'Our Networks are the nerve center of our operations'

Scope International is the Global Shared Services Centre of

the Standard Chartered Group. The company was incorporated in 2000 and commenced

actual operations in Chennai from 2001. In a short span, Scope has grown to

become an organization with over 5,000 employees and constitutes about 8% of

Standard Chartered Group's global workforce. It services all 56 countries of

the Group and processes over 80 mn transactions a year in banking operations

alone. Scope handles a wide range of value-added and complex services from

banking operations to supporting global HR processes, finance and accounting

services, software development and maintenance, and providing IT service and

helpdesk support to the group globally. With these processes in the back drop,

the bank's connectivity and network infrastructure assumes a mission critical

role. Sarabjit Anand, head ITSC (W) and country technology manager India

and GSSC, Scope International, walks through the company's networks in an

interview with Shrikanth G of Dataquest. Excerpts

How critical are the Networks for you?



A few years back we decided to move towards a network centric model as part

of our strategic vision. The initiative was started three years back, and we

moved our core and critical application to our data center. The network centric

model called for strategic and well defined networks for seamless operations.

For instance, from a group perspective, we established 10 core sites spread

across countries like India, Hong Kong, Singapore, Malaysia, UK, UAE among

others. With this model in the backdrop, our network needs to be up and running

always to meet the stringent requirements of Standard Chartered Bank's (SCB)

operations. So in that perspective, our networks play indeed a mission critical

role as all our apps sit in the centralized data center.

How is the model defined?



Our networks are defined in three layers. For instance, the first layer is
the core, and then we come to the second layer that is distribution. The third

one is the access layer. The ten sites that form the core are fully redundant

MPLS based network. Then we come to distribution and when we take one of the

core sites like India- within it we have two sites- Chennai and Mumbai, these

are next level critical sites that drive the banking operation at the country

level. So all bandwidth lands in these critical sites and then gets distributed

across through the access layer.

How do you manage connectivity in your network model?



With ten core sites, the need for a reliable yet agile connectivity backbone
is a must. We used different types of connectivity and multiple vendors, and

arrived at a connectivity paradigm that meets our needs and requirements.

International connectivity is primarily MPLS and we use IPLC in a small way. Our

domestic connectivity is through leased lines enabling point to multi-point

connectivity. We also use VSAT, but in a very limited way.

Choosing a connectivity vendor is critical for operations, given

that what's your approach here?



Our connectivity infrastructure is made up of multiple vendors. We
deliberately adopted that approach to arrive at a network that gives us

resilience. Moreover, we are not tied with one vendor as multi-vendor approach

gives us flexibility and heterogeneity in-terms of managing our networks

effectively. We have defined standards in every areas of IT and network is no

exception.

When we factor typical banking operations such as the ones

spread across branches, ATMs, e-Banking etc. How big are these challenges from a

network perspective?



SCB is in 21 cities and 81 sites in India. Some of the sites are support
offices. But providing connectivity is indeed a primary requirement for all

branches to successfully connect and interconnect with each other. We have 198

ATMs in India and this calls for systems to always be linked and, hence, network

availability is extremely critical. We have a primary set-up and a back-up, with

dial-ups as the back up. This combination of leased, dial-up and VSATs works

well for us to arrive at securing peak uptimes for our ATM operations.

What about network security?



Security is of paramount importance. One is our own banking operation that's
internal. The other one is external customers linked to us through our websites.

In terms of managing the network security internally, we have well defined

controls and access from inside to outside. For instance, only authorized people

are able to access applications, and it is multi-tiered with various levels of

access. We also have three-tiered firewall for Internet banking operations and

the best proactive practices to control spam and spyware.

In terms of applications, can you give us a snapshot?



We have a core banking application called 'Hogan', which India uses and
it is housed in our data center in Hong Kong. All our locations in India use

this CBS application. We also use payment-based applications like RTGS with our

member banks.

What are your key goals through the network centric model?



Our two key goals here are availability and scalability. I think these are
the two most critical factors that define the efficiency of the networks. For

instance, the networks we have created should be available to the banking

operations spread across many locations. From accessing the apps to a simple

function like email, the networks form the backbone and any downtime in that is

totally not acceptable. Meanwhile, any network has to be scaleable to the

business expansion and growth coming out of it. For instance, SCB is going on an

aggressive growth mode and our networks have to meet that without any stress. To

take stock of things from a network perspective, a while back, we have launched

a network re-engineering program aimed at further improving our overall network

backbone. We are enabling our network to future demands. For instance, once the

re-engineering program gets completed by this April, we would have bandwidth

capacity that would be 5 to 20 times higher than what we are currently having.

On new technologies...



Since we adopt a standardized approach, any new technology has to be tested
fully and we need to be comfortable to deploy that on an organizationwide basis.

Hence, our use of technologies like Wi-Fi is limited to conference rooms. And

here also, we have adopted restricted wireless usage through token-based access

and not everybody can access it. As for technologies, we are looking closely at

VoIP, where in we have created some proof points and are exploring the

possibilities of deploying it.

What are the key challenges CIOs face today in managing their

enterprise networking infrastructures?



I think the twin challenges revolve around availability and scalability. Any

network planning has to be done with growth in mind. Here you need to create a

network that is able to take up the current requirement and provide round the

clock availability. But QoS changes very frequently, and one needs to constantly

scale up the networks to accommodate new requirements. For instance,

applications and access patterns keep changing. Yet another key challenge I

consider lies in prioritization of the networks. One should be always aware of

what goes on in the network and need to put in place pro-active monitoring

mechanisms in place that ensures a secured and reliable network backbone.