Connectors Infor LN and M3 to Azure Data Lake Storage Gen2

Infor LN and M3 to Azure Data Lake Storage Gen2: full history landed and reconciled, then kept current.

Lima Backbone reads Infor LN and M3 through ION API gateway, types the data correctly on landing, writes it into Azure Data Lake Storage Gen2 as containers with raw and history paths, and keeps it current from the source’s changes. Every run ends with a row count and a control total per table.

OUT OF INFOR LN AND M3ItemsItem master · item warehouseBills of materialProduct structure BODsRoutingsRouting BODsStockInventory balance BODsLIMA BACKBONEExtractION API gateway · Business object docume…Checknine checks before anything loadsShapetypes, keys, deletes, partitionsLoadParallel partitioned writes with a manif…INTO AZURE DATA LAKE STORAGE GEN2Azure Data Lake Storage Gen2Containers with raw and history paths; P…
The path from Infor LN and M3 to Azure Data Lake Storage Gen2. Objects come out through the source's methods, are typed and checked, land in Azure Data Lake Storage Gen2 and are kept current from the source's changes.
220M rows
615 tables · ~30 GB in ~30 min on 64 vCPU
performance estimate
80 GB/day
moved and transformed every day
production · semiconductor enterprise
8
checks on landing
product
1
count and control total per table per run
product
What Azure Data Lake Storage Gen2 expects

The model the load has to fit.

ModelContainers with raw and history paths; Parquet or Delta partitioned by date and key.LoadParallel partitioned writes with a manifest; counts and control totals per table.BackRestore on demand into any target.
The hardest part of this path

Keeping file sizes right for the engines that read them.

The Azure Data Lake Storage Gen2 connector page →

What comes out of Infor LN and M3

Objects and where they go.

The first load is assembled from these objects through ION API gateway. Change capture (BOD publications) keeps the daily volume small afterwards.

Planning objectSource objectsDestination in Azure Data Lake Storage Gen2
ItemsItem master · item warehouseraw container Parquet or Delta → history
Bills of materialProduct structure BODsraw container Parquet or Delta → history
RoutingsRouting BODsraw container Parquet or Delta → history
StockInventory balance BODsraw container Parquet or Delta → history
Open purchase ordersPurchase order BODsraw container Parquet or Delta → history
Open sales ordersSales order BODsraw container Parquet or Delta → history
Production ordersProduction order BODsraw container Parquet or Delta → history
ForecastsForecast BODsraw container Parquet or Delta → history
Customers and suppliersBusiness partner BODsraw container Parquet or Delta → history
CalendarsCalendar objectsraw container Parquet or Delta → history
The path, step by step

Six steps, one report at the end.

ConnectRegister Infor LN and M3 and Azure Data Lake Storage Gen2 with credentials from the vault; test both connections.
ExtractFirst load through ION API gateway; then BOD publications.
CheckNine checks run before anything loads; a failed row becomes an exception with an owner and the load continues with the valid rows.
ShapeTypes are set from the source, keys kept complete, deletes and late changes applied in order.
LoadParallel partitioned writes with a manifest; counts and control totals per table.
Keep currentChanges from Infor LN and M3 are applied on the schedule you set; every run ends with the same report.
The checks that run before anything loads

8 checks, each with a plain example.

Every check either passes a row or turns it into an exception record with an owner. Nothing is dropped silently.

CheckKindWhat it catches, with an example
Types set from the sourcetypingSAP dates and times stored as text become dates and times; currency decimals shifted by the currency's rule
Keys complete and uniquekeysclient or organisation columns kept in the key; zero-padded keys preserved
Deletes appliedconsistencyrows deleted in the source are marked or removed in the copy
Late changes land in orderconsistencya change that arrives after a later one is applied by its source timestamp
Wide and cluster tables completecompletenessevery column of a wide table lands; cluster tables read through the released module
Partitions cover the rangecompletenessevery date partition present; gaps reported
Counts and control totals matchreconciliationrows and amounts per table compared between source and target
Schema changes noticeddrifta new or changed column is flagged before it breaks a downstream query
Cadence and offer

How the copy stays current, and how the work is priced.

After the first landing, Azure Data Lake Storage Gen2 receives the source's changes: BOD publications. A migration month costs the same as a quiet month on the same tier.

Terms on this page

Plain meanings.

LandingWriting rows into the target's raw layer as they arrived, typed, before any business transformation.
Referential orderLoading parents before children so every row finds what it refers to.
Control totalThe sum of a key amount column compared between source and target, along with the row count.
Exception recordA row that failed a check: kept, routed to an owner, replayed once fixed.
Change captureReading only the rows that changed, from a log, change pointers or timestamps.
Questions

Asked about this path.

How long until the first load is live?

Days for the first tables; a full-history migration is sized by table count and volume, and read in partitions inside the run windows your administrators allow.

What does the data team see?

Counts per step, the check that stopped a row, the owner and the replay point, and the lineage of any number back to its source column.

Who runs it after go-live?

Your team, on the schedule you set, with the same report after every run; or your systems integrator; or Lima's team under the subscription.

Land your first tables in Azure Data Lake Storage Gen2.

One call scopes the dataset, the run window and the tier. The reconciliation report at the end is the acceptance test.