What is Teradata Viewpoint?
Teradata Viewpoint is a browser-based monitoring and administration interface built from portlets. For query diagnosis, its value is connecting a late report to a session, its current state and the evidence needed for the next investigation. Available systems, portlets and actions depend on your installation and assigned roles.
Worked example: investigate a late report in Viewpoint
Documentation basis: Teradata Viewpoint User Guide 26.07, published August 2026. This is an illustrative workflow for a Teradata Database system, not a test performed on a customer system. Menu names and capabilities should be checked against your installed release.
Suppose a scheduled sales report has missed its expected delivery time. The reporting team supplies the database username, approximate submission time and, if recorded by the application, its session identifier or query band. The first question is whether the database request is queued, blocked, executing or already returning results.
1. Find the report’s session
Open Query Monitor, select the correct system and choose By Session > All. Use the table filters to narrow the list by User or Session ID; compare Start and any available Query Band with the reporting team’s information. Click the matching session row for details. A shared reporting username alone does not establish that you have found the correct request.
Confirm the monitoring data is current. The Sessions collector supplies sampled session information. Teradata documents that an inactive collector can leave zero sessions or stale data. Collection intervals also limit what you can observe; a request can finish between samples.
Get the complete Teradata tuning book, free
Teradata Query Performance Tuning: The DWHPro Guide
By Roland Wenzlofsky & Artemios Vogiatzis
206 pages on statistics and demographics, access paths and indexes, query execution plans, and joins. Get the full PDF and practical Teradata updates from DWHPro.
Confirm your email to receive your download link. Privacy policy. Unsubscribe in any newsletter.
About the free book2. Separate waiting from execution
Use State and In State, rather than treating Duration as proof of expensive execution. The documented state filter bar distinguishes Active, Block, Delay, Parse and Resp, among other states. Select a state to narrow the session table.
On a narrow screen, scroll the table sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Observed state | Interpretation | Next investigation |
|---|---|---|
| Delay | The request is waiting to be processed. | Inspect the available Delay tab and workload context; ask the workload administrator to investigate admission rules and concurrency. |
| Block | The request is waiting for a locked resource. | Inspect Blocked By and identify the blocking session or chain before discussing transaction ownership with the application team. |
| Active | The request is processing. | Compare CPU, I/O, spool and skew across samples, then inspect available SQL and Explain details. |
| Resp | Execution has completed and spooled results are being returned. | Investigate result volume and application fetching alongside the reporting team. |
For example, finding the report in Block would change the investigation from rewriting its SQL to understanding the blocking transaction. Query Monitor also offers By Blocker; the documented blocker information distinguishes root causes from sessions that are themselves blocked. Escalate with the session identifiers and observation time. An abort, workload change or release from a queue requires a separate operational decision and appropriate authority.
3. Inspect metrics that answer the question
For an Active request, use Configure Columns to expose relevant Teradata session metrics. Examples include Request CPU, Request I/O, Spool, Request CPU Skew and Request I/O Skew. The delta metrics, Δ CPU and Δ I/O, describe activity since the last sample; they are different from request totals. A large value identifies a lead to investigate, not a diagnosis by itself.
Open the session’s SQL, Explain or Skew details where available. SQL and Explain collection depends on administrator thresholds and collection for that period. An absent tab is not evidence that the request had no SQL or execution plan. Use Explain’s estimated and available actual rows to identify a step needing closer analysis; do not equate its Actual Time, documented as CPU time, with report wall-clock duration.
4. Follow the request into retained history
If the request has finished, use the Query Log portlet where configured. Select the relevant date, filter the logged queries and drill into the matching request. Query Log requires database query logging enabled, PDCR archival and collected data retained in Viewpoint. Its date selector is separate from Rewind. Missing history can therefore be a logging, collection or retention gap.
In release 26.07, the separate Completed Queries portlet documents Aster, Machine Learning Engine and QueryGrid processes; do not assume it is the history portlet for every Teradata SQL request. For database logging fundamentals, see Teradata DBQL query logging.
5. Leave with a specific next investigation
Record the system, request identity, observation time, state and supporting metrics. Route queueing to workload review, blocking to transaction analysis, and sustained execution cost to SQL, plan, statistics and data-distribution analysis. Compare equivalent completed runs only when suitable history exists. For help turning those observations into a measured improvement, see Teradata performance tuning services.
Viewpoint is where two other subjects become visible: how load utilities are classified as their own workload and what a hot standby node does when an AMP fails.
Two of the things Viewpoint surfaces have their own articles: the space limits behind “no more room in database” and how access rights are granted and inherited.
Deciding what to change in your data platform?
Compare optimisation, modernisation and migration against your workload and business requirements.
Get the complete Teradata tuning book, free
Teradata Query Performance Tuning: The DWHPro Guide
By Roland Wenzlofsky & Artemios Vogiatzis
206 pages on statistics and demographics, access paths and indexes, query execution plans, and joins. Get the full PDF and practical Teradata updates from DWHPro.
Confirm your email to receive your download link. Privacy policy. Unsubscribe in any newsletter.
About the free book