PEOPLESOFT ETL CONNECTOR

    PeopleSoft ETL Connector — Production-Grade, PS_-Native, Audit-Signed

    Pre-built peoplesoft etl connector. Full PS_ table catalog (FSCM + HCM + Campus), scheduled parallel jobs, effective-dated history native, audit-signed manifests, PSOPRDEFN read-only security, FBDI/HDL/Parquet/JSON output. Deploy in 2–4 days.

    PS_ catalog
    FSCM + HCM + Campus tables
    Parallel jobs
    Up to 64 worker pods
    Signed manifests
    Tamper-evident audit chain
    2–4 days
    Install to production

    Why a purpose-built peoplesoft etl connector matters

    Generic ETL tools have PeopleSoft adapters that handle basic table read. They don't know SetID share scope, effective-dated history, PeopleCode-derived defaults, or PSOPRDEFN security model. The result is data that loads successfully and means the wrong thing.

    Informatica, Talend, Pentaho and the rest ship PeopleSoft adapters that authenticate to the DB and execute SQL against PS_ tables. They handle the connection. They don't handle the semantics. A query against PS_VENDOR returns rows that need additional SetID-aware filtering — without it, you get vendor rows from SetIDs your target Business Unit shouldn't see. A query against PS_JOB returns current rows only unless you write custom EFFDT join logic — without it, you miss the effective-dated history that drives HCM analytics. A query against PS_JRNL_LN returns ChartField columns whose values were derived by PeopleCode and don't match what the journal entry screen shows — without documentation, your downstream reconciliation breaks.

    Syntra ETL's peoplesoft etl connector ships PeopleSoft-native semantics built in. SetID-aware extraction with target-environment routing. Effective-dated history as a first-class extraction option with explicit current-only / full-history / as-of-date choice per PS_ table. PeopleCode-derived defaults documented and surfaced in metadata, so the downstream consumer knows what the value means. PSOPRDEFN security model respected: read-only DB credential with Permission List minimization, audit logging of every PS_ table access for SOC 2 compliance.

    Quarterly updates track PeopleTools PUM and PeopleSoft 9.2 image releases — when SAP/Oracle publishes a new PS_ table column or alters a column semantic, the peoplesoft etl connector ships an update with the change documented and regression-tested. No maintenance burden on the customer. No chasing PeopleTools release notes. The peoplesoft etl connector backed by an SLA.

    What the peoplesoft etl connector delivers

    1
    Full PS_ catalog
    FSCM (PS_LEDGER, PS_JRNL, PS_VOUCHER, PS_VENDOR, PS_ASSET) + HCM (PS_EMPLOYEES, PS_JOB, PS_PAY_CHECK) + Campus (PS_STDNT_*). Quarterly updates per PUM release.
    2
    PeopleSoft-native semantics
    SetID-aware, effective-dated, PeopleCode-aware, PSOPRDEFN-scoped. Generic ETL adapters don't know these — they get them wrong silently.
    3
    Scheduled parallel jobs
    Cron-based scheduler. Up to 64 worker pods running concurrent extractions. Per-PS_-table modified-since delta watermark.
    4
    Audit-signed manifests
    Every run produces hash-signed manifest in tamper-evident chain. SOX/IRS/FERPA retention evidence ready.

    The peoplesoft etl connector — six core capabilities

    What ships pre-built. No bespoke SQR or App Engine development. No multi-month PeopleSoft adapter project.

    📚

    Full PS_ catalog coverage

    Every production PS_ table across FSCM (GL, AP, AR, Assets, Procurement), HCM (Workers, Jobs, Payroll, Time, Absence) and Campus (Students, Enrollment, Financial Aid). Plus PSDBFIELD/PSRECDEFN metadata for schema-aware extraction.

    🔌

    Connection patterns

    Direct DB connection (Oracle, SQL Server, DB2) with read-only PSOPRDEFN credential. Integration Broker REST Service Operations for record-level access. Component Interface APIs where business-logic-aware access is needed.

    📅

    Effective-dated native

    EFFDT/EFFSEQ history handled as first-class extraction option. Per PS_ table: current-only, full-history, or as-of-date snapshot. Original audit columns preserved.

    🗂️

    SetID-aware routing

    Every PS_VENDOR/PS_CUSTOMER/PS_ASSET row tagged with SetID context for target-environment routing. Generic ETL adapters miss this and ship cross-SetID contamination.

    Parallel job orchestration

    Configurable worker pool (default 8, scalable to 64). Per-table watermark-driven delta extraction. Independent failure recovery per worker. Idempotent re-runs.

    🔏

    Audit-signed manifests

    Per-run JSON manifest with row counts, sum totals, hash signatures, watermark range, PSOPRDEFN audit context. Tamper-evident chain — each run references previous run hash.

    The peoplesoft etl connector — install to production in five steps

    From PSOPRDEFN provisioning to first scheduled delta run in production, typically 2–4 days.

    1

    PSOPRDEFN provisioning — Day 1

    PeopleSoft admin provisions read-only PSOPRDEFN account scoped via Operator Class + Permission Lists to the minimum PS_ table set in your extraction plan. Read-only DB credential stored in cloud KMS — Syntra never holds it in plaintext.

    2

    Connector deployment — Day 1–2

    Connector runtime deployed to your cloud or on-prem environment (Kubernetes, ECS, Cloud Run, bare VM). Output destinations configured: S3/GCS/Azure Blob for files, Snowflake/BigQuery/Redshift for warehouse direct, Fusion FBDI/HDL drop for migration.

    3

    Scope + schedule config — Day 2

    Per-PS_-table extraction scope: fiscal years, business units, SetIDs, full-history vs current-only. Per-table output format: Parquet, JSON Lines, FBDI, HDL. Schedule per table: one-shot, hourly, nightly, weekly, monthly.

    4

    First bulk extract — Days 2–4

    Initial full-snapshot extraction across all configured PS_ tables in parallel. Multi-year PS_LEDGER and PS_PAY_CHECK history throttled to off-peak windows. Per-partition signed manifests produced.

    5

    Steady-state production — Day 4 onward

    Scheduled delta runs execute on cron, capturing modified-since rows since last watermark. Prometheus metrics live. Grafana dashboards live. Alerts route via email/Slack/PagerDuty. Failures surface as tickets — no silent drift.

    Output formats supported by the peoplesoft etl connector

    Configurable per PS_ table per run. Multiple outputs per run supported (e.g. Parquet to warehouse AND FBDI to Fusion).

    📊

    Parquet → object storage

    Columnar Parquet partitioned by Business Unit + Fiscal Year + Period to S3/GCS/Azure Blob. Ideal for downstream analytics in Athena, BigQuery, Snowflake, Spark.

    📡

    JSON Lines → streaming

    Newline-delimited JSON preserving full PS_ table row shape to Kafka, Kinesis, Pub-Sub. Ideal for event-driven downstream ETL pipelines.

    📦

    FBDI → Oracle Fusion

    FBDI ZIPs for GL, AP, AR, Assets, Procurement. Validated against current Fusion 26x release schema. Ready to drop in Fusion ESS submission folder.

    👤

    HDL → Oracle Fusion HCM

    HDL bundles for Worker, Assignment, Position, Element Entry. Date-effective HISTORY_DATA segments preserved. Ready for HDL job submission.

    🗄️

    JDBC → data warehouse

    Direct JDBC load to Snowflake, BigQuery, Redshift, Synapse with automatic schema creation and incremental MERGE based on watermark.

    ❄️

    Tiered archive

    Long-term archive to S3 Glacier, GCS Coldline, Azure Archive for SOX (7yr), IRS (4–7yr), FERPA (indefinite), state public-records (10–30+yr) retention compliance.

    Frequently asked questions

    What is a peoplesoft etl connector and what does it do?+

    A peoplesoft etl connector is a packaged software component that authenticates to a PeopleSoft 9.2 environment, extracts data from PS_ tables and Integration Broker service operations, transforms it through configurable mapping rules, and loads it into a target destination (Oracle Fusion FBDI/HDL, data warehouse, lake, downstream ETL pipeline). Syntra ETL's peoplesoft etl connector ships pre-built support for the full PS_ table catalog (FSCM + HCM + Campus), every supported pagination and effective-dated pattern, scheduled batch + real-time delta capture, PSOPRDEFN-scoped read-only security, multiple output formats (Parquet/JSON/FBDI/HDL), Prometheus observability, SOC 2 audit logging, and an SLA. Customers deploy it in 2–4 days versus the 12–20 weeks a hand-built equivalent typically takes.

    What PeopleSoft tables and APIs does the peoplesoft etl connector support out of the box?+

    The full production PS_ table footprint across FSCM, HCM and Campus Solutions. Financials: PS_LEDGER, PS_LEDGER_BUDG, PS_LEDGER_ADB, PS_JRNL_HEADER, PS_JRNL_LN, PS_VOUCHER, PS_VOUCHER_LINE, PS_VCHR_ACCTG_LINE, PS_VCHR_DST, PS_VENDOR, PS_VNDR_LOC, PS_VNDR_ADDR, PS_VNDR_BANK_ACCT, PS_PYMNT_VCHR_XREF, PS_CUSTOMER, PS_CUST_OPTION, PS_AR_ITEM, PS_ASSET, PS_BOOK, PS_DEPR, PS_PO_HDR, PS_PO_LINE, PS_REQ_HDR. HCM: PS_EMPLOYEES, PS_PERSON, PS_NAMES, PS_ADDRESSES, PS_PERSONAL_DATA, PS_JOB, PS_POSITION_DATA, PS_DEPT_TBL, PS_JOBCODE_TBL, PS_LOCATION_TBL, PS_PAY_CHECK, PS_PAY_EARNINGS, PS_ABSENCE_HIST, PS_TL_RPTD_TIME. Campus: PS_STDNT_CAR_TERM, PS_STDNT_ENRL, PS_ITEM_SF, PS_ACAD_PROG, PS_TRANSCRIPT. Plus PSDBFIELD/PSRECDEFN/PSRECFIELD metadata. Integration Broker REST service operations and Component Interface APIs for record-level access where direct table access is restricted.

    How does the peoplesoft etl connector handle scheduling and parallel extraction jobs?+

    Built-in scheduler with cron syntax. Per-PS_-table extraction scope: which fiscal years, which business units, which SetIDs, full effective-dated history or current-only. Parallel job orchestration: multiple PS_ tables extract concurrently across configurable worker pool (default 8 workers, scalable to 64). Per-table modified-since watermark for delta extraction (LASTUPDDTTM, LASTUPDOPRID, or EFFDT/EFFSEQ pairs for effective-dated). Common schedules: nightly delta extract feeding data warehouse, weekly full-snapshot for backup, monthly PS_LEDGER and PS_PAY_CHECK archive batch for SOX/IRS retention compliance. Each scheduled run produces a signed manifest with counts, sums and hashes per partition; run logs feed SOC 2 audit trail; failures alert via email/Slack/PagerDuty/webhook.

    What audit-signed manifests does the peoplesoft etl connector produce?+

    Every extraction run produces a signed JSON manifest per PS_ table per partition. Manifest contents: PS_ table name, partition key (Business Unit + Fiscal Year + Period typically), row count extracted, sum totals for key columns (POSTED_TOTAL_AMT, GROSS_AMT, MONETARY_AMOUNT, HEADCOUNT), per-row hash signatures collected as a Merkle root, source LASTUPDDTTM watermark range, PSOPRDEFN audit context (user, session timestamp, IP), output file paths with hash signatures, total extraction wall-clock time. Manifests are written to immutable cloud object storage with KMS-encrypted signatures and chained — each run's manifest references the previous run's manifest hash, forming a tamper-evident audit chain. Critical for SOX retention evidence and for downstream reconciliation when the peoplesoft etl connector feeds a Fusion migration.

    What's the difference between Syntra's peoplesoft etl connector and a generic ETL tool with a PeopleSoft adapter?+

    Generic ETL tools (Informatica, Talend, Pentaho) ship PeopleSoft adapters that handle basic table read — but they don't know PeopleSoft semantics. They don't understand SetID share scope, so a query against PS_VENDOR returns rows that need additional SetID-aware filtering. They don't handle effective-dated history natively, so PS_JOB extract returns current rows only unless you write custom EFFDT join logic. They don't recognize PeopleCode-derived defaults, so values in the table differ from values in the application UI without explanation. Syntra ETL's peoplesoft etl connector ships PeopleSoft-native semantics: SetID-aware extraction with target-environment routing, effective-dated history as a first-class extraction option, PeopleCode-derived defaults documented and surfaced in metadata. Plus quarterly updates tracking PeopleTools releases — no maintenance burden on the customer.

    Does the peoplesoft etl connector require deployment inside our PeopleSoft environment?+

    No. The peoplesoft etl connector runs in your cloud or on-prem environment (containerized for Kubernetes, ECS, Cloud Run, or bare VM) and connects to the PeopleSoft database tier via a standard read-only DB credential or to PeopleSoft Application Server via Integration Broker REST service operations. No code deployed to PeopleSoft. No PeopleTools changes required. No PSOPRDEFN account changes beyond provisioning a read-only credential. The peoplesoft etl connector leaves no trace inside the PeopleSoft environment beyond the standard PSOPRDEFN audit log entries that any DB session produces. Critical for change-control-sensitive environments (regulated industries, government) where any in-PeopleSoft code change requires a quarterly release cycle.

    How does the peoplesoft etl connector handle PeopleSoft attachments?+

    PeopleSoft attachments (PS_ATTACHMENTS, PS_ATT_VERSIONS) are streamed by the peoplesoft etl connector as binary objects to cloud object storage (S3, GCS, Azure Blob) with a hash-signed sidecar JSON capturing PS_ATTACHMENT key, mime type, original filename, file size, and PSOPRDEFN audit context. Attachment-to-parent-record relationship preserved as a manifest entry — PS_VOUCHER VOUCHER_ID → PS_ATTACHMENT key → object storage URL. Critical for SOX-relevant documentation chain (Fusion GL line → Journal → PS_JRNL_LN → PS_VOUCHER → PS_ATTACHMENT support PDF) and for FERPA-relevant student record retention (PS_STDNT_TRANSCRIPT → PS_ATTACHMENT transcript image). Throughput in production: 500K–2M attachments per hour per worker pod depending on average attachment size.

    What output destinations does the peoplesoft etl connector support?+

    Multiple, configurable per PS_ table per run. (1) Cloud object storage: Parquet to S3/GCS/Azure Blob, partitioned by Business Unit + Fiscal Year + Period, ideal for analytics in Athena/BigQuery/Snowflake/Spark. (2) Streaming pipelines: JSON Lines to Kafka/Kinesis/Pub-Sub, ideal for downstream event-driven ETL. (3) Oracle Fusion-native loaders: FBDI ZIPs for GL/AP/AR/Assets/Procurement, HDL bundles for HCM, REST API payloads for incremental delta loads. (4) Data warehouse direct: JDBC connectors for Snowflake, BigQuery, Redshift, Synapse with automatic schema creation and incremental MERGE. (5) Long-term archive: tiered cloud storage (S3 Glacier, GCS Coldline, Azure Archive) for SOX/IRS retention with retrieval workflow.

    Deploy the peoplesoft etl connector on your environment

    30-minute call. Walk through your PeopleSoft footprint, PS_ table volumes, output destination and SLA requirements — and have a working extract running on your dev environment within a week.