproxy.1wt.eu/download/1.3/doc/configuration.txt

Within the HAProxy configuration file, there are 3 master sections: Global, Defaults, and Listen. Listed
below are the values in the default configuration file on DTI’s website, with explanations for settings:

global
log 127.0.0.1 local0 info Logging
maxconn 4096 Maximum Number of connections on the front end
daemon Background process forking, allows nbproc parameter
nbproc 4 Number of processing cores on the blade. The MPCBL0030 has
2 dual core processors, making 4 cores total.
user root
#debug Commented. Enable for debugging
#quiet Commented. Enable for debugging

defaults
log global enables per instance logging of events and traffic
mode http set running mode or protocol
option httplog enable logging of HTTP request, session state, and timers
option dontlognull disable logging of null connections
retries 10
option redispatch enables session redistribution if there is connection failure.
requires a non-zero value for retries parameter.
maxconn 2000 maximum per-process number of concurrent connections
contimeout 5000 maximum time to wait for a connection attempt to a server to succeed
clitimeout 50000 maximum client side inactivity time
srvtimeout 50000 maximum inactivity time on the server side

Page 42


listen
listen webfarm <IP of HALB eth3>:80 tells HAProxy to bind to port 80 of the eth3 IP address
mode http set running mode or protocol
balance roundrobin set the methodology of load balancing
server WS1 <IP of WS1 eth3>:80 weight 1 these two arguments tell HAProxy where to send the
traffic and in what proportion (weight)
server WS2 <IP of WS2 eth3>:80 weight 1

Here is where we need the IP address of HALB, WS1 and WS2. Since this is AdvancedTCA, we have
about 6 Ethernet ports on each board to choose from! If we’ve architected our chassis properly, how-
ever, eth0 or eth3 should be the only ports whose IP addresses are of concern, presently. If our servers
are going to be serving via the Base channel, eth0 is our port, and if we’re serving on fabric, eth3 should
be our port. Let’s assume we’re serving on fabric—we need the IP addresses of eth3 on HALB, on
WS1, and on WS2. Let’s imagine our current addressing looks like:

HALB: eth3 10.10.1.2 netmask 255.255.254.0
WS1: eth3 10.10.1.3 netmask 255.255.254.0
WS2: eth3 10.10.1.4 netmask 255.255.254.0

Given this addressing scheme, in the listen section of the haproxy.cfg file, our arguments that require
IP addresses will be as follows:

listen webfarm 10.10.1.2:80
...
server WS1 10.10.1.3:80 weight 1
server WS2 10.10.1.4:80 weight 1

Once completed, save the configuration and start HAProxy:

[root@HALB ~]# service haproxy start
Starting haproxy: [ OK ]

Now all that is left to do is to reconfigure your settings such that your URL points to the eth3 interface
of HALB instead of WS1. In doing so, we have constructed a TCP/HTTP round-robin load balancing so-
lution in relatively short order.

Page 43

Our final setup resembles the following diagram:

Conclusion
HAProxy is a very simple and lightweight tool that can be used to quickly deploy TCP/HTTP load bal-
ancing solutions. While it can be deployed on nearly any hardware, AdvancedTCA can present a prob-
lem for those not very familiar with ATCA networking and configuration. HAProxy offers far more options,
arguments, and parameters than those discussed in the configuration file above – it is far deeper than
I expected at first glance, and I’m sure there are many who have far more advanced configurations than
the simple one I have put together above.

While the above setup is simple, it will work and work well, costing far less than purchasing dedicated
load balancing appliances or new ATCA switches with Layer 4 Load Balancing integrated into the stack.
While a hardware solution will be more reliable in the long run, HAProxy is, by its very name, made for
high availability uptime and should be a very good holdover until such time that a hardware solution can
be deployed.

Article Written by:
Doug Mays
Field Application Engineer

Page 44


About Diversified Technology, Inc.
Diversified Technology, Inc. (DTI) is an embedded hardware company whose strength lies in the
cohesive approach we use with our customers. This cohesive approach means DTI works hand-in-hand
with customers to ensure they are getting the best performance, highest reliability, shortest
time-to-market and the most efficient use of computing hardware for their program's embedded
application.

DTI, an Ergon Company, was founded in 1971 and has a history of design experience with PICMG
standards. DTI has been prolific in introducing technology and standard off-the-shelf products based on
PCI/ISA, COM Express, CompactPCI, AdvancedTCA and other standardized form factors. In addition
to COTS solutions, DTI is also a leader in the custom development of advanced computing boards and
embedded systems for unique, application-specific designs. DTI single board and system products are
pointed toward mission critical applications such as telecommunications, factory automation,
networking, ruggedized military, high-end gaming and medical control operations.

Joe Speak
by Joe McDevitt, Chief Technology Officer
Many internal and external people have asked me why I ran for office to be PICMG’s VP of Marketing
and I can’t help but to think of the movie “A Beautiful Mind” staring Russell Crowe as John Nash, a
mathematical genius, eventual Nobel Prize winner, and sufferer of paranoid schizophrenia. Late in his
college career, Nash grows increasingly frustrated as he seeks in vain for his revolutionary idea. Nash’s
“A-Ha” moment — as the movie purports — happens one night while drinking beers with his fellow
mathematician students. While enjoying tall, frosted beers, five girls enter the bar. The sole blonde is
taller and more striking than the other four, and the mathematicians begin to discuss strategies for
securing a dance with the taller girl. They conclude that Adam Smith is correct and everyone needs to
work in one’s self interest, leaving the prettiest girl to decide who is the fittest. Nash ponders that
approach for a moment and then explains that if they all attempt to dance with the striking blonde, they
will block each other’s efforts and no one will succeed. Further more, the other girls will reject their
secondary efforts, as no one wants to be the “second fiddle.” In the movie Nash concludes that they
should all agree to completely ignore the blonde and approach the other girls first. Nash states that
Adam Smith’s theory is incomplete and that success comes not from everyone doing what is best for
themselves alone, but in doing what is best for themselves and the group. PICMG is about those
competitive developers gathering and agreeing to work together, and thereby attracting customers into
an ecosystem - much like Nash stated and is the main reason why I ran.

Page 45

DTI Conference Schedule
MS National Guard RTECC
Natchez, MS Montreal, QC
May 1 to 2, 2009 June 2, 2009

NGB Logistics RTECC
Keystone, CO Ottawa, ON
May 4 to 8, 2009 June 4, 2009

WINDPOWER 2009 MS Chiefs of Police
Chicago, IL Biloxi, MS
May 4 to 7, 2009 June 8 to 11, 2009

RTECC National Sheriff Association
Greenbelt, MD Ft. Lauderdale, FL
May 5, 2009 June 22 to 23, 2009

RTECC SMD Conference
Boston, MA Huntsville, AL
May 7, 2009 August 17 to 20, 2009

Border Security DSEi 2009
Phoenix, AZ London, UK
May 13 to 14, 2009 September 8 to 11, 2009

RTECC NGAUS Conference
Phoenix, AZ Nashville, TN
May 19, 2009 September 11 to 13, 2009

RTECC RTECC
Fort Huachuca, AZ San Diego, CA
May 21, 2009 September 29, 2009

Interop Modern Day Marine
Las Vegas, NV Quantico, VA
May 19 to 21, 2009 September 29 to October 1, 2009

Fires Conference RTECC
Ft. Sills, OK Long Beach, CA
June 1 to 4, 2009 October 1, 2009

Page 46

DTI EMBEDDED ROADMAP
LBC9426 LBC9526 LBC9626
Dual-Core Xeon / Dual CPU Future Dual CPU/Chipset Future Dual CPU/Chipset
PICMG 1.0

LBC9316 LBC9416 LBC9516
Core 2 Duo Future CPU/Chipset Future CPU/Chipset

LBC9216
Pentium 4 or Celeron

TrexSys Platforms
1U, 2U and 4U Systems

CPB4612 CPB4912
CompactPCI

Pentium M Future CPU/Chipset

CPB4712 CPS4224
Dual-Core Xeon Switch, 10G Egress (4),
19 Slot Chassis
CPB4812
Core 2 Duo

PlexSys Platforms
1U, 2U, 4U Systems

VPX CPU VPX CPU
Intel Chipset/CPU Future Chipset/CPU

VPX Switch
VME

ATC6231/ATC6239 ATC7239 ATC8239
Dual and Quad-Core Future CPU/Chipset Future CPU/Chipset

ATS1936 ATS1930 ATS7160
ATCA

10G Switch AMC (3) Carrier, AMC (2), Base GigE,
PCIe and 10 GigE Fabric 10G Serial
AMC & RTM Boards
Application Specific ATI1914
AMC (1), 10G to 1G
Targa Platforms Expansion Board
5 and 14 Slot Systems

CMXLX45 CMXLX55
Small Form

COM Express Board Future CPU/Chipset
Factors

ETXLX15
ETX Board

ETXM690
ETX Board

2009 2010 2011 2012 2013

*DTI’s Product Roadmap is subject to change. Please contact Diversified Technology, Inc. for a current revision.
sales@dtims.com | 1.800.443.2667 | 476 Highland Colony Parkway, Ridgeland, MS 39157

1. 8 0 0 .4 4 3 .2 6 6 7

w w w. D i v e rs i fied Te c h n o lo g y.c o m

Diversified Technology, Inc. US Postage
PAID
476 Highland Colony Parkway
Jackson, MS
Ridgeland, MS 39157 Permit #134

DTI Custom Publication, 2009

More Related Content

What's hot (18)

Viewers also liked (7)

Similar to DTI Custom Publication, 2009 (20)

DTI Custom Publication, 2009