Technical & Dev · Fusion · Advanced
Oracle Fusion Technical: Reports, Data, Integrations and Extensions
Nineteen modules, nineteen labs, and a technical work package end to end
The layer beneath the functional courses. Data gets into Fusion, comes back out as a report, and another system talks to it - and none of that is configuration you can click your way through. This is for the consultant or developer who can already navigate Fusion and now has to load a client history, explain why a report is slow, or make an integration survive a quarterly update. Vendor documentation lists the tools. This covers which one answers which question, what order the work goes in, and which decisions cannot be taken back once transactions exist.
EGP 3,000
177 lessons · 365 min
What you will be able to do
- Choose between BI Publisher, OTBI and a direct extract for a given reporting question, and defend the choice.
- Build a BI Publisher data model and template that still performs when the data set grows.
- Load master and transactional data with FBDI, and prove afterwards that it landed correctly.
- Plan a migration with a reconciliation that a finance owner will accept.
- Call Fusion from another system, and let Fusion call out, without inventing a bespoke integration.
- Extend Fusion inside supported boundaries, and recognise when a request has crossed one.
- Refresh a non-production environment without leaking production data or losing configuration.
- Read a quarterly update for risk and test the right things before it lands.
What you should know first
- You can navigate Fusion and have seen at least one implementation from the inside.
- SQL you can read and write - joins, aggregates, and why a query is slow.
- An environment you are allowed to break. Several lessons demonstrate operations that are disruptive in a shared instance, and one module is about refreshing environments, which cannot be rehearsed anywhere else.
- No prior BI Publisher, OIC or Java experience is assumed.
What you will cover
1. Module 0 · Orientation
What this course covers, what it deliberately leaves to the functional courses, and how to get an environment you are allowed to break.
- 1How this course works3 min
- 2Getting an environment you can break3 min
- 3Where technical work sits in an implementation3 min
- 4The vocabulary, once3 min
2. Module 1 · The Fusion technical landscape
The layers a technical consultant works across, and which problems live in which one. The module that stops you solving a reporting problem with an integration.
- 1The layers, end to end3 min
- 2What SaaS actually forbids, and why2 min
- 3Where each kind of request belongs2 min
- 4Release trains and what they mean for you2 min
- 5Reading Oracle documentation without drowning2 min
- 6The technical consultant toolkit2 min
- 7Lab briefing · Mapping a real request2 min
3. Module 2 · Security, roles and what you are allowed to see
Role-based access as it actually behaves, and the diagnostic sequence for "why can this person not see that". The module that saves you from granting a role to fix a data problem.
- 1Roles, privileges and duties2 min
- 2Data security versus function security2 min
- 3Data access sets and business unit security3 min
- 4The security diagnostic sequenceFree preview2 min
- 5Reading the security console2 min
- 6Security for reporting: the part people miss2 min
- 7Custom roles: when, and how to survive an update2 min
- 8Provisioning and role assignment at scale3 min
- 9What breaks: the three security failures you will meet2 min
- 10Lab briefing · Diagnosing three access failures2 min
4. Module 3 · The reporting stack: which tool answers which question
The decision module. Four tools overlap and choosing wrongly costs weeks, so this settles the choice before any of them is opened.
- 1The four tools and what each is for2 min
- 2Which tool answers which questionFree preview2 min
- 3Subject areas versus tables2 min
- 4When the answer is "not a report"2 min
- 5Who maintains it afterwards2 min
- 6Performance, before you build2 min
- 7What breaks: the report that worked in test2 min
- 8Lab briefing · Choosing for six real requests2 min
5. Module 4 · BI Publisher: data models, templates and bursting
FLAGSHIP. The tool most technical work ends up in, and the one where an early decision about the data model cannot be tuned away later.
- 1What a data model actually isFree preview2 min
- 2SQL data sets: writing the query2 min
- 3Parameters and lists of values2 min
- 4Bursting: one run, many recipients2 min
- 5RTF templates: the practical subset2 min
- 6Excel and eText templates2 min
- 7Sub-templates and reuse2 min
- 8Layout, pagination and the printed page2 min
- 9Localisation: Arabic, dates and numbers2 min
- 10Performance: making a slow report fast2 min
- 11Security in BI Publisher2 min
- 12Scheduling and delivery2 min
- 13What breaks: five BI Publisher failures2 min
- 14Lab briefing · A supplier statement, end to end2 min
6. Module 5 · OTBI and subject areas
The tool a business user can maintain after you leave, and the reason to prefer it whenever it can answer the question at all.
- 1What a subject area is2 min
- 2Finding the right subject area2 min
- 3Building an analysis2 min
- 4Joins across subject areas, and when you cannot2 min
- 5Calculations, bins and hierarchies2 min
- 6Dashboards and prompts2 min
- 7Security in OTBI: it is already applied2 min
- 8Real-time versus reported figures2 min
- 9Performance and the analysis nobody can run2 min
- 10When OTBI is the wrong answer2 min
- 11Lab briefing · An analysis and a dashboard2 min
7. Module 6 · Reports at scale: scheduling, delivery, performance
What changes when a report stops being run by a person and starts running itself at 2am for a thousand recipients.
- 1The scheduler, and what a job really is2 min
- 2Scheduling patterns that survive2 min
- 3Delivery: email, FTP and content server2 min
- 4Knowing a scheduled report failed2 min
- 5Volume: when the extract stops fitting2 min
- 6Incremental extracts and high-water marks2 min
- 7What breaks: the 2am report2 min
- 8Lab briefing · Schedule, deliver, and prove it ran2 min
8. Module 7 · FBDI: the file-based loader end to end
FLAGSHIP. The main road for getting data into Fusion, and the one where an error is not a rejected file but posted transactions nobody expected.
- 1The FBDI sequence, and why order mattersFree preview3 min
- 2Finding and reading the right template2 min
- 3Mandatory columns, and the ones that only look optional2 min
- 4Preparing the file without corrupting it2 min
- 5Loading the interface tables2 min
- 6Running the import and reading the output2 min
- 7Errors: finding the row that failed3 min
- 8Correcting and reloading safely2 min
- 9Master data versus transactional loads2 min
- 10Volume: loading a hundred thousand rows2 min
- 11Proving the load landed2 min
- 12Rollback, and why there usually is not one2 min
- 13What breaks: six FBDI failures2 min
- 14Lab briefing · A supplier and invoice load2 min
9. Module 8 · HDL for HCM data
HCM loads its own way, and the differences from FBDI are not cosmetic - effective dating changes what a load even means.
- 1Why HCM has its own loader2 min
- 2The .dat file format2 min
- 3Business objects and their dependencies2 min
- 4Effective dates: the part that catches everyone2 min
- 5Source keys and matching existing records2 min
- 6Loading, monitoring and the error report2 min
- 7Corrections and the delete you cannot do2 min
- 8HDL versus HSDL, and when to use each2 min
- 9What breaks: four HCM load failures2 min
- 10Lab briefing · Load a workforce, then correct it2 min
10. Module 9 · Migration strategy and reconciliation
FLAGSHIP. The module about deciding what to migrate, in what order, and how to prove it afterwards to somebody who signs for it.
- 1What to migrate, and what to leave behind2 min
- 2Balances versus transactions2 min
- 3The cutover sequence2 min
- 4Reconciliation: proving a migration landedFree preview2 min
- 5Data cleansing, and who owns it2 min
- 6Mapping legacy values to Fusion2 min
- 7Trial loads and the rehearsal nobody schedules2 min
- 8Sub-ledger and general ledger agreement2 min
- 9Signing off: what "done" means2 min
- 10What breaks: the migration that reconciled and was wrong2 min
- 11Lab briefing · Plan and reconcile a migration2 min
11. Module 10 · REST and SOAP: calling Fusion, and being called
The integration surface, without an integration platform. What Fusion exposes, how to use it, and how to keep it working across updates.
- 1What Fusion exposes, and what it does not2 min
- 2REST vs SOAP: choosing without arguingFree preview2 min
- 3Authentication and the service account2 min
- 4Reading the REST API catalogue2 min
- 5Your first call, and its error2 min
- 6Payloads, child resources and partial updates2 min
- 7Bulk operations and rate limits2 min
- 8Fusion calling out: business events2 min
- 9Idempotency and retries2 min
- 10Versioning and surviving a quarterly update2 min
- 11What breaks: five integration failures2 min
- 12Lab briefing · Read, write and break an integration2 min
12. Module 11 · Integration patterns without OIC
What you can build with what Fusion already has, and the honest boundary where an integration platform stops being optional. Keeps OIC a separate product rather than a prerequisite.
- 1The four patterns you will actually need2 min
- 2File-based integration, done properly2 min
- 3Scheduled extract to an external system2 min
- 4Point-to-point API calls, and their limits2 min
- 5Error handling without an orchestrator2 min
- 6When you genuinely need OIC2 min
- 7What breaks: the integration nobody owns2 min
- 8Lab briefing · Build one without a platform2 min
13. Module 12 · Extensions: sandboxes, Page Composer and the boundaries
Changing what users see, inside what Oracle supports. The module that teaches you to recognise a request that has crossed the line before you have spent a week on it.
- 1What a sandbox is, and what it is notFree preview2 min
- 2Publishing, and the point of no return2 min
- 3Page Composer: what it can change2 min
- 4Descriptive and extensible flexfields2 min
- 5Value sets and validation2 min
- 6Application Composer versus Page Composer2 min
- 7Visual Builder: where it fits2 min
- 8Recognising an unsupported request2 min
- 9What breaks: extensions and quarterly updates2 min
- 10Lab briefing · Extend a page and survive an update2 min
14. Module 13 · Groovy in Application Composer
The scripting that runs inside Fusion. Small, constrained, and the place where an innocuous-looking rule becomes a performance problem for everyone.
- 1Where Groovy runs, and where it does not2 min
- 2The language subset, quickly2 min
- 3Validation rules that help rather than annoy2 min
- 4Triggers and the order they fire in2 min
- 5Object functions and reuse2 min
- 6Performance: the rule that runs a million times2 min
- 7Debugging without a debugger2 min
- 8What breaks: three Groovy failures2 min
- 9Lab briefing · A rule, and its performance2 min
15. Module 14 · Approvals and BPM workflow
Routing work to people. Configuration rather than code, and the module where a wrong rule is discovered by an invoice that nobody approved for three weeks.
- 1How approvals are actually structured2 min
- 2Approval rules and how they are evaluated2 min
- 3Supervisory and position hierarchies2 min
- 4Parallel, serial and FYI participants2 min
- 5Delegation, escalation and timeouts2 min
- 6Notifications and what they say2 min
- 7Testing an approval rule properly2 min
- 8Diagnosing a stuck transaction2 min
- 9What breaks: the approval nobody received2 min
- 10Lab briefing · Build and break an approval chain2 min
16. Module 15 · ESS jobs and scheduling
The scheduler everything else runs on. Understood properly it explains most "why did that not happen" questions in Fusion.
- 1What Enterprise Scheduler does2 min
- 2Submitting, monitoring and cancelling2 min
- 3Job sets and dependencies2 min
- 4Custom ESS jobs for your own reports2 min
- 5Parameters, defaults and the date trap2 min
- 6Reading a failed request3 min
- 7What breaks: the job that stopped running2 min
- 8Lab briefing · Register and schedule a job2 min
17. Module 16 · Environment management: P2T, refreshes and what breaks
FLAGSHIP. The module that cannot be rehearsed anywhere but a real estate, and the one where a mistake sends production data somewhere it must never be.
- 1The environments, and what each is for2 min
- 2What a P2T actually copies2 min
- 3The checklist before anybody clicksFree preview2 min
- 4Data masking and why it is not optional2 min
- 5Preserving configuration across a refresh2 min
- 6Integrations after a refresh2 min
- 7Scheduling refreshes around a project2 min
- 8What breaks: four refresh failures2 min
- 9Lab briefing · Plan a refresh you could defend2 min
18. Module 17 · Release readiness: quarterly updates and regression
Updates arrive on a schedule you do not control. Reading them for risk, and testing the right things rather than everything.
- 1Reading an update for riskFree preview2 min
- 2What is exposed, and what is safe2 min
- 3Building a regression pack that stays useful3 min
- 4Testing in the update window2 min
- 5New features: opting in deliberately2 min
- 6Communicating an update to the business2 min
- 7What breaks: after the update landed2 min
- 8Lab briefing · Assess a real release note2 min
19. Module 18 · Capstone: one technical work package, end to end
A single realistic brief that touches reporting, loading, integration and extension - delivered, reconciled and handed over.
- 1The brief2 min
- 2Sequencing the work2 min
- 3Building it2 min
- 4Proving it2 min
- 5Handing it over2 min
- 6Where to go next2 min