When your organization is responsible for helping customers make business-critical decisions, your internal delivery processes need to be on point.
This global enterprise mobile computing company creates solutions that connect users with the data they need, guiding them with clear and accurate insights for the best next action.
The Challenge: Delivering quality software on time
The company’s IT team had long-standing challenges with manual migration and a lack of visibility. One Senior Manager had been with the company for about 11 years in the Middleware team. When he started, there was a small Middleware footprint, using SOA 10g and some small Java applications. Since that time, the team upgraded to SOA 11g and planned an upgrade to 12c. Additionally, they had introduced technologies such as Oracle Data Integrator (ODI) and their footprint had grown to 18 on-prem instances.
The impending 12c upgrade brought all their challenges to the forefront. They had managed their upgrade from 10g to 11g manually, which the Senior Manager described as a “complete nightmare”—a tedious manual approach resulting in a mismatch in code migrations, uncertainty about what code base was in which instance, relying on spreadsheets to track changes, and long execution times.
The processes were painful and frustrating for the Middleware team’s upgrade and had ripple effects on projects across other IT groups and the rest of the business. When deployments were not accurate the first time, shared services groups had to run steps like deployments multiple times, tests had to be re-run, and project timelines were extended, impacting final delivery for the business. “If the correct code base was not in a target environment, migrations to that environment would have to happen multiple times,” explained the Senior Manager. “This impacted the DBA team who had to re-migrate, the test team who had to run tests again, and it delayed any downstream activities.” These issues left the organization with a constant skepticism – they never knew if the right code base would be in the target instance, and if project timelines would be delayed.
The Solution: One centralized DevOps platform for automated and transparent delivery processes
The customer's Oracle Business Intelligence Enterprise Edition (OBIEE) team had been using FlexDeploy with positive results and suggested the Middleware team use it with their upgrade. The team did a brief POC which “went really well,” and quickly started using FlexDeploy in production with SOA.
The Outcome: Improved visibility, faster time to market, and increased stability
The Middleware team initially used FlexDeploy with SOA and ODI on-prem, and later introduced Oracle Cloud Infrastructure (OCI) and Oracle Integration Cloud (OIC). “With SOA and ODI we are in a stage where we’re doing CI/CD,” said the Senior Manager. “We can do continuous deployment without any issues and are moving to get there with OIC at a good pace.”
The team found that FlexDeploy is the right tool for them to grow their CI/CD practices and take them to the next level. With FlexDeploy, they replaced manual processes vulnerable to human error with ones that are automated and repeatable. They no longer track environments via spreadsheets, but instead have visibility to what code base is running where, reducing the risk that needed code will be overwritten or that an environment isn’t ready for a deployment.
The organization has gained benefits across five main areas using FlexDeploy:
- Reduction in Time: A 70% reduction in migration efforts occurred by injecting automation and removing manual steps. Where migrations used to take five to six days, they are now taking four to five hours. This is reducing manhours required on projects and freeing the group to take on additional tasks and deliver faster.
- Stability of the Code Base: With FlexDeploy, the team never needs to guess what code is in which environment. They have confidence when they deploy to a target environment that the expected code will be there the first time. This has eliminated additional time spent to troubleshoot, re-migrate and test where errors arise.
- Cross-Functional Improvements: The ripple effect of delays created when the Middleware group needed to re-test or re-deploy has been eliminated, increasing cross-functional team effectiveness, confidence, and satisfaction.
- End-to-End Process Flow: The team has integrated FlexDeploy with other tools in their toolchain. Users are no longer jumping between tools and instead have a holistic, fluid process.
- Improved Speed of Delivery to Customer: FlexDeploy has eliminated the delivery slippage that would occur in 10-15% of the customer's projects. Ultimately, increasing project delivery speed and ensuring new features and enhancements are available to customers quicker.
The Senior Manager summed the improvements by saying, “We’re at a point where we just know it’s going to work when something is deployed. We’re not guessing anymore. Deploy one or one hundred and we know it will work.”
Next Steps: Expanded usage in the Middleware Team and across the IT organization
The Middleware team plans to continue expanding their use of FlexDeploy, first by connecting it further across their toolchain – integrating with test tools such as Tosca and security and authentication tools. They also plan to expand usage of FlexDeploy’s built-in reporting and auditing functionality to support process-level audits and traceability.
Based on the success the Middleware team has seen, other groups within the IT organization are looking at FlexDeploy. “Maybe the best testimonial I can give is that other teams are looking at and starting to use FlexDeploy for their own delivery," the Senior Manager says. "They saw our results and wanted to get those same benefits themselves.”
- Industry: High Tech
- Employees: 8,800
- Oracle SOA Suite
- Oracle Integration Cloud
- Oracle Cloud Infrastructure
- Oracle Data Integrator (ODI)
- Oracle Business Intelligence
Life After FlexDeploy
- Reduced overall project deployment time by 30X—from six days to five hours
- Visibility to what code is running in which environment
- Increased speed of delivery to customer
- Completely eliminated project schedule delays that previously occurred in 15% of projects
- Confidence that deployments will work the first time, every time
“We’re at a point where we just know it’s going to work when something is deployed. We’re not guessing anymore. Deploy one or one hundred and we know it will work.”