|
To
view this update as a Web page, copy this link into your
browser:
http://www.qtsnet.com/stayinformed/quiknews/quiknews
april2005.htm.
To
subscribe or unsubscribe, please follow instructions at
the bottom of this page.
Welcome to the April edition of QTS QuikNews, our
monthly e-mail newsletter. In this monthly e-mail, you
will receive an update of what's new at QTS - new
products we support, new patches and upgrades, solution
ideas and promotions to save you money, and information
about our company and our clients.
In this issue:
-
QTS and Partner News
-
Events
-
President's Corner
-
QuikSecure Tip of the Month
-
Patches and Upgrades
-
Product Support Lifecycle Watch
-
Solution Spotlight
-
Special Offers
-
Partner Spotlight
QTS AND
PARTNER NEWS
WELCOME TO NEW CUSTOMERS
QTS offers a “welcome aboard” to the following new
customers:
·
Ballet Makers
QTS PRESIDENT TO INTRODUCE MICROSOFT VP AT WORLDWIDE
PARTNER CONFERENCE
QTS is pleased to announce that Neil Rosenberg will be
introducing Microsoft’s Vice-President of Server and
Tools Marketing, Andy Lees, for his keynote presentation
at Microsoft’s Worldwide Partner Conference this July at
Minneapolis, MN.
QTS IS HIRING!
QTS is recruiting for senior level technical personnel,
as well as an entry to mid-level sales/service support
specialist. If you know anyone who might be a good fit,
please have them submit their resume to Liz Meechan, our
Office Manager. Liz can be reached at
lmeechan@QTSnet.com, or (973)984-7600 x223.
MICROSOFT CASE STUDIES
MICROSOFT PUBLISHES CASE STUDY FOR QTS WINDOWS/EXCHANGE
2003 MIGRATION AT CHRISTIAN HEALTHCARE CENTER
Healthcare Center Improves Security and Performance with
Network Upgrade.
For
the Case Study, visit
http://www.microsoft.com/resources/casestudies/
CaseStudy.asp?CaseStudyID=16544
QTS
SUCCESS STORIES
AFS INTERCULTURAL PROGRAMS - USA
QTS
implemented Citrix MetaFrame
and Microsoft Windows Terminal Services to provide AFS
with secure remote access and to reduce the effort in
deploying and managing applications for remote office
users and volunteers.
http://www.qtsnet.com/about/success stories/afs
success story.pdf
Back
to Top
EVENTS
QTS
and Microsoft
are hosting a seminar entitled Why Renew: A Review
of Benefits, Available Promotions and Microsoft Product
Roadmaps and Key Initiatives. Join
us for this informative, 2 hour event to learn why your
organization should renew or upgrade into Microsoft
Software Assurance through the Open License Value
program. Microsoft and QTS will review the benefits of
Software Assurance (it’s more than just new version
upgrades!), discuss available promotions that can save
you money, and review the roadmap for upcoming product
releases and features. We will also present on key
Microsoft initiatives such as Trustworthy Computing
(Microsoft’s security efforts) and the integration of
Microsoft’s Business Solutions offerings. This event is
focused on customers deciding whether to maintain their
Software Assurance coverage, but will provide valuable
information to any Microsoft customer! Join us for lunch
on Wednesday May 4th at the High Lawn
Pavillion in West Orange,
NJ. For more information or to register, contact your
QTS Account Manager, or click
here.
QTS,
AKA Enterprise Solutions and Microsoft are hosting a
CIO Luncheon on Tuesday, May 24th. This
is a lunchtime event at the C-Side Café in Jersey City,
NJ (adjacent to the Exchange Place PATH station) is for
IT Directors, CIOs and CFOs
in organizations with 200 or more users. The focus of
this strategic event is on Microsoft’s efforts around
systems management (the Dynamic Systems Initiative) and
desktop integration of Office with Microsoft’s back-end
systems (Microsoft Business Portal and Information
Bridge Framework). For more information or to register,
contact your QTS Account Manager or click
here.
QTS,
Econium,
Softmart, New Horizons and
Microsoft are hosting two half day seminars on Tuesday,
June 7th at New Horizons in Iselin, NJ. In
the morning, session, QTS will present on improving
Operational Efficiency and Productivity with Windows
Server 2003 and Exchange Server 2003, as well as discuss
management solutions. In the afternoon session,
Econium will present on how
to improve employee productivity with collaborative
solutions including SharePoint,
InfoPath and Live Communications Server, while QTS
presents on Security and Secure Remote Access.
Softmart and New Horizons
will discuss Software Assurance benefits, including
training vouchers, at both events. For more information
or to register, contact your QTS Account Manager or
visit
www.clicktoattend.com and enter event code 102552
for the
morning session (QTS), and 102554 for the
afternoon session (Econium/QTS).
Back
to Top
PRESIDENT'S CORNER
Last
month, our column started to review Ten Steps to Build a
Highly Available Network, and covered the first three
steps. To review that column, click
here. In this column we’ll cover the next four
steps, and next month we’ll address the remaining three
steps.
4.
Incorporate Server Hardware Fault Tolerance
For
your servers, the most likely component to fail is your
disk subsystems. First, design all of your servers with
redundancy in your disk drives – ideally hardware RAID 1
or RAID 5 or greater depending on the amount of data and
on your performance requirements. Second, many vendors
(including HP – our standard) offer an “online spare”
function so that if a drive in a RAID Array fails, the
data from the failed drive is automatically written to a
backup drive that’s ready and waiting. You trade off
some capacity here for higher reliability and better
performance in the event a failure occurs. At many of
our customers with our standard configuration, we’ve
seen drive failures result in no downtime, and no
performance loss, because we were able to simply unplug
the failed drive and plug in a blank one, and users
never knew any better. Also, in larger networks keep a
shelf spare of each model of hard disk, as well as other
common spare parts, to minimize risk of downtime.
Another trick we use in medium to larger environments is
to order consistently in server models, even if it
involves some overspending. For example, if a customer
is going to need 8 servers, we might order 3 HP DL380s
(medium disk capacity) and 5 DL360s (low disk capacity,
lesser cost and less rack space). However, we might
also order 8 (or 9) DL380s, so if a system board fails,
we can pull replacement parts from a spare server, or
from a lower priority server (for example, a third
server in a Citrix server farm) temporarily to get the
high priority server back up and running. If your
server models are all different, you don’t have this
flexibility. There’s obviously a higher cost to this
approach, but it’s like insurance – very worthwhile when
you need it.
5.
Maximize Server Fault Tolerance
If
these approaches are not enough, then it is time to look
at server clustering or load balancing. In a clustering
scenario, multiple servers work together as a team,
providing access to data on a shared storage device
(usually a SAN – Storage Area Network) – so if one
server fails, the others keep going and users can work
uninterrupted. Microsoft has built solid clustering
support into Windows Server 2003 Enterprise Edition, and
we’ve done this numerous times for our customers –
particularly around Exchange since email is so critical
to business these days. However, the same type of
approach can be taken for database servers such as SQL
Server, and for file/print services. In a clustered
configuration, users point to a virtualized server name
and IP address, and the physical servers all work
together to handle service requests to that address.
Clusters can typically be Active-Active, where all the
servers work together, or Active-Passive, where a server
sits back and waits for the primary server to fail. All
of this needs to be properly designed and architected
based on the applications, the traffic load, and the
needs of the business.
For
situations where the servers are not hosting data,
setting up “farms” of servers in a load balanced
configuration works well, whether it is a “web farm” of
servers running Network Load Balancing under Windows
Server 2003 (this is now part of Standard Edition), a
Citrix MetaFrame Server
Farm, or a “cluster” of firewalls (sorry – people use
terms way too interchangeably in our industry). Very
often, clustering and load balancing are complementary
technologies – for example, you might set up a load
balanced web server farm in your security DMZ hosting
static web pages for an application, and set up round
robin DNS to make those web servers look like one site –
with the ability to add more and more identical servers
as performance and scalability needs dictate. Behind
the web farm, on your internal network, may be a cluster
of SQL database servers, connected via
Fibre Channel to a SAN
hosting the database. In any case, the key is
identifying and where appropriate eliminating single
points of failure in key network resources.
6.
Replicate Data for Disaster Recovery
Server failover can be taken to a further level by
replicating data from one physical location to another.
This is typically implemented as part of a Disaster
Recovery Plan (see next months’ final part of this
column), and there are a number of vendors providing
software solutions for this including
NSI’s
DoubleTake (which is also
OEM’d by HP and other hardware vendors),
NeverFail and others. The
key here is defining the architecture, since replication
can consume precious WAN bandwidth. Do
your WAN or VPN links have
enough bandwidth for your normal traffic as well as
replication traffic? How do these products handle
replication and synchronization? Security also needs to
be considered – do you want critical data replicated
over the Internet? Some platforms offer hardware level
replication – typically SAN to SAN – and cut out the
software layer. This may or may not be the right
solution. I would observe that clustering and
replication are two distinct solutions to distinct
problems – Clustering is to ensure maximum availability
of resources at your site, while replication is to deal
with the site being unavailable. Often people think
replication solutions make clustering unnecessary – but
keep in mind that a replication failover may be far too
complex to deal with when a simple server outage
occurs. They are in fact different solutions to
different problems. And neither may be justified in a
given case – for example, if all you need is file system
replication for a moderate amount of data,
then Windows Server 2003’s
DFS (Distributed File System) may be all you need to
address that problem.
7.
Utilize Server Management Tools
Server and network management tools should be utilized
to proactively ensure uptime. This can range from the
basic steps, such as installing HP (formerly Compaq)
Insight Manager or Dell OpenManage
agents on your servers and setting up a console, to the
more advanced such as setting up a
multiproduct management platform such as HP
OpenView, CA
Unicenter or Microsoft
Operations Manager. We recommend at the very least
leveraging Insight Manager or
OpenManage, so you can proactively see when
hardware problems are developing (usually disk drive
failures) and can take action before downtime occurs.
However, Microsoft Operations Manager (MOM) is a great
in-between step for heavy Microsoft shops that want to
build a proactive management system but aren’t prepared
for the investment required with the larger platforms.
If your environment justifies a more elaborate solution,
then look at the ROI – a few hours of downtime is
usually all it takes to justify an appropriately scaled
investment here. However, one point of caution – no
investment in this area will be beneficial if the IT
team doesn’t use the solution – this is a tool, not a
magic bullet, and as such it needs to be properly
integrated into the IT Department workflow and
procedures, and people need to be properly trained and
committed to use it. Anything less is a waste of
money.
Next
month, we’ll explore the final 3 areas where you can
take steps to maximize the uptime and availability of
your network and its applications and data.
As
always, feel free to email me your comments or thoughts
at
nrosenberg@QTSnet.com. Thank you.
Neil Rosenberg
President & CEO
Quality Technology Solutions
Back
to Top
PARTNER SPOTLIGHT
This
month QTS is spotlighting its knowledge
management and applications development partner,
Econium.
Econium is a Gold-Certified
Microsoft Solutions Provider that builds technology
solutions that promote information sharing, distribution
and collaboration across teams of people.
Econium delivers real time
collaborative solutions in web conferencing, application
sharing and content distribution, high quality video
conferencing, telephony integration with phone systems,
portals and presence information to help clients reduce
travel costs and improve team communications. The end
result is a more agile and effective organization. They
have delivered vertical solutions for sales and
marketing teams around brand portals and digital asset
management, and have provided portals and repositories
for healthcare companies, consumer packaged goods and
government & not-for-profit organizations.
Econium
is a four year old firm that is made up of seasoned
business technology experts who focus heavily on
understanding the business processes and needs
initially, then applying solutions that will leverage
existing technology investments. They work with a
variety of collaborative technologies including
Microsoft SharePoint Portal
Server and Windows SharePoint
Services, Live Conferencing Server and Live Meeting
Server, Office System 2003, Exchange, SQL Server and
Content Management Server. Their clients include Kraft
Foods, Pfizer Pharmaceuticals and Adams Candy.
Econium
is headquartered in Northern New Jersey and also has an
office in New York City.
Econium
relies on QTS to create the technical infrastructure on
which its solutions run. Econium
develops and deploys solutions which are geared toward
lowering the costs for companies to manage, distribute
and share the information they need to conduct business.
For more information on Econium
please view their web site at
www.econium.com, email them at
sales@econium.com, or call 973-812-9691. Or,
contact your QTS Account Manager.
Back
to Top |
