.svg)
PRODUCT RELEASE · OCTOBER 2026
Checks that reach your self-managed lakehouse, run history that shows exactly what happened upstream, and data quality reported at the level of the products the business uses.
An analyst who doubts a number in a report can stop and ask someone. An agent that reads the same table acts on it. That difference turns three questions reliability teams used to answer in a weekly review into questions they need answered on demand. Is this data monitored at all? Which upstream run produced it? And is the data product built on it healthy, or is nobody checking?
Most programs answer those questions in pieces. Tools that copy data into their own compute, or that see inside one platform, stop where the private-cloud lakehouse begins. Pipeline views show the latest run instead of the run that matters. Quality results sit at the rule level while the business works with data products.
ADOC 26.10.0 expands the environments teams can monitor, makes pipeline executions easier to investigate and connects quality results with the business data products that depend on them. Enterprise access, reporting, developer context and platform reliability improvements support these workflows.
ADOC 26.10.0 answers all three on the platforms you already run, across three layers:
- Coverage: bring self-managed lakehouses into monitoring through Apache Kyuubi, with every check running on Kyuubi’s own engines inside your cluster.
- Run context: trace a pipeline failure to the exact upstream run, see retries on the graph, and keep your investigation in view in large pipelines.
- Product-level quality: see data quality across your published data products in one dashboard, with five more templates for the reviews your team runs every day.
01 Coverage: your private-cloud lakehouse, in the same monitoring program
Regulated enterprises are consolidating workloads onto open lakehouses they run themselves, behind a SQL gateway that enforces strict authentication and authorization over terabytes of data a day. A tool that needs to copy that data out, or hold its own path to storage, doesn’t fit. ADOC 26.10.0 adds Apache Kyuubi as a data source. ADOC crawls the databases behind your Kyuubi gateway and monitors them with data reliability policies through the Pushdown data plane engine, so Kyuubi runs every check on its own engines and its assets join the same catalog as every other source you monitor.
What’s included:
- Pushdown execution through Kyuubi, so a platform owner adds quality coverage without giving another tool access to storage.
- Crawling of the Kyuubi databases you select, so a data engineer works from a current catalog of what sits behind the gateway.
- Profiling, sample data, data quality policies, and reconciliation, so a data engineer can inspect a table and enforce rules on it in the same workflow used for other sources.
- Optional Keycloak (OAuth2) authentication with credentials retrieved from a secret manager, so a security team approves a connection that follows its credential policies.
Full List of Acceldata Integrations
02 Run context: the more holistic pipeline investigation
When a policy fails on a downstream table, the investigation moves upstream, through DAGs that trigger other DAGs. If the graph opens the latest run instead of the one that produced the data, and hides the retries, the engineer spends the first hour rebuilding context. ADOC 26.10.0 links each dataset-triggered Airflow DAG run to the exact upstream run that started it, shows retry counts on the graph, and gives large pipelines the room to investigate.
What’s included:
- Run-level links for dataset-triggered Airflow DAGs, so a data engineer opens the exact upstream or downstream run from the lineage graph instead of the latest one. ADOC confirms links on Airflow 2.11 and later and infers them from the dataset and execution interval on Airflow 2.10.
- Task retry counts on graph nodes for Airflow and AWS Step Functions, so an on-call engineer sees which tasks succeeded only after retries and opens their events from the status icon.
- Persistent dependency highlighting, plus an Isolate option, so a platform engineer keeps the nodes under investigation and their direct connections in view while panning and zooming.
- Horizontal and vertical graph layouts, so a large pipeline shows more nodes on a wide screen with less scrolling.
- A resizable, collapsible details panel, so the Timeline has room to show task durations and parallel execution, or the graph gets the full view.
03 Product-level data quality: one dashboard for the products the business uses
Business teams consume data products, not rules. A quality program that reports at the rule level leaves product owners translating, and it makes gaps in monitoring hard to spot. ADOC 26.10.0 adds a Data Product Quality Dashboard that summarizes quality across your published data products, alongside five more templates for the reviews operation teams run every day.
What’s included:
- The Data Product Quality Dashboard, with reliability scores, assets grouped by reliability, policy execution results, and governance coverage by policy type, so a product owner reviews the whole portfolio in one place and sees where policy coverage is thin.
- Details panels and chart-segment filtering, so a product owner moves from a value on a widget to the items behind it without building a new view.
- Data Products widgets in the widget library, so a team adds product quality to any existing dashboard.
- Schema Drift, Profiling, Freshness, and DQ Freshness Check templates, each with a description of the data it shows, so an operations team starts its daily review from a ready-made view.
- A Monthly Data Quality KPI Report template, so a governance lead brings quality trends to the monthly review with data owners.
In this release
Full details for each capability are in the documentation below.
- Apache Kyuubi Data Source, New, 26.10.0
Checks that run inside your private cloud. ADOC crawls Kyuubi databases and monitors them with data reliability policies through pushdown, with optional Keycloak authentication and secret manager credentials.
→ Read the full Kyuubi announcement
→ Docs: https://docs.acceldata.io/adoc/documentation/apache-kyuubi - Pipeline Run Context, Improvement, 26.10.0
The exact run behind a failed table. Run-level links for dataset-triggered Airflow DAGs, retry counts on the graph, persistent highlighting, and layouts built for large pipelines.
→ Docs: https://docs.acceldata.io/adoc/documentation/pipeline-run-details
→ Airflow run links: https://docs.acceldata.io/adoc/documentation/airflow - Data Product Quality Dashboard and New Templates, New, 26.10.0
Quality across your published data products in one view, plus Schema Drift, Profiling, Freshness, DQ Freshness Check, and Monthly KPI templates.
→ Docs: https://docs.acceldata.io/adoc/documentation/data-reliability-tab
