By Jérome Délabarre, team lead for Industrialization & Tools, Société Générale
Editor’s note: This article is part of an ongoing series of guest posts by HP Software customers about Automation and Cloud Management use cases.
About the author: Jérome Délabarre manages a team in charge of Industrialization and Tools around Automation, Orchestration and Application Industrialization. He has been At Société Général for six years and has successfully implemented a fully automated Code Deployment solution that delivers quality, efficiency and agility to the Line of Business.
Build, deploy, test. Repeat. Repeat. Repeat. Is your deployment process easily repeatable?
With more than 500 applications in production, Société Générale (Socgen) needed to reduce the time it took to deploy these applications.
Here is how we implemented an automated application deployment solution from HP to improve the time to deploy by nearly eight times using a seamless, standardized approach.
Automating Ops to support new development methods
Socgen Retail Banking operates 6000 servers, multiple datacenters and more than 500 multi-tiered applications running on Windows, Unix (AIX) and Linux. New Agile development methods were resulting in short sprint cycles, and new behavior tests were being added to existing unit and functional tests to perform automated end-to-end testing on fresh environments. As a result, we were getting an increase in requests to deploy more frequent, small updates to production environments, which lead to hosting infrastructure changes.
In Operations, we needed an end-to-end automation solution to handle the infrastructure topology and application code lifecycle. An automated synchronous de-provisioning process was required in order to reuse resources to deliver fresh installation environments and run the tests. We also needed to redefine and shorten our ITIL change management process, as well as revisit our capacity planning process for resource pooling.
In short, if Dev was running sprints, Operations needed to change the discipline to compete in the 4x100m relay!
How Socgen uses automation and orchestration
Before cloud automation, a project manager would make requests via ticketing, tools, email or even phone. This process resulted in a set of tasks separately completed by system, middleware, database, storage and network engineers. With automation, the project manager makes a request through the Cloud Service Automation (CSA) portal, which fully automates the process. We achieve synchronous process between demand and execution, with no requirement to set time into account demand or monitor the player in charge of the application. The entire process is nicely repeatable.
Orchestration also plays a key part, building a service from the bottom up from the infrastructure to the application. With this IT process automation, the configuration layer is a self-contained package that is built, adapted and integrated into the application package. All material is deployed via a robot.
Here is brief look at how applications are deployed with CSA (working up from the bottom of the graphic).
Mandatory installation of monitoring & backup products
Infrastructure topology creation based on GTS IaaS Cloud
At this point, product binaries are deployed but not configured yet
Automated product installation based on automation flows triggered from the Service Portal
3. Configuration of application code deployment
Application Code Deployment using Neptune on a newly created infrastructure topology
Configure products based on application specific data stored in the self-contained package (connectors, JMS files …)
4. Production tools
You can execute exploitation action before deployment
The relationship with the Line of Business has completely changed. Now the Line of Business developers deploy a complete artifact including all the components of applications in dev. Production deploys the identical complete artifact, include all the components of applications, in the approval and production environment. We plan on making this even more seamless with continuous deployment with HP Codar.
Today, we have a total of 41 services across four CSA Catalogues, and are delivering benefits to all our stakeholders:
Standardized and production-grade delivery
Reduced release management cycles
Upfront package validation to detect errors
Differential application upgrades
Rollback always possible
Package pre-staging during off-peak hours
Compliance enforced, platforms auditable
Developers gained autonomy on integration and assembly environments
Traceability of application deliveries
Logs accessible to Dev and Ops
We are able to have a virtual machine ready to use within 1 hour if it only requires an OS, if needs database, middleware and a web tier, with the associated configuration, we can get that ready in four hours.
Overall, we’re handling 300-plus changes each month, with 150 applications ready to deploy and have reduced the number of application deployment incidents by 95 percent — proof that we have successfully made our deployment processes easily repeatable.
Nimish Shelat is currently focused on Datacenter Automation and IT Process Automation solutions. Shelat strives to help customers, traditional IT and Cloud based IT, transform to Service Centric model.
The scope of these solutions spans across server, network, database and middleware infrastructure. The solutions are optimized for tasks like provisioning, patching, compliance, remediation and processes like Self-healing Incidence Remediation and Rapid Service Fulfilment, Change Management and Disaster Recovery.
Shelat has 23 years of experience in IT, 20 of these have been at HP spanning across networking, printing , storage and enterprise software businesses. Prior to his current role as a Manager of Product Marketing and Technical Marketing, Shelat has held positions as Software Sales Specialist, Product Manager, Business Strategist, Project Manager and Programmer Analyst.
Shelat has a B.S in Computer Science. He has earned his MBA from University of California, Davis with a focus on Marketing and Finance.