|
Today's highly connected organizations rely on vast amounts of data to
maintain their daily operations. Any failure of the communications network could
mean significant losses. For example, when employees and partners are unable to
access e-mail or company systems it can have serious effect on productivity.
Retailers stand to lose revenue if transaction information from point-of-sale
operations is not received. A disabled network also impacts production
schedules, causing loss of valuable time.
Although more difficult to quantify, communications failure can even damage a
company's public image. Customer perception of a business can be greatly
influenced by their experiences online. Shoppers can be lost forever if the
website is not working properly because competitive offerings are just a
mouse-click away.
Network downtime is the result of two types of failures.
First, whenever an Internet Service Provider (ISP) experiences problems and
the broadband connection goes down, their business customers can lose Internet
connectivity and valuable work time.
Second, even if a company still has Internet access, a network outage can
cause their virtual private network (VPN) to fail, leaving remote workers and
business partners unable to access central office resources. This can have
serious consequences if vendors or departments such as accounting or payroll are
unable to access data to complete end of quarter bookkeeping, for example.
In order to support daily operations and maintain business continuity, an
organization must be able to move vast amounts of data quickly, reliably and
securely. But the complexity of managing a comprehensive network infrastructure
is taxing IT departments like never before. Medium-sized companies, especially,
are feeling the pressure. These companies have the same tasks to perform as
larger organizations, but do not have the resources.
And it is IT managers that are held accountable for the consequences, even if
what occurred was entirely outside of their control. For this reason IT managers
are looking to implement “insurance policies” for their
organizations-namely integrated, robust security solutions capable of keeping
critical applications up and running at all times. IDC projects that spending on
business continuity solutions of this kind will increase between 2002 and 2007
to reach more than $118 bn.
Toward a Smart Solution
According to a 2003 survey conducted by SonicWALL, customers purchasing
business continuity solutions reported that network reliability is the number
one factor that influences buying decisions. Network reliability is achieved
using failover redundancy for all key components, thereby ensuring that no
single point of failure impacts network availability.
|
Best
Practices... |
|
... that organizations
of all sizes should incorporate in a number of core technologies
-
ISP failover: a
dual connection to the Internet either through two different ISPs, or
with two separate interfaces to two geographically dispersed locations
of the same ISP. Regardless of the approach, the result ensures an
automatic back-up if one line fails for any reason. By distributing
the risk, a company minimizes its vulnerability to a network outage.
-
VPN redundancy: allows
remote/branch offices and business partners to establish a VPN
connection to a secondary gateway at corporate headquarters if the
connection to the primary gateway fails. The switch from one VPN
tunnel to another should be transparent to employees, partners,
customers-and even to the network administrator. Maintaining a
continuous connection with the central office ensures that remote
employees and business partners can access the vital information they
need, when they need it.
-
WAN failover:
two different types of media can be involved. If the primary Internet
access connection is T1 or broadband then the secondary connection
might be a more economical alternative such as analog or ISDN. This
approach is often used by retail businesses with numerous POS
locations. An alternative is to use different service providers for
independent connections so that any problem encountered by one
provider does not affect other areas of business.
-
Load balancing: optimizes
both primary and secondary connections so that neither sits idle,
enhancing the network connectivity investment. The secondary
connection provides back-up insurance but also improves network
performance by sharing the traffic load. Cross-media redundancy
provides another type of load balancing. In this scenario, failover
capabilities utilize different types of Internet services.
-
Hardware
failover: to ensure network reliability, hardware failover
provides two components that serve the same function. In the case of
the active unit failing, the passive unit automatically detects and
assumes responsibility for forwarding traffic. This redundancy can be
achieved by deploying two identical routers or firewalls.
Alternatively, a company might opt for internal redundancy with two
WAN interfaces in the same box. Hardware failover can even be used in
“active-active” mode to enable load balancing, thus creating a
highly efficient method for distributing WAN traffic.
-
Stateful
synchronization: a feature that provides automatic failover to a
backup ISP if the primary connection goes down. The failover
transition is invisible to end-users and protects transactions in
process from being lost or corrupted. Stateful synchronization is of
especial importance to retail businesses.
In summary, the
technology to implement failsafe procedures for taking care of network
connectivity is today available to businesses of all sizes. Such solutions
should provide network reliability via failover redundancy; ease of use;
manageability/flexibility; enterprise-class features and functionality and
excellent price/performance. |
Generally speaking this is achieved by implementing redundancy at every
network connection point. At the hardware level, a network might have dual
routers or firewalls. A business may also choose to introduce some external
redundancy, such as two different ISP connections into the same router or
firewall.
A good business continuity solution for small and mid-sized networks needs to
provide a comprehensive yet cost-effective set of features. Solutions must scale
to meet any size of organization. Customers don't want to pay for more
bandwidth than necessary, yet they need a flexible solution that can keep pace
with their company's changing requirements.
The solution must also be quick to deploy and easy to administer, especially
if the organization has limited IT resources. Above all, the security system
should be easy to manage.
Products that have been developed for large enterprises are often offered to
small- and medium-sized businesses in “lite” formats. However, such
products are usually designed to offer comprehensive feature sets to major
enterprises. This means they often do not meet the needs of organizations with
limited resources that cannot afford to support them and have no need for many
of the features in any case.
Another approach is to customize a solution by cobbling together
best-in-class components from various vendors. The downside is that this
is often complicated, costly, and fraught with integration and maintenance
issues that IT managers generally prefer to avoid. The training requirements
alone can be prohibitive for a resource-limited organization. Ideally, a fully
integrated business continuity solution will scale to exact price/performance
specifications. A package that provides all of the necessary functionality-and
perhaps even some bonus capabilities for enhanced productivity- is more likely
to meet the IT manager's needs.
Shubhomoy Biswas
maildqindia@cybermedia.co.in
The author is country manager, SonicWALL India Page(s) 1
| 
