HDFC Bank's Journey to Oracle Private Database Cloud
By Nilesh Agrawal on Nov 08, 2012
One of the key takeaways from a recent post by Sushil Kumar is the importance of business initiative that drives the transformational journey from legacy IT to enterprise private cloud. The journey that leads to a agile, self-service and efficient infrastructure with reduced complexity and enables IT to deliver services more closely aligned with business requirements.
Nilanjay Bhattacharjee, AVP, IT of HDFC Bank presented a real-world case study based on one such initiative in his Oracle OpenWorld session titled "HDFC BANK Journey into Oracle Database Cloud with EM 12c DBaaS". The case study highlighted in this session is from HDFC Bank’s Lending Business Segment, which comprises roughly 50% of Bank’s top line. Bank’s Lending Business is always under pressure to launch “New Schemes” to compete and stay ahead in this segment and IT has to keep up with this challenging business requirement. Lending related applications are highly dynamic and go through constant changes and every single and minor change in each related application is required to be thoroughly UAT tested certified before they are certified for production rollout. This leads to a constant pressure in IT for rapid provisioning of UAT databases on an ongoing basis to enable faster time to market.
Nilanjay joined Sushil Kumar, VP, Product Strategy, Oracle, during the Enterprise Manager general session at Oracle OpenWorld 2012. Let's watch what Nilanjay had to say about their recent Database cloud deployment.
“Agility” in launching new business schemes became the key business driver for private database cloud adoption in the Bank. Nilanjay spent an hour discussing it during his session. Let's look at why Database-as-a-Service(DBaaS) model was need of the hour in this case -
- Average 3 days to provision UAT Database for Loan Management Application
- Silo’ed UAT environment with Average 30% utilization
- Compliance requirement consume UAT testing resources
- DBA activities leads to $$ paid to SI for provisioning databases manually
- Overhead in managing configuration drift between production and test environments
- Rollout impact/delay on new business initiatives
The private database cloud implementation progressed through 4 fundamental phases - Standardization, Consolidation, Automation, Optimization of UAT infrastructure. Project scoping was carried out and end users and stakeholders were engaged early on right from planning phase and including all phases of implementation.
- Standardization and Consolidation phase involved multiple iterations of planning to first standardize on infrastructure, db versions, patch levels, configuration, IT processes etc and with database level consolidation project onto Exadata platform. It was also decided to have existing AIX UAT DB landscape covered and EM 12c DBaaS solution being platform agnostic supported this model well.
- Automation and Optimization phase provided the necessary Agility, Self-Service and efficiency and this was made possible via EM 12c DBaaS. EM 12c DBaaS Self-Service/SSA Portal was setup with required zones, quotas, service templates, charge plan defined. There were 2 zones implemented - Exadata zone primarily for UAT and benchmark testing for databases running on Exadata platform and second zone was for AIX setup to cover other databases those running on AIX. Metering and Chargeback/Showback capabilities provided business and IT the framework for cloud optimization and also visibility into cloud usage.
More details on UAT cloud implementation, related building blocks and EM 12c DBaaS solution are covered in Nilanjay's OpenWorld session here.
Some of the key Benefits achieved from UAT cloud initiative are -
- New business initiatives can be easily launched due to rapid provisioning of UAT Databases [ ~3 hours ]
- Drastically cut down $$ on SI for DBA Activities due to Self-Service
- Effective usage of infrastructure leading to better ROI
- Empowering consumers to provision database using Self-Service
- Control on project schedule with DB end date aligned to project plan submitted during provisioning
- Databases provisioned through Self-Service are monitored in EM and auto configured for Alerts and KPI
- Regulatory requirement of database does not impact existing project in queue
This table below shows typical list of activities and tasks involved when a end user requests for a UAT database. EM 12c DBaaS solution helped reduce UAT database provisioning time from roughly 3 days down to 3 hours and this timing also includes provisioning time for database with production scale data (ranging from 250 G to 2 TB of data) -
And it's not just about time to provision, this initiative has enabled an agile, efficient and transparent UAT environment where end users are empowered with real control of cloud resources and IT's role is shifted as enabler of strategic services instead of being administrator of all user requests. The strong collaboration between IT and business community right from planning to implementation to go-live has played the key role in achieving this common goal of enterprise private cloud. Finally, real cloud is here and this cloud is accompanied with rain (business benefits) as well !
For more information, please go to Oracle Enterprise Manager web page or follow us at :Twitter | Facebook | YouTube | Linkedin | Newsletter