By John Mark V. Tuazon
September 1, 2010

Back in the day, Ergie Ong, CIO of PJ Lhuillier Group of Companies (PJL), a national network of more than 1,200 Cebuana Lhuillier pawnshops and remittance centers, recalls that the network requirements of companies are tied to the bare minimum, owing to simple applications being developed for the enterprise, coupled with the fact that telecommunications companies are unable to provide high bandwidth connection to end-users.

But as developments in technology naturally advanced, the demands for a higher-bandwidth network infrastructure became apparent, due to various business applications being developed to help streamline firms’ business processes. “Applications are proliferating and the network is critical for delivering these to business users,” explains Bernie Trudel, technology lead for the data center at Cisco, APAC.

One such application is video, which, according to the latest Visual Networking Index (VNI) conducted by Cisco, will exceed 91% of total global consumer Internet traffic by 2014. Videoconferencing, a tool starting to gain traction within the walls of the enterprise, is set to grow sevenfold within the same time period, the report says.

“Aside from the traditional office automation and businesscritical applications, we are seeing much more voice and video traffic as organizations are increasing their use of media-rich collaboration and social networking applications,” Trudel relates.

FINE-TUNING APPLICATIONS

While large corporations are tripping over themselves in the pursuit of deploying the latest technologies to supposedly improve their business processes, Ong of the PJ Lhuillier Group says they are focusing their strengths on developing fine-tuned applications that exactly meet their business requirements.

“We wanted to make sure that we maintain and pay for cost-effective network solutions,” Ong reiterates. “Given that, our system is based entirely on the minimum lines that we actually use.”

Ong’s development team, he relates, made it a point to minimize data transmission into the system by developing from the ground up a central application that supports and integrates all their products—which include pawning, remittance, insurance, bills payment and electronic loading systems, among others.

With the fine-tuned system, Ong claims they are able to process more than 10,000 of transactions per day using only a low bandwidth line, with that figure going up to about 50,000 transactions per day during peak seasons. Their branch system can process online transactions using dial-up lines when the primary line is down. “With our central in-house application, we can deploy new products in just one to two days, centrally updating it from our data center,” he adds.

The local office of First Advantage Corporation (FADV), a global information processing firm, lives by the same philosophy. Jan Espino, the company’s regional IT manager for Asia Pacific, shares that since their subsidiary’s focus is on employment screening for clients in the region, they only employ a core application that integrates all the functions needed by end-users.

“Our core application is hosted off of our data center in Bangalore, India, and connected using an international MPLS connection,” he explains. The core application developed inhouse automates the screening process, he goes on, and is accessed via a Web browser.

Their core application only uses half of their bandwidth, Espino points out, while the other half is relegated to accessing e-mail and Internet connectivity, which employees mainly use to augment their research materials.

ALLOTTING BANDWIDTH

Using light applications that are not heavy on network bandwidth use is one foolproof solution, but in the interconnected environment of the World Wide Web—where applications such as social networking, video and the like are gaining massive popularity—users are wont to make a significant dent on the network’s performance.

“Communication tools like video conferencing and email generates a steady stream of traffic, which no doubt consumes a significant amount of bandwidth,” says Desmond Toh, marketing director of network devices vendor DLink. “Having the risk of bandwidth competition, this may lead to poor and irregular application response time. These applications can easily saturate available bandwidth and starving other applications.”

Designating specific bandwidth allocations for missioncritical applications and light network requirements, therefore, becomes an all-too important task. PJL’s Ong, in fact, has taken this point to heart, banning the use of social media and other such bandwidth-hungry applications in their offices.

“Our managers asked if we can check if the employees are accessing the Web for work-related activities,” he recalls. “We found out that a lot of people are accessing Facebook, so we denied access to it right away.”

Upon banning the social networking site, Ong shares that there was a significant improvement in their network infrastructure, so they chose to stick with the policy. “But we allow access if users go through the approval process, which sees if the requirements to access the sites are valid,” he adds.

FADV’s Espino has limited the non-official use of the Internet in their company for productivity and security reasons. Among the sites included in their blocked list—which is managed through Active Directory—are video streaming sites, instant messaging applications, external emails, and social networking.

“We have an acceptable use policy which defines what acceptable Internet use is, to make sure users use it judiciously” he says. “We do not have wireless access points where users can just use their laptops with, and we have built firewalls to ensure security is not compromised through the network.”

Espino claims banning these sites do not affect the productivity of their employees, and that in fact, “it should even boost productivity because they are focused with their work.”

EAGLE’S EYE VIEW

Having policies in place ensure that critical business applications are serviced, but ensuring that they are complied with to the letter requires that network administrators monitor activities within their network, to gain visibility about bandwidth pain points within the infrastructure.

“The network management team imperatively requires tools that enable them to control the valuable network resources in order to ensure that each application receives its corresponding level of priority and the most effective use of bandwidth is made,” suggests Cisco’s Trudel.

D-Link’s Toh concurs: “It is important for the organization to have the ability to monitor traffic within the Local Area Network (LAN), the Wide Area Network (WAN), and certain points in the Internet,” he surmises.

In this scenario, the old saying that knowing the problem is half the solution becomes relevant. Without an effective monitoring tool, enterprises would rely solely on users abiding by their policies, or external influences respecting the private use of their infrastructure, which is not always the case.

“Organizations cannot afford their network uptime to be compromised. In order to achieve this, it is essential for companies that have large data networks to monitor network traffic and usage for more efficient troubleshooting when there are outages in the system,” Toh clarifies.

PLJ uses a freeware tool for monitoring their network traffic, Ong shares. “We use it to determine which of our branches go online and offline,” he explains.

Their proactive approach in using apps with small bandwidth requirements bodes well for them as well. “We tend to avoid high-bandwidth requirements, so we make sure our systems use minimal bandwidth,” Ong says, adding that they recently had to turn down a proposition from the management to centrally manage their branches’ CCTV cameras, saying it is not a practical application to use their network for.

D-Link’s Toh says there is wisdom in having a central management for the network, especially in large corporations, “because it is not feasible to execute system checking tasks manually as it is very time consuming.” “The application should be able to perform polling of network/devices status periodically. With the information gathered, they can address the issues accordingly,” he adds.

ALL GROUNDS COVERED

FADV’s Espino says it is best to approach network automation and management in a twopronged approach: employing the best-of-breed network devices that provide bandwidth throttlingand prioritization, among others, and, more importantly, managing the usage of end-users.

“No matter how sophisticated your system is, if users can just install applications, access Facebook, and the like, you will still face congestion,” he points out.

For PJL’s Ong, the best mindset is always expecting failure of the system, where contingencies are necessary. After all, no matter how advanced network technologies become, they are still prone to bogging down unexpectedly.

“One of our big advantages is having contingencies whenever our branches go offline,” Ong narrates. This involves employing mobile phones as alternative transmission gateways whenever systems go down. Branch employees would simply have to send text messages to the system in order to process transactions. Additionally, Ong says, they have developed a text-to-mail service in order to enable e-mail messaging during downtimes.

“Cases of systems going offline and downtimes are rampant especially in the Philippines,” he notes. “Having contingencies in place is important if online processing is critical to your business.”

Possibly Related Posts:

• IT Budgets on the Rise
• Veering into Virtualization
• BUSINESS AS USUAL: Picking up where you left
• IT Outsourcing Reaching Out to Reach Further
• Forecast 2012

 Print This Post