Central Platform Job Scheduling and Execution Hub (JobSpider)
Background jobs for the whole platform are scheduled and run from one point. It reads the target service's address from service discovery, and not from fixed configuration.
- Domain
- Scheduling and execution of background jobs
- Target customers
- Technology management, platform operations, financial and operations managers of the services
- Main capabilities
- Central hub for periodic jobs, run history with business results, automatic retry of pending financial vouchers
JobSpider is the central hub for scheduling and running background jobs across the whole platform. Instead of each service (financing, investment, warehouse, receivables collection, contract management and the like) building and maintaining its own scheduler and nightly jobs separately, all of these jobs are scheduled, run and logged from one central point. The service finds each target service's address dynamically from the platform's service discovery system, with no manual or fixed addressing.
Today 39 scheduled jobs are registered in JobSpider, running across about ten different platform services: daily processing of financing installments and penalties, FantapHub's consolidated nightly settlement, ScoreCore's daily credit scoring pipeline, reconciliation and compensation of customer reward payouts, automatic retry of pending financial receipts in treasury, warehouse and sales, monthly fixed-asset depreciation, and a full set of daily and hourly receivables collection and contract management jobs (identifying arrears, calculating debt age, executing the collection strategy, reminders of contract due dates, guarantee monitoring, late-payment penalties and the like).
Every execution of a job, whether by schedule, by pressing the manual run button, or by retry, passes through one path and records exactly one history record: whether it succeeded or failed, what business summary it returned (for example, how many items were processed), and, in case of error, the message and its details. This means operations management can see from one page which jobs ran last night and what the result of each was.
Key capabilities
Central scheduling hub
All periodic platform jobs (daily, hourly, every few minutes) are scheduled and run from one service, instead of being scattered across every service.
Dynamic discovery of target services
The address of any service to be called is read from the platform's service discovery system, and not from a fixed address in configuration.
Complete history of every run
Each run of a job (scheduled, manual or retry) records a history entry with the result, a business summary and, in case of error, its details.
Business summary of each run
Instead of only succeeded or failed, each job returns its own meaningful result (for example, how many installments were processed or how many items were settled), which is visible on the history page.
Self-correcting financial jobs
Several jobs are designed specifically to automatically retry pending financial receipts (treasury, warehouse, sales, fixed assets, customer club), so that incomplete accounting vouchers do not stay pending unnoticed.
Receivables collection and contract management job chain
A coordinated set of daily and hourly jobs that run, in the correct order and one after another, the identification of arrears, calculation of debt age, execution of the collection strategy, due-date reminders and monitoring of contractual obligations.
Automatic cleanup and maintenance
Periodic deletion of old run history, and the ability to delete history records in bulk from the operations panel.
Delegated security model
JobSpider has no separate sign-in and authorization system of its own. Each service that owns the data checks the user's access itself, and then requests the operation from JobSpider service-to-service. Permission ownership therefore always stays with the specialized service.
Business value
Without a central hub, every technical team has to build its own scheduler, monitoring and run history for its service's background jobs separately. That duplicates engineering cost and leaves management without a unified view of the state of the whole platform. JobSpider solves this once for all services: a single place to see whether nightly and periodic jobs ran, what the result of each was, and where attention is needed. Its self-correcting jobs on pending financial vouchers reduce the operational risk of an accounting entry getting silently stuck.
What sets it apart
- A single point of visibility and control for all of the platform's background jobs, instead of a scheduler scattered across every service.
- History of each run with a real business result, and not only a success or failure flag.
- Automatic, safe retry jobs (with no erroneous double posting) that directly help keep financial vouchers healthy.
- A security model that always keeps the access decision with the service that owns the data, and never in the central scheduler itself.
See this module on demo data
In a demo session we walk through your organization's scenarios on Dara's demo environment and answer your technical and finance teams' questions.