Case Studies

Powering India Inc

Indian enterprises believe in putting world-class server infrastructure in place. This is evident from the examples of The Stock Exchange, Mumbai, Goodlass Nerolac Paints and Sundaram-Clayton.

BSE’s consolidation gain story

When you are handling one of India’s biggest stock exchanges, managing a host of different servers is the last thing you need on your plate. This was the issue that S B Patankar, Director, IS, The Stock Exchange, Mumbai, faced having to deal with decentralised applications.

BSE’s server problems had roots in the organisation’s rapid growth over the years. As the organisation grew, its servers also mushroomed. Each department had its own servers resulting in a situation where the stock exchange found itself with server management issues and inefficient resource utilisation. According to Patankar, “One application needed higher power, while the other did not. So the hardware was not being utilised efficiently.”

Identifying the bottlenecks

This was the time when BSE started identifying critical applications that had to be operated from its DR site. Consolidation of applications and servers was diagnosed as the cure for BSE’s IT ills.

Server consolidation was undertaken to begin with. However, BSE had to sort out some tangles to get its consolidation right. “To cite an example, derivatives run through the day and we need lot of compute power for that. At the end of day, the settlement system requires more power. This is the time when trading does not happen. So if we keep more CPU power for derivatives, it is going to waste,” Patankar explains.

The right stuff

The stock exchange opted to use HP rp8400 servers based on the PA-RISC 8700 CPU to resolve the issues that they faced. Two rp8400 servers in cluster mode were used. The VPAR tool that came with the servers was used to create virtual partitions allowing BSE to allocate CPUs, memory and I/O according to the different application requirements.

Visible savings

At present, BSE has consolidated from 14 to just 2 servers with a fail over. If a server goes down, the applications fail over to the second server.

BSE uses dynamic partitioning so that trading is given the maximum CPU power during the day. At the end of day, the settlement system gets more power. “The same CPUs that were lying idle earlier are now utilised for various applications at different points of time. Thus the system is more responsive, hardware costs are reduced and the hardware is utilised more efficiently,” Patankar adds. According to him, the ROI calculation for the servers has been simple. Earlier, BSE used to have three-year AMCs for its 14 servers. The new rp8400 servers with three-year warranty were purchased for the cost of that AMC. Apart from this, BSE also saved on administration because of the changeover. “The new servers are cheaper and easier to manage. In addition, we almost doubled the power available for each of the applications,” Patankar affirms.

BSE also saved in terms of DR because they did not need to buy 14 servers for a secondary site. As the organisation was able to reduce the number of servers at the main site to two, all that was required was to replicate this smaller set-up at the DR site. It saved here too. The servers used previously are being used for DR now with necessary upgrades such as additional CPUs and memory.

Anil Patrick R can be reached at

anilpatrick@networkmagazineindia.com

SCL’s adaptive server environment

Automobile component manufacturer, Sundaram-Clayton Limited (SCL), wanted to squeeze the most from its ERP system. To this end, it built its ERP (SAP R/3) atop an Itanium 2 based server architecture.

The suitable architecture

SCL set about evaluating an appropriate server architecture from solution providers such as HP and IBM. “The evaluation was based on techno-commercial matrices which looked at the performance of the solution versus the cost,” explains T G Dhandapani, CIO at SCL.

The solution was based on HP’s Itanium 2-based Integrity servers. Each dual processor server enables the architecture to support No Single Point of Failure (NSPoF) and ensures maximum system uptime.

The servers offer multi-OS compatibility, allowing standardisation of IT infrastructure on a single platform. The cost of deploying two sets of servers for the two business divisions was about Rs 1.8 crore.

Rapid-fire introductions

Infrastructure reliability also derives from the fact that the 2-way servers are clustered. As soon as CPU use hits a pre-set threshold, the system is designed to send an SMS to the system administrator. HP OpenView is used to manage the network and SAP’s Solution Manager to monitor application performance.

The development server, which ran IBM AIX, had to be migrated to a HP Itanium platform. Clustering the servers took some work.

Adapting servers for the better

About 150 team members at SCL access the new set-up, and another 100 will be added as SAP goes live at two more group companies and four manufacturing units in Hosur and Pune.

“The transaction time has reduced by almost a third. The IT infrastructure has become reliable, and availability is high as the clustering of applications has resulted in a reduction in downtime. This helps us fully utilise the benefits of ERP. The server architecture is easy to scale since an increased workload in future will only need to incrementally add hardware,” comments Dhandapani.

Soutiman Das Gupta can be reached at

soutimand@networkmagazineindia.com

Paint my server

Jason Gonsalves GM, IT & Costing GNPL

With the who’s who of global players entering the Indian market, Goodlass Nerolac Paint Limited (GNPL) faced competition on an unprecedented scale during the late 1990s. Like many Indian players, the organisation had no option but to adopt global systems and standards.

The situation led Jason Gonsalves, GM, IT & Costing, GNPL, to consolidate the organisation’s distributed applications. “IT was identified as one of the key vehicles to bring about the change to deal with global competition. Technology had matured sufficiently by 2000 to consider a centralised environment and a process-based approach,” says Gonsalves.

Today GNPL is an Indian market leader operating in two segments—decorative and industrial paints. It is a leading player in the industrial segment with over 45 percent marketshare. Clubbing the industrial and decorative paint segments together, it is the second largest player in the paint industry with a marketshare of approximately 20 percent. The organisation operates out of five factories and 65 depots spread across India.

Order from chaos

GNPL decided to get its infrastructure in shape in 2000. At that time, the organisation ran a legacy application built around technologies such as Cobol and Unix. This was a decentralised IT set-up with applications spread across 65 locations and 67 servers. Five of its factories had five different servers, and the head office had its own servers. “All these necessitated a backbone and that’s where the servers came. The driving point was not to change the servers, but to drive business needs. Luckily, one strength was that the platform was the same throughout the organisation,” explains Gonsalves.

GNPL decided to move on from the legacy decentralised mode of working to a SAP ERP system. At the same time, it deployed data warehousing from SAS. Because the company had a Unix-Cobol platform throughout the organisation, the idea of a centralised architecture was easy to sell. The projects were started together in early 2000, and they went live by November that year.

The supporting backbone

The server farm was started from 2000 onwards starting with the SAP R/3 platform. Seven Sun servers were used for this infrastructure.

A clustered database with a backup solution from Sun was deployed. However, due to the high data growth (14 to 16 GB per month) generated by SAP R/3, GNPL went in a data archiving system using an Intel class server from Wipro.

A supply chain solution using SAP APO was next in line in 2003. This called for augmentation of the server architecture. For this, GNPL added four more servers from Sun. “When we did server sizing, we were getting values as high at 32 GB of RAM. Today it is being used in a server which has 40 GB of RAM,” Gonsalves adds.

GNPL switched from SAS data warehousing to SAP BIW. This required three more Sun servers, which has resulted in the creation of a farm of 20 servers which GNPL is in the process of consolidating. Says Gonsalves, “Work is on jointly with Sun and Wipro for consolidating the servers. We are trying to consolidate SAP R/3 and SAP APO on to the same server.”

Anil Patrick R can be reached at anilpatrick@networkmagazineindia.com

Soutiman Das Gupta can be reached at soutimand@networkmagazineindia.com