The growing scale of today's data centers puts increasing demands on administrators; deployment and maintenance operations must be efficient and non-disruptive. Oracle's Fleet Patching and Provisioning (FPP), formerly known as RapidHome Provisioning (RHP), standardizes and simplifies software distribution and management. Automation and efficiency minimize the impact of maintenance and scale to large deployments.
Oracle Fleet Patching and Provisioning, or FPP, helps you control your database fleet lifecycle through automation, standardization, and out-of-place patching. All FPP operations have been tested and validated according to Oracle's Maximum Availability Architecture blueprints. Customers implementing Oracle Fleet Patching and Provisioning can effortlessly patch hundreds or thousands of databases per maintenance window with the minimum human interaction, enabling consistent time and money savings.
Please join us in this blog post on how FPP is used @ INGKA group.
INGKA Group is one of 12 franchisees operating IKEA stores and other sales channels under franchise agreements with Inter IKEA Group, the owner of the IKEA Concept and worldwide IKEA franchisor. The Group operates three businesses in one: IKEA Retail, Ingka Centers, and Ingka Investments.
IKEA retail is its core business, and it currently operates 482 IKEA stores, shops, and planning studios in 31 countries.
INGKA Group’s IT division supports 4 Data Centers worldwide with many distributed environments with one or the other database. The team is responsible for administrating, maintaining, and patching about 4,000 Oracle databases. The DBA team started using Fleet Patching and Provisioning (FPP) to alleviate this job.
We recently had the chance to talk to Jonas Persson, System Engineer – Oracle database at INGKA Group IT, about their use of Fleet Patching and Provisioning.
We mainly used Oracle Enterprise Manager (OEM) to provision databases, which worked very well. However, we noticed that OEM, due to its broad range of functionality, has lots of moving parts that needed to be kept up to date, which took a lot of work to manage in our complex environment. This was when we started looking for a different solution, focused on provisioning and patching.
I wouldn’t say that we used fleet maintenance as a whole; we used some parts, but mainly the out-of-the-box procedures, such as the provisioning procedures in OEM. We didn’t use out-of-place patching. We always patched in place, which is probably not the best from a timing point of view, but it worked for an environment like ours. When we started using FPP, the main objective was to solve the provisioning issues. We are using one FPP installation in one of our data centers in Sweden, which we then use to provision the other data centers and the distributed sites. We have an isolated lab environment where we evaluate and test features, new functionality, and patches.
About 3,000 of the 4,000 databases are provisioned using FPP – our virtualized environment is taken care of in a slightly different way. But we use RAC and single instance – the latter without ASM.
I like to see it as the central piece of the estate. One of the repositories of the estate, as it’s easy to use, patch version agnostic, as you always deal with it in the same way. The zip-copy functionality is one of our favorites – it makes especially sense for our distributed environments and dramatically reduces the time to deploy new Oracle Homes – providing a zip file as input is straightforward. Since FPP uses a gold image approach, it is an excellent deal for standardization!
Since we know that 23c is a major release, we will start working with it immediately. We will install and test it to investigate and get accustomed to the new features. Then, we will decide which feature we will enable and use. To keep the answer short, for major releases, we start working from the very beginning.
Yes and No, we work with the latest, but only twice a year. We have RU on a 6-month schedule, which differs from the quarterly you are offering.
No, for some of our applications, we looked into using it, but most of the applications don’t fulfill the prerequisites for using this feature.
We use Data Guard for our patching strategy, services, connection draining, etc. As a result, we minimize our downtime. Together with RAC, it enables us to make patching as smooth as possible.
The key advantages we see are standardization, ease of use, patch automation, and repeatability. FPPs' gold image out-of-place patching methodology also makes rolling patching and rollback of patches very easy and fast.
We estimate to at least win 15-20% in efficiency gains if not more.
Start small, get your hands dirty, and try to use the documentation!
Thank you, Jonas, for your time and testimony using FPP at INGKA IT.
Jonas Persson is a Systems Engineer, with close to three decades of database experience, focusing on assisting the shift from traditional database hosting infrastructure into a scalable, modern, API-driven and cloud-like database platform, shaping the future of IKEA.
Martina Keippel has worked in the IT industry for over 35 years and has held positions in Sales, Marketing, and Business Development within Product Management. Today she is leading a team responsible for references and programs in Product Management.
Next Post