Angel Broking, a stock broking and wealth management firm with a retail-focused business model, has decided to move its disaster recovery (DR) system for its mission-critical applications to the cloud. In the first phase of implementation, which commenced February 2011, Angel Broking outsourced its DR system infrastructure to two service providers, Reliance and BSNL. However, the primary infrastructure will continue to reside at the company’s Mumbai data center. By October 2011, when the first phase is expected to conclude, the company would have moved its business-critical applications to a single service provider that meets all its requirements.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
The first phase involves moving all critical systems, functions and processes to the cloud, on approximately 80 servers attached to HP storage. The hardware includes HP and Dell servers running on Linux as well as Windows platforms.
Prior to the decision to move its DR system to the cloud, Angel Broking operated in a decentralized environment. Operations were spread over almost 20 data centers across the country, and managing everything was no easy task. Ketan Shah, Angel Broking’s associate director for IT and B2B business, says, “With technology moving more towards convergence, along with bandwidth and networks becoming stable, it made more sense to consolidate the entire infrastructure under one roof. This is our long term plan.”
Angel Broking’s overall IT infrastructure consists of 250-300 servers. Earlier, with multiple data centers, the DR system was already taken care of, as there was no single point of failure. With consolidation, the new dependency on a lone data center created a single point of failure, and hence the need for a DR system.
A DR system on the cloud comes with its own set of risks and vulnerabilities, so issues of security and data integrity are paramount. Proper management of the service provider also assumes importance. “When the DR infrastructure is in your entire control, the processes can be changed as per your requirement. When another party is involved in the DR processes, it is time-consuming, and one has to make sure it does not affect uptime, which is very critical,” Shah points out.
At this stage, Angel Broking has not defined all the processes, but is rather concentrating on getting the optimum hardware (server, storage and power) infrastructure in place for its cloud-based DR system. Subsequently, the DR system’s replication processes, activities and other considerations will be deliberated on, and finalized.
On the point of Angel Broking’s readiness for a DR system on the cloud, Shah says, “It is more about maturity and learning. When you are going to outsource DR infrastructure, there will be different kinds of challenges that one is constantly badgered with. However, with fore vision and careful planning, one can easily avoid these.”