Feature

Insurance against disaster

With the July floods in Mumbai playing havoc with the life of the common man, organisations have had to take a hard look at their technology initiatives. The Insurance sector is no different, and it has realised the importance of maintaining DR facilities. Priya Jain reports

The Mumbai rains that paralysed the city a few months back also revealed the loopholes in the supposedly infallible technology initiatives of various companies, especially those in the BFSI segment. After the deluge we heard stories of how customers faced problems trying to withdraw or deposit money; there were also reports of a growing list of insurance claimants.

Many insurance companies had to call for help from other cities for making surveys. Though these companies insure us, do they have systems guaranteeing that they themselves are safeguarded from natural disasters? Do they have ready databases available at other sites, or for that matter, do they have a disaster recovery (DR) site?

Here we try to find out how these companies cope with disasters. Do they have the infrastructure in place to deal with such calamities in future? And for those who do not, are they on their way to planning for future emergencies?

Infrastructure basics

A DR plan is focussed only on how to recover the IT components so that business processes can be recovered. Ajay Patil, Head, Systems, Birla SunLife Insurance says, “DR can mean different things to different companies. In our organisation we have certain processes in place to manage the impact of a disaster. We do not have any specific DR software except backup and recovery. We use Veritas and Tivoli software for backup/replication. All our applications and production data is backed up to the DR site. Our branches and central office are connected on the WAN. We have equivalent hardware and sufficient capacity to switchover our entire production load. Every night the production data is replicated at the DR site, and this is tested at least once in six months.”

LIC and GIC were the early players in the insurance sector, hence the need for them to overhaul their systems. Reveals S K Kapahi, LIC’s Executive Director for IT/BPR, “We are in the process of finalising a DR plan. Mumbai floods or otherwise, a DR system is necessary for any organisation having large IT initiatives. The main objectives under DR are to be able to provide services in the shortest possible time even if a disaster strikes. For selecting sites a number of factors such as safety, security and infrastructure have to be taken into consideration, and different sites may satisfy these criteria. Ultimately, the choice is left with the organisation based on its business considerations.”

LIC, the biggest and oldest insurance company in India, is in the process of getting its IT infrastructure in place. The organisation has undertaken a lot of IT initiatives, one of them being warehousing with Wipro and Teradata at a cost of Rs 35 crore spread over three years. The need for warehousing is to have a centralised database. The vendors would be collecting data from LIC’s 2,000 centres, placing it in a centralised location, and then building an MIS application over it. They then also plan to implement a CRM solution. This would help them in the integration of their delivery channels. Presently, they have call centres taking queries from customers, but once the data warehousing is centralised and linked to a network all the interactions can be tracked. Under the current system, LIC has a unique ID number for each customer, which allows them to know all about the interactions between their officials and the customer.

HDFC Standard Life has multiple UPS systems (main and backup) which can supply power for the main servers and telecom equipment for two hours. Physical access to the server room is controlled by an access control system. For desktops and laptops, guidelines have been provided to end-users to reduce risk. Critical systems have backup which can be started when the main system fails. Redundancy is provided for routers, switches, etc., and WAN links are provided with redundancy at critical locations.

At ICICI Lombard General Insurance, the core insurance applications are configured on multiple servers with a Network Load Balancing Service to ensure the even distribution of transaction load during peak hours. The databases for their core applications are kept on a storage area network (SAN). The SAN ensures data integrity in case of an error on the database servers. According to Anuj Gulati, Head of Operations & Technology at ICICI Lombard, they take regular backups of their database or application servers based on pre-defined policies governing the specific application.

States Gulati, “In case of a failure, we have dedicated systems for urgent restoration on site. This gives us the ability to re-implement all our business-critical applications with a minimum turnover time at our DR site or production site in the event of total system failure. We also keep a copy of our backup media at an offsite location (Hyderabad) to ensure maximum data safety and security. Further, there are processes laid down and infrastructure in place to allow for system restoration within pre-defined timelines.”

Investment in DR

Investment is the most important component of any decision-making process— and also the most negotiated one. DR services do not come cheap. Insurance companies look realistically at the cost of people, equipment, environment, maintenance, power, software licences, communications, etc. A decision based solely on price can have implications on the vitality of a DR plan.

Patil reveals that having a DR site was part of the business plan even at the time Birla SunLife started its operations in 2001. “What a DR plan has done is to ensure that our turnaround time for the customer will not suffer even in the case of a disaster. We have invested Rs 2 crore as the one-time set up cost, and around Rs 50 lakh per year is our recurring cost on DR infrastructure.” Though Gulati was not ready to disclose the company’s DR investment, their total IT infrastructure investment is about Rs 20 crore.

According to Gartner, two out of five enterprises that experience a disaster go out of business within five years. Since prevention of disaster or managing the same is so critical for the survival of any business, it is high time for insurance companies to update their data bank of lessons learnt from previous disruptions so that swift action can be initiated in case of similar situations in the future.

There are processes laid down and infrastructure in place to allow for system restoration within pre-defined timelines Anuj Gulati Head, Operations & Technology ICICI Lombard General Insurance	Every night the production data is replicated at the DR site, and this is tested at least once in six months Ajay Patil Head, Systems Birla SunLife Insurance

Need for effective DR

Today, verticals like BFSI, telecom and energy are becoming the main drivers of the DR market in India. These businesses have direct transactions, so customer data needs to be mirrored to a remote location in order that their business does not get affected come what may. Enterprises are also looking at data mirroring sites in multiple locations depending upon their requirements. Certain companies have many sites that are continuously operating at the same time, and, therefore, if a disaster were to occur the other site or location would take over.

Gulati points out that having a DR plan in place is a necessity. “During the Mumbai floods, when our call centre based in north Mumbai was rendered inaccessible as a result of waterlogging, we expanded the call handling capacity at our parallel facility in Hyderabad. All the calls to our Mumbai call centre were automatically diverted to Hyderabad, thus ensuring 100 percent uptime for our customers.”

Mirror, mirror on the wall

With increasing branches and the need for smoother functionality, many of the insurance company have dedicated DR sites. The location of the alternate site is considered to be crucial for efficient recovery of data during disasters like the Mumbai floods. Patil states, “We have a warm DR site belonging to the Aditya Birla Group at Hyderabad, which is a seismically different zone from Mumbai; it is connected by a 2 Mbps link.”

He elaborates: “Our turnaround time for claims was not affected by the July 26 Mumbai floods. Since our production servers were affected, we had to partially activate our DR site for some functions.”

Gulati agrees about the importance of an alternate DR site. “Technology is the backbone for our policy administration and claims servicing. Our primary data centre is located in Mumbai. This data centre has been designed keeping in mind best practices relating to power backup, data port, air circulation, air-conditioning and backups for various utilities. A similar infrastructure has been replicated in Hyderabad at the DR centre.”

Many companies were not ready to disclose information about their DR infrastructure, but some showed increased awareness of the need to have a DR system in place. However, some like LIC are already moving towards acquiring better DR systems.

It is difficult to deny that the July deluge in Mumbai was an eye-opener for large organisations, including those in the insurance sector. No wonder RBI has taken up the matter seriously, and issued a circular initiating the process regarding the urgent need for disaster recovery systems in banks.

priya@expresscomputeronline.com