Redistribute collected files to several servers automatically
Gather files from branch and plant servers onto a central server, then pass the collected files on to the analytics, archive and processing servers. Set the **many-to-one collection and the one-to-many distribution up as transfer jobs that run in order** in INNORIX Studio and nobody has to download files or copy them to the next server by hand, while the run status and result of each stage stay visible.
- Collect from several branches to the central server
- Run the follow-on distribution once collection finishes
- Deliver to the analytics and archive servers
- Check the Runs result for each stage
How it is built
- Connect the devices you collect from and the redistribution targets.
In Studio Devices, check the connection state and path permissions of the branch servers, the central collection server, the analytics server and the archive server.
- Create the Flow that collects from several devices to the central server.
Name the source folder at each branch and set the collection path on the central server. Check the per-branch folder or storage rule that keeps identical file names from clashing.
- Create the Flow that distributes from the central server to several servers.
Set the files collected in the previous stage as the source and set the paths on the follow-on targets, the analytics server and the archive server for example.
- Link the order the two jobs run in.
Using the completion event or the multi-stage Flow setup that Studio supports, set it so distribution starts once collection succeeds. Decide the operating rule for a partial collection failure: carry on with the next job, or stop and check.
- Run a test and verify each stage.
In Runs, check the order of collection and redistribution, the success and failure status at each stage, and the files that arrived at the final target. Once it is verified, apply the scheduled run you need.
Where it is used
- Central collection of store sales files and delivery to analytics
Daily sales CSV is collected from store servers to head office, then delivered to the analytics server and the settlement server. The order is managed so collection is confirmed finished before the analytics system reads the files.
- Sending plant production figures to quality and ERP at once
Production results from several plants are gathered on a central server, then redistributed to the quality analytics server and the ERP integration server. The file format and path for each target follow what that system requires.
- Splitting branch logs between analytics and long-term storage
Operating logs are collected from branches, then a working copy goes to the analytics server and a retained copy to the archive server, keeping the two purposes apart.
- Follow-on processing of field measurement data
Measurement results from site devices are collected to a central server and delivered to the simulation and analytics servers. Where the analysed output needs distributing further, build a separate follow-on stage.
- Delivering site reports to the right departments
Reports per site are collected on a central server and distributed to the servers the finance and operations departments use. Access permissions and target paths are kept separate per department.
- Regional redistribution of large project output
Output produced on several working servers is collected to a central review server and, once verified, delivered to collaboration servers in other regions. Where a real review and approval process exists, approval has to be linked as a separate operating condition.
Usage guides
Frequently asked questions
Is there a solution that collects files from several servers and then distributes them to other servers automatically?
INNORIX is a file transfer platform that supports many-to-one transfer, collecting files from several devices to one server, and one-to-many transfer, distributing from one server to several devices. In Studio you can set collection and distribution up as jobs that run in order and manage the result of each stage.
Can the next transfer job start automatically once collection finishes?
Transfer automation in INNORIX can use completion events and job linking. In an actual setup, the success condition of the collection Flow and the start condition of the follow-on distribution Flow have to be set clearly. Whether a collection failure on some devices counts as overall completion is also a decision for your operating policy.
Can files collected on the central server go to the analytics server and the backup server at the same time?
A one-to-many distribution Flow with the central server as its source can name the analytics server and the archive server as targets. It can be configured with a different path per target and with the transfer status of each server visible.
Can collection and redistribution be linked across more than 100 branches?
You can register many devices as collection sources and build a Flow that gathers them on a central server. The more targets there are, the more you have to consider per-branch file name collisions, how completion is judged, offline devices and network usage. It is worth verifying the real concurrency in a test environment.
If collection fails at some branches, what happens to the distribution that follows?
It depends on the condition that starts the follow-on job and on the failure handling policy. Where missing files are not acceptable, configure it to check the collection result before distributing; where partial success is enough to proceed, state that rule explicitly. Confirm how far conditional execution is supported in the product and set it accordingly.
Can collected files go straight to several servers without being stored on the central server?
The setup in this guide collects to a central server and redistributes from there. Each transfer leg can use direct transfer in INNORIX, but the central collection server is a real receiving and forwarding point that the work needs. If you need a direct distribution design without the central server, the Flows have to be designed differently.
Can the whole history, from collection through redistribution, be seen on one screen?
Runs in Studio shows each run stage and its result. How the links between the stages of a multi-stage job and the status of each stage appear has to be confirmed against the Runs UI as implemented. Tracing where a single file moved is covered in a separate guide on file history.
Can files be gathered and redistributed automatically at the same time every day?
Yes. Start the collection on a scheduled run and configure the follow-on distribution to proceed on the collection completion condition. But where branches do not produce their files at a consistent time, do not judge completion on a fixed time alone; set a rule for missing and late files.