SIEMENS

Teamcenter 12.2

Cacheless Search
Deployment
PLM00301 • 12.2
Contents

Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

What is cacheless search? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
How cacheless search works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Using cacheless search in rich client and Active Workspace . . . . . . . . . . . . . . . . . . . . . . 1-5
Using JT based cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6
Using NX based cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6
Understanding bounding boxes and TSO files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Understanding spatial search indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Tuning the indexer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Understanding cacheless search and Multi-Site Collaboration . . . . . . . . . . . . . . . . . . . . 1-11
Deploying cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-11
Setting up product-scoped cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-12
Setting up full database cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-13

Working with Product Scoped Cacheless Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

Preparing your data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Upgrade legacy transforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Find and clean up cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Check and generate missing UGPART-BBOX creation form (NX) . . . . . . . . . . . . . . . . . . . 2-2
Generate TruShape (TSO) files (NX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Enabling product-scoped cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Make the product indexable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Creating bounding boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Creating bounding boxes for NX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
Creating bounding boxes and TSO files for JT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6
Creating a spatial index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Populating the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Listing the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
Processing the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
Deleting an index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Setting up live background updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Setting up Dispatcher to manage updates (NX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Setting up Dispatcher to manage updates (JT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
Setting up cron jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
Configuring cacheless search in a Multi-Site Collaboration environment . . . . . . . . . . . . . . . . 2-13

Working with full database cacheless search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1

Preparing your data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Upgrade legacy transforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Find and clean up cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Check and generate missing UGPART-BBOX creation form (NX) . . . . . . . . . . . . . . . . . . . 3-2
Generate TruShape (TSO) files (NX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3

PLM00301 12.2 Setting Up Cacheless Search 3

 Contents
Contents

Creating bounding boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3

Creating bounding boxes for NX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Creating bounding boxes and TSO files for JT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Creating a spatial index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Populating the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Listing the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Processing the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Deleting all indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
Setting up live background updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
Setting up Dispatcher to manage updates (NX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
Setting up Dispatcher to manage updates (JT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
Setting up cron jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
Configuring cacheless search in a Multi-Site Collaboration environment . . . . . . . . . . . . . . . . 3-12

Oracle Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
How do I validate a spatial index? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
Why is qsearch_process_queue taking so long? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
How do I improve search engine performance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
Why is my search slow? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2

4 Setting Up Cacheless Search PLM00301 12.2

Chapter 1: Getting Started

What is cacheless search?

Cacheless search allows you to perform spatial search of the live (latest) configured product data
that is persistent in the Teamcenter database. It is not necessary to maintain secondary structure
data caches or to create a separate database.
The cacheless search engine must be used instead of the QPL and appearance-based search
engines that were available in earlier versions of Teamcenter. All the functionality provided by the
earlier search engines is also available with cacheless search. Unlike appearance and QPL searches,
cacheless searches are performed on recent data, and not on data that was saved the previous day.
The audience for this information is administrators and Teamcenter services personnel who are
tasked with deploying cacheless search.
You can configure cacheless search for one of three modes:
• NX based
Configure this mode if you exclusively use NX as your CAD solution.

• JT based
Configure this mode if you exclusively use CAD solutions other than NX.

• Mixed
If you work in a multi-CAD environment, you must configure both NX and JT modes.

To use cacheless search, populate the database with the following objects:
• Teamcenter bounding boxes (bbox data)
Bounding boxes must be created for all structures you want to search. Bounding boxes are
attached to their owning item revisions and to their geometry source, either a CAD or JT dataset.
To create bounding boxes, run the create_or_update_bbox_and_tso utility.

• (Optional) TruShape (TSO) voxel files. TruShape files provide more accurate results but occupy
additional space in the database.
To create TSO files, run the create_or_update_bbox_and_tso utility.

• Spatial search indexes

• Index queue
Teamcenter creates the index queue to hold objects that require their indexes to be created or
updated. Depending on your deployment strategy, the items may be restricted to one or more
selected products in the database.

PLM00301 12.2 Setting Up Cacheless Search 1-1

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

Note
This queue is transient and gets used by the index update process as the objects in
the queue get processed and incorporated in the spatial index.

Once you have generated the initial data, you must keep it updated to reflect changes to the CAD
designs. There are two methods of doing this:
• Using Dispatcher

• Using a cron job or service

You can choose one of two deployment strategies when creating bounding boxes and search indexes:
• Full database
Use this strategy if you have relatively small products or designs for only a few products.

• Product-scoped cacheless search

Use this strategy if you want to exclude data from products that are complete or inactive from the
search results.

How cacheless search works

Cacheless searches allow you to query the configured product structure in Structure Manager,
Design Context, and several other applications to retrieve matching items and item revisions. You
can specify any combination of the following search parameters:
• Spatial parameters including:
o Box volumes, either inside, outside, or intersecting the box

o Above, below, or intersecting defined planes

o Proximity to another object

o Approximate size

• Attributes including item or item revision ID, name, occurrence notes, and form attributes

• Classification attributes

• Saved queries

Teamcenter uses bounding boxes and TruShape voxel maps that are created during the processing
of CAD designs to implement cacheless searches. Internally, it uses part-level bounding boxes to
construct and maintain the necessary spatial indexes.
If you use NX and the Teamcenter Integration for NX, you can configure NX to generate bounding
box data automatically. If you search JT files generated by another CAD tool, Teamcenter generates
the bounding box data when spatial searches are first enabled. The generation process may take a
significant time if you have large quantities of data.

1-2 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

Note
If you work in NX, you should not manipulate the JT files that Teamcenter creates from the
NX files.

The search initially identifies bounding boxes that intersect the bounding boxes of the selected
objects, limiting the search results to objects with geometry in close proximity. This allows Teamcenter
to display the initial results relatively quickly.
If you choose TruShape filtering, the search then tests the objects with intersecting bounding
boxes to determine if they have intersecting TruShape parameters. Depending on your TruShape
configuration, Teamcenter may simplify the geometry of each object into regular cubes (voxels) that
provide a simplified representation of the actual shapes. The TSO files contain voxel representations
of each part that has geometry.
The accuracy of the search results depend on the cell size you define in NX or the voxel size you
configure in Teamcenter if you are using JT files. The smaller the cell or voxel size, the more accurate
the results. However, smaller sizes result in larger TSO files and slower response times.
The following actions trigger an update of the bounding box data, search index, or both:
• The creation of a new Teamcenter item causes the creation of a structure index. If your system
automatically creates a dataset under the item revision when you create items, this action triggers
the creation of a bounding box, even if the bounding box is empty. This occurs whether you
create the item directly in Teamcenter or in any of the Teamcenter integrations.

• Any structure edits, the addition or subtraction of occurrences, and absolute occurrences
modifications cause the creation of the structure index .

• The creation of an NX or JT dataset triggers the creation of the bounding box. This action also
triggers a structure update to roll the bounding box information up to its parent structures.

• Any modification of an NX or JT dataset triggers an update of the bounding box. This action also
triggers a structure update to roll the updated bounding box information up to its parent structures.

Teamcenter also makes these updates if modified data is imported from another system.

Proximity Search and TruShape

Proximity search returns parts within the proximity range from the external surface of the target
part as shown in the following figure:

PLM00301 12.2 Setting Up Cacheless Search 1-3

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

Proximity search with TruShape returns only the blue part shown in the following figure:

Proximity search without TruShape (only the bounding box filter) returns all the parts, including
the red parts, as shown in the following figure:

1-4 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

Using cacheless search in rich client and Active Workspace

Cacheless search allows you to locate sufficient information in the context of the product you are
designing or building to validate fit, form, and function for current data.
The following applications allow you to make cacheless searches.
• Structure Manager

• Design Context.

• Manufacturing Process Planner

• 4G Designer

• Active Workspace

You enter spatial criteria, attributes values, or a combination of both, and then click Search.
Teamcenter displays matching parts as a list of BOM lines. You can send selected lines to the
viewer for inspection.

PLM00301 12.2 Setting Up Cacheless Search 1-5

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

Using JT based cacheless search

If you use CAD tools other than NX, you must install the Dispatcher Server and JtToBboxAndTso
translator to create the necessary bounding box and TruShape data. You must also install the
QSEARCH_process_queue service to create an index for spatial searches. Teamcenter reads the
bounding box information from this index. Teamcenter stores the calculated TruShape data as
a Trushape-Data named reference in a DirectModel dataset.
The accuracy of the search depends on the voxel size you configure in Teamcenter. A small voxel
size provides more accurate search results, but the TruShape files are larger and the response
times, slower.

Note
The bounding box and TruShape data generated for a JT dataset are ignored if an NX
dataset exists for the same item revision.

On high performance hardware with a well-tuned database, Teamcenter can create bounding boxes
for 1,000 parts to 5,000 parts per hour from JT data.

Using NX based cacheless search

If you use NX, it calculates the bounding box for a part and stores it in Teamcenter as a
UGPART-BBOX named reference in a UGMASTER dataset. Teamcenter reads the bounding box
information from the UGPART-BBOX form. NX can also be configured to create TruShape data for
the part and store it as a Trushape-Data named reference in the UGMASTER dataset.
The accuracy of the search depends on the cell size you configure in NX. A small cell size provides
more accurate search results, but the TruShape files are larger and the response times, slower.
If you import existing NX assemblies, set the QS_NX_RELATION_POST_ACTION_ENABLED
preference to True and Teamcenter creates the necessary bounding box data automatically.

1-6 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

In many deployments, it is not necessary to use Dispatcher to periodically create and update
search indexes, as the Teamcenter Integration for NX provides a similar creation mechanism.
However, advanced NX features such as datum points, weld points, arrangements, and deformable
components are available only on JT datasets, and not on NX datasets. If you use advanced NX
features, ensure you extract the bounding box data from the JT dataset.
Use the create_or_update_bbox_and_tso utility with the JTTOBBOX+JTTOTSO option selected
to generate bounding box data that includes advanced NX features. You can then run the
qsearch_process_queue utility Dispatcher service to create the corresponding search index.
On high performance hardware with a well-tuned database, Teamcenter can create bounding boxes
for 10,000 parts to 20,000 parts per hour from NX data.

Understanding bounding boxes and TSO files

To populate the bounding boxes and TSO files, Teamcenter loads the JT file and then extracts
the bounding box information. It adds that information to the Teamcenter bounding box table
(PBOUNDINGBOX) and then creates the TSO file as a named reference in the JT dataset, as follows:

You can inspect the bounding box data for a selected part by adding bl_itemrev_bounding_box
to the displayed properties in Structure Manager. This property shows the following values of the
bounding box in the local coordinate space of the node:
• x minimum
• y minimum
• z minimum
• x maximum
• y maximum
• z maximum

PLM00301 12.2 Setting Up Cacheless Search 1-7

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

Optionally, you can refine the accuracy of the proximity search when using TSO files. To configure
the cell size of the TSO file, change the setting of the TC_JT_voxel_size preference from its
default value of 12.75 millimeters. Some testing with sample geometry may be necessary to find
the optimum cell size.

The following example shows how TSO data refines the geometry:

Understanding spatial search indexes

In its simplest form, a spatial index is the bounding box that encloses the bounding boxes of all item
revisions of an item.

1-8 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

A spatial index is also the bounding box that encloses all the extents of the constituent substructure
(that is, subassemblies).
When Teamcenter calculates spatial indexes, it rolls up the boxes up wherever they are used.
Consequently, the boxes of Rev A, Rev B, and Rev C are rolled into the index of item A. This index
is saved separately to represent item A’s volume and is then propagated upward. It is transformed
into the coordinate space of its parent in the structure. The parent’s own index is extended to
accommodate the transformed indexes of all its children.

The spatial indexes are updated when Teamcenter processes requests on the index update queue.
The update queue contains items, occurrence threads, and appearance path nodes (APNs) that
are processed in batches. The queue is processed in batches on a first in, first out (FIFO) basis.
Teamcenter does not wait until the queue is full, but is configured with a maximum batch size for
optimum scalability. Only certain types of objects are placed in the queue, for example, changes to a
bounding box attached to an item revision causes Teamcenter to place the corresponding item in the
queue. Likewise, a change to a transform causes Teamcenter to place the corresponding occurrence
thread in the queue when the occurrence is saved in the database.
You create spatial indexes when you run the qsearch_process_queue utility to process the queue
and can optionally validate the indexes by running the same utility with the check_structure_indexes
option specified.

Note
There is a single indexer (qsearch_process_queue process) for each database.

Internally, a spatial search with spatial indexes is a BOM expansion guided by the index. Teamcenter
traverses down one level and finds any indexes that meet the specific spatial criteria. It prunes out
nonmatching indexes and continues the traversal down to the next level. Pruning removes subtrees
and discards unnecessary parts of the structure, contributing to the speed of the search. It also
processes overrides similarly. After configuration of the necessary lines, Teamcenter validates the
results against actual bounding boxes and TruShape data.
To correctly configure the data, Teamcenter generates a partial BOM structure. The time it spends
performing a search typically includes:

PLM00301 12.2 Setting Up Cacheless Search 1-9

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

• Creating the partial BOM structure (50 to 70% of the total time).

• Spatial filtering by bounding boxes (10 to 20% of the total time).

• Configuring variants and processing properties (10 to 20% of the total time).

The lower the variation in geometry between different revisions of the same item, the better the
search engine performs. The greater the correlation between the product structure and the geometry
breakdown, the better the search engine performs.
Similarly, attribute searches require a partial BOM structure configuration, using objects that satisfy
the given attribute search criteria. Teamcenter first searches for the objects in the database and
then traverses up all BOM configurations. Some of the traversals may be outside of the structure
you are searching. However, implementation of product scoping avoids unnecessary traversals.
Teamcenter traverses up until it reaches the product in which the required results are found and
the final result is the same.

Tuning the indexer

On high performance hardware with a well-tuned database, Teamcenter can index 50,000 objects to
150,000 objects per hour. Actual performance on a production system depends on the rate of change
of structure and geometry, for example, occurrences, occurrence transforms, item revisions, bounding
boxes (JT and CAD), and absolute occurrence transform overrides. You can use Oracle auditing and
SQL scripts to measure the hourly or daily changes based on the last modified date of each object.
From those measurements, you can estimate the number of changed items, occurrence threads, and
APNs per hour you must index (the indexer queue only contains those objects). You can also identify
the optimum mixture of queued objects for the best performance.
You should then measure the actual maximum indexing throughput, which comprises two
considerations:
• Throughput of operations that do not alter indexes, which you can identify by running the
qsearch_process_queue utility with the –queue_update option.

• Throughput of requests that cause index changes, which you can identify with the CAD tool
or a custom script or utility.

Tip
Avoid accumulating a large number of entries in the indexing queue. The faster
updates are processed, the sooner live updates are available to users. During the
deployment process, you should monitor the size of the queue by periodically running the
qsearch_process_queue utility with the -list_queue option selected. If the queue size
is increasing steadily, you may have to reconsider your deployment strategy. In a test
environment on high performance hardware, Teamcenter can process between 60,000 and
120,000 requests per hour without the queue building up.

If you have a relatively small product structure (a thousand or less occurrence threads), you may
improve performance by switching to a top-down search. This configuration limits the scope of the
search to the product only. However, Teamcenter must first traverse the unconfigured structure in
the database to establish the search scope. This approach may be more time-consuming with large

1-10 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

product structures; if so, it may be better to tune the default bottom-up search for best performance
in such cases.
By default, Teamcenter does not include all possible indexes that you might need for attribute
queries. Additional database indexes have an associated performance cost, and you should consider
carefully whether you need them. Use the Oracle database tools in a live production environment to
assist with these decisions.

Understanding cacheless search and Multi-Site Collaboration

If you work in a Multi-Site Collaboration environment, you can make cacheless spatial and attribute
searches of products or programs that are owned or replicated at remote sites. Where appropriate,
local and remote objects are incorporated in the search results.
If you enable this feature:
• Teamcenter checks access permissions to the remote BOM lines. If you do not have the
necessary permissions, you cannot import the affected remote lines. If appropriate, you can
review the search results and import only the lines to which you have access permissions.

• Identical revision and variant rules must be defined at all sites or searches may return inconsistent
results. Also, the content of these revision rules and variant rules must always be consistent
across all the sites.

• The content of saved queries must be consistent across all sites, or searches may return
inconsistent results.

• You can only initiate the search in Design Context and not in Structure Manager or Multi-Structure
Manager.

You do not need to manually configure IDSM to use remote cacheless searches. When you install
or upgrade the system using Teamcenter Environment Manager (TEM), it installs the callbacks
necessary for remote cacheless search. Each site participating in remote cacheless search must
install or upgrade the database using the same version of TEM.

Deploying cacheless search

When you deploy cacheless search, you must choose between a product-scoped deployment
strategy and a full database deployment strategy.
• Full database deployment
Teamcenter generates bounding boxes for all geometric entities in batches, ignoring items
without geometry. It places all update candidates in the indexer queue. Update candidates
are any item revisions with bounding boxes.

• Product-scoped deployment
You mark one or more products as indexable, and Teamcenter processes only the relevant items.
No indexes are created for old and inactive programs, saving system resources.

PLM00301 12.2 Setting Up Cacheless Search 1-11

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

The feasibility of processing the complete database depends on the size and number of bounding
boxes to process. As a guideline, product-scoped deployment is probably the best strategy if less
than 30% of your data is in active use.
To obtain a baseline time, you can test a representative sample of data and then estimate the time
needed to process all the bounding boxes and indexes. For example, if the entire database cannot
be processed in one weekend, you may prefer the product-scoped deployment strategy. Even with
this strategy, the initial preparation and processing of a single product may take hours or even days,
depending on the number of components in the product.
An alternative strategy is to use a product-scoped deployment when you first populate the database,
that is, to index your data one product at a time. You can then switch to a full database deployment
when the database is populated with all products.

Setting up product-scoped cacheless search

At some customer sites, a significant proportion of the product structure data in the databases is no
longer in use, for example, it belongs to programs that are complete and no longer active. If you index
this data for spatial searches, it has an adverse impact on the speed at which indexes are created
and also on search performance. This enhancement allows you to limit spatial and attribute searches
to active programs, which may correspond to end items, top-level items, or top-level product items.
When product-scoped search is deployed, users can search on all levels of the product.
By default, all data is unsearchable and you must run the qsearch_process_queue utility on the
chosen product structures to mark all items and occurrences that are part of those structures as
indexable. The create_or_update_bbox_and_tso utility then processes only bounding boxes
attached to indexable product structures and their components. This utility also identifies missing
bounding box and TSO data for indexable products.
Once an active structure is indexed, you can make incremental updates to its spatial search index,
rather than generating a complete new index each time. Incremental index updates are restricted to
product structures previously identified as active. To process the queue, the administrator runs the
qsearch_process_queue utility with the -process_queue option.
To deploy cacheless search for a product-scoped strategy:
1. Ensure you have enabled or upgraded cacheless search with Teamcenter Environment Manager
(TEM) before you begin the deployment procedure.

2. Ensure that live background updates are turned off while you manually process the product data.

3. Prepare your data for processing by updating legacy transforms and removing cycles from the
database. This step must be completed carefully or the results of searches may be incorrect.

4. Enable product-scoped cacheless search.

5. Make the product indexable. This step makes all components of the chosen product available for
cacheless searches.

6. Create the bounding boxes and TSO files for the chosen product.

7. After the bounding boxes and TSO files are available, populate the queue with the
entries for the chosen product. Do not run the qsearch_process_queue utility until the
create_or_update_bbox_and_tso has finished.

1-12 Setting Up Cacheless Search PLM00301 12.2

 Getting Started

8. Process the queue for the chosen product.

9. If you want to make other products in the database available for cacheless searches, repeat steps
3 through 6 for each product in turn.

10. Query for and process any bounding boxes remaining in the database that do not belong to the
products you have already processed.

11. Configure live background updates with Dispatcher or cron jobs to process changes to the
indexed data.

Note
In a typical system, you may run a cron job every 15 minutes to process updates. The
value could be adjusted, depending on how frequently data changes and the extent
of those changes.
Always turn off live background updates if you run the qsearch_process_queue utility
to manually process the queue.

Setting up full database cacheless search

Siemens PLM Software recommends that you process both the bounding boxes and spatial indexes
in batches if the database has a large number of datasets. The generation of bounding boxes and
spatial indexes in batches helps prevent the utilities from running out of memory or causing other
scalability issues. If you have a small database of less than 50,000 UIDs, you may process all
the datasets without any batching.
To deploy cacheless search for a full database strategy:
1. Ensure that you have enabled or upgraded cacheless search with Teamcenter Environment
Manager (TEM) before you begin the deployment procedure.

2. Ensure that live background updates are turned off while you manually process the product data.

3. Prepare your data for processing, by updating legacy transforms and removing cycles from the
database. This step must be completed carefully or the results of searches may be incorrect.

4. Query for a list of dataset UIDs to process for the entire database.

5. Create the bounding boxes and TSO files for the database.

6. After creating the bounding boxes, regenerate the database statistics to improve search
performance.

7. Populate the queue with the entries for the entire database.

8. Process the queue you populated in the previous step. Do not run the qsearch_process_queue
utility until the create_or_update_bbox_and_tso has finished.

9. After generating the indexes for a product, regenerate the database statistics to improve search
performance

PLM00301 12.2 Setting Up Cacheless Search 1-13

 Chapter
Chapter 1: 1: Getting
Getting Started
Started

10. Configure live background updates with Dispatcher or cron jobs to process changes to the
indexed data.

Note
In a typical system, you may run a cron job every 15 minutes to process updates. The
value could be adjusted, depending on how frequently data changes and the extent
of those changes.
Always turn off live background updates if you run the qsearch_process_queue utility
to manually process the queue.

1-14 Setting Up Cacheless Search PLM00301 12.2

Chapter 2: Working with Product Scoped Cacheless Search

Preparing your data

Before you use cacheless search on an existing database, you must run a few checks and clean the
data.
To begin with, you must upgrade all existing legacy transforms, since the cacheless search
functionality uses only the new PLM XML transform. Next, you must run the qsearch_process_queue
utility to find and clean up all cycles (circular references). Also, check that all NX datasets have
the UGPART-BBOX form.

Note
Ensure that you have the typical Teamcenter environment variables set for all command
line utilities.

Upgrade legacy transforms

1. Find all legacy transforms by running the -list_legacy_transforms option in the
qsearch_process_queue utility in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-list_legacy_transforms -outfile=c:\temp\legacytransforms.txt.

2. Run the nxmgr_update_transforms utility to upgrade the transforms in the following format:
nxmgr_upgrade_transforms -u={USER_ID} -p={PASSWORD} -g={GROUP} -bypass=yes
-i=c:\temp\legacytransforms.txt -upgrade_release=yes.

3. Run the -list_legacy_transforms option again to ensure that all transforms are upgraded.

Find and clean up cycles

1. Run the find_cycles option in the qsearch_process_queue utility to find all the cycles.
To find cycles in the entire database, run qsearch_process_queue -u={USER_ID}
-p={PASSWORD} -g={GROUP} -find_cycles 2> c:\temp\list_of_cycles.txt.
Processing the entire database might take a long time. Instead, you can find cycles by item
ID, list of item IDs, or item ID UIDs.
To find a cycle by item id, run qsearch_process_queue -u={USER_ID} -p={PASSWORD}
-g={GROUP} -find_cycles–item_id=AKT75443 2>c:\temp\AKT754432_cycles.txt.

2. Clean up the cycles as per your organization’s plan, considering that cleaning up cycles affects
the content of the data.

PLM00301 12.2 Setting Up Cacheless Search 2-1

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

Tip
(Product-scoped cacheless searches only)
The -check_structure_indexes argument of the qsearch_process_queue utility flags
cycles in a structure with an index-box-check found cycle report. This argument starts
the utility at one or more specified items and works down the entire substructure. It finds
apparently incorrect indexes and, at the same time, also finds cycles. Conversely, the
find_cycles argument works up the substructure from the specified item; if you specify
only top-level items, it returns no results. Also, the find_cycles argument takes all items
by default.
In general, arguments with structure in the name interpret given item IDs as top-level
items, and perform the relevant action on the entire structure below each item. Options
without structure in the name simply perform the relevant action on each item in isolation.
Therefore, the -check_indexes argument simply checks each item's index but does not
find any cycles; the -check_structure_indexes argument also identifies any cycles.
To check for cycles under a top-level item, enter a command similar to the following
example:
qsearch_process_queue -dump_substructure -item_id=top-level-item-id(s) 2>&1
) | grep " item " | sed -e "s/ *item *//" | sort -u >all_substructure_item_uids_file
qsearch_process_queue -find_cycles -uid=@all_substructure_item_uids_file
The -dump_substructure argument produces an output similar to the following example,
followed by a list of the item UIDs of all items in the structure. The -find_cycles then reads
and processes the list of items.
# descend_structure: at level 1
item gsWFWitwAAgcRA
occThread gBUFWitwAAgcRA
occThread gBfFWitwAAgcRA
# descend_structure: at level 2
item g4SFWitwAAgcRA
item wESFWitwAAgcRA
occThread ARXFWitZAAgcRA
occThread ABRFWi9dAAgcRA
occThread gJZFWitwAAgcRA
occThread gJdFWitwAAgcRA
apn wYXFWivRAAgcRA
# descend_structure: at level 3
item wIaFWitwAAgcRA
item w4aFWitwAAgcRA

Check and generate missing UGPART-BBOX creation form (NX)

1. Run the create_or_update_bbox_and_tso utility in the following format to find all NX datasets
that do not have the UGPART-BBOX form:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=NXBBOXFORM -mode=query -output_dir=c:\temp\ugpart_bbox.

2-2 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

2. Set the QSEARCH_update_enabled preference to False and then run the

run_tc_publishing_utility.bat utility in the following format to generate the missing forms:
run_tc_publishing_utility.bat -u=%tc_admin_user% -p=%tc_admin_pwd% -g=dba -i="!file!"
-publish_tr=yes -record_pa=yes -log="!file!_refile.txt"

Generate TruShape (TSO) files (NX)

1. In NX, click Assemblies->Site Standards->TruShape tab.

Note
The use of refined search based on TSO files is optional for spatial searches.

2. Set the cell size and cell units, and click Generate Component Shape Representations on
Save.
TSO files are generated upon every save.

Note
The default cell size in NX is 0.0 mm. A cell size of 5.0 mm is recommended for initial
testing. You can define your own cell size that is best for your design practice. Bear in
mind that smaller the cell size, the longer the proximity search takes.

3. Run the run_tc_publishing_utility.bat utility in the following format to generate TSO files for the
existing NX dataset:
run_tc_publishing_utility.bat -u=%tc_admin_user% -p=%tc_admin_pwd% -g=dba -i="!file!"
-publish_tr=yes -record_pa=yes -log="!file!_refile.txt"

Note
The run_tc_publishing_utility.bat utility is supplied with NX and more information is
provided in the Teamcenter Integration for NX documentation in the NX Help.

Enabling product-scoped cacheless search

Prior to running the product-scoped cacheless search, you must enable it by running the
qsearch_process_queue command once for every database in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-enable_product_scoping.
This generates the QSEARCH_filter_queries_by_product_scoping preference in Teamcenter with
a value of 1 (true). It also creates the following tables in the Teamcenter database:
• PQSEARCHINDEXABLEITEM

• PQSEARCHINDEXABLEOCCTHREAD

PLM00301 12.2 Setting Up Cacheless Search 2-3

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

• PQSEARCHINDEXABLEAPN

Note
If you want to disable the product scoped-cacheless search, run the
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-disable_product_scoping -drop_table command once for every database.

Make the product indexable

When you make a product indexable, all the components of the product are placed into a database
table and tracked as updates are made. This improves performance by only searching the required
data and not the complete database.
1. Run the qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-clear_all_indexes and qsearch_process_queue -u={USER_ID} -p={PASSWORD}
-g={GROUP} -clear_queue commands to clear any existing index or queues.

2. Run the qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}

-make_indexable -product={PRODUCT_ITEM_ID} 2> make_indexable.out command to
make the product indexable.
If there are any missing product components, the check_indexable option lists it. You must
index it again for it to be picked up.

3. Run the qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}

-check_indexable -product={PRODUCT_ITEM_ID} 2> check_indexable.out command to
validate that the product is indexed.

4. Run the qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}

-list_indexable -product={PRODUCT_ITEM_ID} 2> list_indexable.out command to list all
the contents of the indexed product.

Note
• Run the qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-make_non_indexable -product={PRODUCT_ITEM_ID} command if you want to
make the product non indexable.

• After you index a product, it is recommended that you regenerate the database statistics
to improve the search performance, as described in the database documentation.

Creating bounding boxes

The Teamcenter bounding boxes are generated by populating the bounding box table in Teamcenter
with the information from the UGPART-BBOX. The create_or_update_bbox_and_tso utility is used
to generate the Teamcenter bounding boxes. This is a task is performed once, when you initially
populate the database.

2-4 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

Creating bounding boxes for NX

The figure below illustrates the process of copying the information from the UGPART-BBOX form to
the Teamcenter PBOUNDINGBOX table.

Before you start creating the bounding boxes, it is recommended that you do the following:
• Set up the environment variable to manage the syslogs, TC_KEEP_SYSTEM_LOG=Y.

• Set up a temporary directory, for example TC_TMP_DIR=C:\temp.

Query dataset UIDs for creating bounding boxes

1. Run the create_or_update_bbox_and_tso utility to query the database and find out how many
bounding boxes need to be processed for the entire database.
Run the utility in this format: create_or_update_bbox_and_tso -u={USER_ID}
-p={PASSWORD} -g={GROUP} -translation_mode=NXBBOXTOBBOX -mode=query
-product=%ITEM_ID% -output_dir=%TC_TMP_DIR%.

Note
This can be done only if the product has been made indexable.

2. Check the output directory for the report log file.

Each report has 50,000 UID entries. Till the query is completed, reports are generated, and the
report names are incremented by 1.

Create bounding boxes

1. Set the QSEARCH_update_enabled preference to False.

2. Run the create_or_update_bbox_and_tso utility in the following format:

create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=NXBBOXTOBBOX -mode=process -product={PRODUCT_ITEM_ID} .
Using the -product option, the bounding boxes are generated by the specific product.

PLM00301 12.2 Setting Up Cacheless Search 2-5

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

Delete bounding boxes

If you delete bounding boxes, the operation removes the bounding boxes, multi-bounding boxes, and
the relations that link them together.
Depending on your requirement, you may run any of these commands:
• Run the create_or_update_bbox_and_tso command in the following format to delete all the
bounding boxes in the database:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-delete_all_bboxes .

• Run the create_or_update_bbox_and_tso command in the following format to delete the

bounding boxes for a specific dataset:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-mode=delete -dataset=hBH1IcSSoQUWxA .

• Run the create_or_update_bbox_and_tso command in the following format to delete bounding

boxes for a specific product:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-mode=delete -product={PRODUCT_ITEM_ID} .

Creating bounding boxes and TSO files for JT

To create bounding boxes and TSO files, you must first load the JT file, extract the bounding box
information, add that to the Teamcenter bounding box table, and then create the TSO as a named
reference to the JT dataset.
Do the following:
• Set the TC_KEEP_SYSTEM_LOG environment variable to Y to manage the syslog.

• Create a temporary directory similar to TC_TMP_DIR=C:\temp\.

Query dataset UIDs for creating bounding boxes

1. Run the create_or_update_bbox_and_tso utility in the following format to query the database
and find out how many bounding boxes need to be processed for the entire database:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode= JTTOTSO -mode=query -product=%ITEM_ID%
-output_dir=%TC_TMP_DIR%.

Note
This can be done only if the product specified in the command line in the -product
option has been made indexable.

2. Check the output directory for the report log file.

2-6 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

Create bounding boxes

1. Set the QSEARCH_update_enabled preference to False.

2. Run the create_or_update_bbox_and_tso utility in this format:

create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=NXBBOXTOBBOX -mode=process -product={PRODUCT_ITEM_ID}.
Using the -product option, the bounding boxes are generated by the specific product.

Delete bounding boxes

If you delete bounding boxes, the operation removes the bounding boxes, multi-bounding boxes, and
the relations that link them together.
Depending on your requirement, you may run any of these commands:
• Run the create_or_update_bbox_and_tso command in the following format to delete all the
bounding boxes in the database:
create_or_update_bbox_and_tso -u=infodba -p=infodba -g=dba -delete_all_bboxes.

• Run the create_or_update_bbox_and_tso command in the following format to delete the

bounding boxes for a specific dataset:
create_or_update_bbox_and_tso -u=infodba -p=infodba -g=dba -mode=delete
-dataset=hBH1IcSSoQUWxA .

• Run the create_or_update_bbox_and_tso command in the following format to delete bounding

boxes for a specific product:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-mode=delete -product={PRODUCT_ITEM_ID} .

Creating a spatial index

After generating the bounding boxes from NX or JT, you can generate the index using the
qsearch_process_queue utility. This provides the option to create, update, or delete the index boxes.

Populating the queue

After you create the bounding boxes you populate the queue. The queue entries are Very Large
Arrays (VLA) and contain the item, the occurrence thread, or the appearance path node that is related
to the particular CAD geometry that the bounding box represents. Each VLA has a maximum of 5,000
object items (depends on the QSEARCH_chunk_size preference value: 5,000 is the default). If the
object items exceeds this value, another entry is created in the update table.
Before you process any updates, ensure the following:
• TC_KEEP_SYSTEM_LOG environment variable is Yes.

• QSEARCH_update_enabled preference is True.

• QSEARCH_foreground_processing_halted preference is True.

PLM00301 12.2 Setting Up Cacheless Search 2-7

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

• QSEARCH_force_simplify_to_one_box preference is True.

• Set a temporary directory, for example TC_TMP_DIR=C:\temp\create_bbox_uids.

Populate the queue by running the qsearch_process_queue command in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-force_queue_product_substructure_update -product={PRODUCT_ITEM_ID}.
This updates the queue with all the necessary objects related to the product.

Note
Populating the queue generates entries into the PQSEARCHINDEXUPDATEENTRY
table for each bounding box.

Listing the queue

After you populate a queue, list the queue by running the qsearch_process_queue utility in the
following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -list_queue –verbose
2> %TC_TMP_DIR%\list_queue.txt
The output is similar to this:
Queue contains 7 objects in 1 entry entry = 00007445 = gQM14MFEoQUWxA object = 00007446
= BpF1IcSSoQUWxA, type = 1 (itemType), date = 18-Mar-2011 10:33:40, box-delta = 0,
state = 0 (unprocessed), session = 00007444 = AII14MFEoQUWxA, is-run-time = FALSE Item
"15259388-BELT ASM-R/SEAT CTR BKL" object = 00007447 = BpH1IcSSoQUWxA, type = 1 (itemType),
date = 18-Mar-2011 10:33:40, box-delta = 0, state = 0 (unprocessed), session = 00007444 =
AII14MFEoQUWxA, is-run-time = FALSE Item "11561331-BOLT ASM"

Processing the queue

When the queue is populated, the next step is to generate the index. This populates the
PQSEARCHINDEX table with the spatial index information for items, occurrence threads, and
appearance path nodes that are impacted by the bounding boxes. The spatial index is the rolled
up bounding box of all children and all their revisions as described in Understanding spatial search
indexes.
In the example below if we start from the bottom and take the bounding box of the dataset
(piece part), its transform then rolls up to its subassembly. We then continue rolling up the
assembly structure and populating the information for each item. The following example uses the
QSEARCH_force_simplify_to_one_box method.

2-8 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

The entries of the PBOUNDINGBOX table are as follows:

The entries of the PQSEARCHINDEX table are as follows:

After processing the queue, it is recommended that you generate the database again to improve
search performance.

Process all updates

• Run the qsearch_process_queue utility in the following format to process all the entries in
the queue:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue.

Process by UID
• Run the qsearch_process_queue utility in the following format to process all multiple UIDs
using a comma-separated list:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue
-uid=0$0F56myQyYmUC.

Process by item id list

• Run the qsearch_process_queue utility in the following format to process multiple Item ID’s
using a comma-separated list:

PLM00301 12.2 Setting Up Cacheless Search 2-9

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue

-item_id=@list_of_items.txt.

Deleting an index
If you want to delete an index in your database, use the qsearch_process_queue command in
the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -clear_product_indexes
-product=%ITEM_ID%.

Setting up live background updates

The live background updates can be done by using Dispatcher, cron jobs, or using a Windows service.
If you use Dispatcher, you must set up to manage the daily updates of the spatial index. Set its timeout
period to a duration long enough for the application to process QSearchProcessQueue requests.

Setting up Dispatcher to manage updates (NX)

To set up Dispatcher for managing updates, ensure that you have:
• Installed the QSearchProcessQueue Dispatcher translator, using TEM as shown in the following
figure.

• Set the QSEARCH_update_enabled preference to True.

• Set the QS_BBOX_GENERATION_FROM_NX_ENABLED preference to True.

The following figure illustrates the process of how an NX dataset is created, saved, and the spatial
index, updated with Dispatcher setup.
1. When a geometry is saved, triggers are automatically set for creating bounding boxes and
TSO within the NX application. It also triggers a structure update to roll up the bounding box
information to its parent structure.

2. BBOX is created or updated.

3. TSO is created or updated.

4. The creation of spatial indexes is triggered in the form of a Dispatcher request.

2-10 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

Note
If you would like to configure the QsearchProcessQueue Dispatcher translator as a
recurring request, set the following preferences as required:
• QSEARCH_queue_recurring_background_updates

• QSEARCH_recurring_background_update_survival_time

• QSEARCH_recurring_background_update_interval

If you want to reduce the objects getting added to the queue and also reduce the number of Dispatcher
requests, specify the item types you want to avoid in the QSEARCH_types_to_avoid_processing
preference.

Setting up Dispatcher to manage updates (JT)

To set up Dispatcher to manage updates, ensure that you have:

• Installed the following Dispatcher translators, using TEM, as shown in the following figure:

o JtToBboxAndTso

o QSearchProcessQueue

PLM00301 12.2 Setting Up Cacheless Search 2-11

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

Note
It is recommended that you use only one translator for the spatial search indexer.

• Set the QSEARCH_update_enabled preference to True if you want to generate TSO files.

The following figure illustrates the process of how a JT dataset is created, saved, and the spatial
index updated with Dispatcher setup.
1. When a geometry is saved, triggers are automatically set for creation of a JT file.

2. The JT file is created and it requests the Dispatcher to generate the bounding boxes and TSO
files.

3. The JT is loaded and the bounding box information is populated in the bounding box table. The
tables in the database are populated with the bounding box information.

4. TSO files are generated and attached as a named reference.

5. A Dispatcher request is triggered, and a spatial index is created.

Note
If you would like to configure the QsearchProcessQueue Dispatcher translator as a
recurring request, set the following preferences as required:
• QSEARCH_queue_recurring_background_updates

• QSEARCH_recurring_background_update_survival_time

• QSEARCH_recurring_background_update_interval

If you want to reduce the objects getting added to the queue and also reduce the
number of dispatcher requests, specify the item types you want to avoid in the
QSEARCH_types_to_avoid_processing preference.

2-12 Setting Up Cacheless Search PLM00301 12.2

 Working with Product Scoped Cacheless Search

Setting up cron jobs

If you want to generate the spatial indexes using a cron job script, set the
QSEARCH_dispatcher_not_available preference to True.

Configuring cacheless search in a Multi-Site Collaboration

environment
If you work in a Multi-Site Collaboration environment, you can make cacheless spatial and attribute
searches of products or programs that are owned or replicated at remote sites. Where appropriate,
local and remote objects are incorporated in the search results.
If you enable this feature:
• Teamcenter checks access permissions to the remote BOM lines. If you do not have the
necessary permissions, you cannot import the affected remote lines. If appropriate, you can
review the search results and import only the lines to which you have access permissions.

• Identical revision and variant rules must be defined at all sites or searches may return inconsistent
results. Also, the content of these revision rules and variant rules must always be consistent
across all the sites.

• The content of saved queries must be consistent across all sites or searches may return
inconsistent results.

• You can only initiate the search in Design Context, not in Structure Manager or Multi-Structure
Manager.

To configure this feature, you must set the following preferences:

• QS_remote_master_site
This preference lists the remote sites and the product or program that can be searched at each of
those sites. It must be defined in a format similar to the following example:
DEFAULT:site1
Product2:master_site2
Product3:master_site3
The default site is always searched.

• QS_remote_master_site_override
This preference allows the user to define a preferred remote site to make cacheless searches. It
must be defined in a format similar to the following example:
Product3:master_site3
Whenever you add a new product or assembly to remotely search in Design Context, the
administrator must update this preference with the new product and its owning master site at all
sites in the Multi-Site Collaboration network.

You do not need to manually configure IDSM to use remote cacheless searches. When you install
or upgrade the system using Teamcenter Environment Manager (TEM), it installs the callbacks

PLM00301 12.2 Setting Up Cacheless Search 2-13

 Chapter
Chapter 2: 2: Working
Working with Product
with Product Scoped
Scoped Cacheless
Cacheless Search
Search

necessary for remote cacheless search. Each site participating in remote cacheless search must
install or upgrade the database using the same version of TEM.

2-14 Setting Up Cacheless Search PLM00301 12.2

Chapter 3: Working with full database cacheless search

Preparing your data

Before you use cacheless search on an existing database, you must run a few checks and clean the
data.
To begin with, you must upgrade all existing legacy transforms, since the cacheless search
functionality uses only the new PLM XML transform. Next, you must run the qsearch_process_queue
utility to find and clean up all cycles (circular references). Also, check that all NX datasets have
the UGPART-BBOX form.

Note
Ensure that you have the typical Teamcenter environment variables set for all command
line utilities.

Upgrade legacy transforms

1. Find all legacy transforms by running the -list_legacy_transforms option in the
qsearch_process_queue utility in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-list_legacy_transforms -outfile=c:\temp\legacytransforms.txt.

2. Run the nxmgr_update_transforms utility to upgrade the transforms in the following format:
nxmgr_upgrade_transforms -u={USER_ID} -p={PASSWORD} -g={GROUP} -bypass=yes
-i=c:\temp\legacytransforms.txt -upgrade_release=yes.

3. Run the -list_legacy_transforms option again to ensure that all transforms are upgraded.

Find and clean up cycles

1. Run the find_cycles option in the qsearch_process_queue utility to find all the cycles.
To find cycles in the entire database, run qsearch_process_queue -u={USER_ID}
-p={PASSWORD} -g={GROUP} -find_cycles 2> c:\temp\list_of_cycles.txt.
Processing the entire database might take a long time. Instead, you can find cycles by item
ID, list of item IDs, or item ID UIDs.
To find a cycle by item id, run qsearch_process_queue -u={USER_ID} -p={PASSWORD}
-g={GROUP} -find_cycles–item_id=AKT75443 2>c:\temp\AKT754432_cycles.txt.

2. Clean up the cycles as per your organization’s plan, considering that cleaning up cycles affects
the content of the data.

PLM00301 12.2 Setting Up Cacheless Search 3-1

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

Tip
(Product-scoped cacheless searches only)
The -check_structure_indexes argument of the qsearch_process_queue utility flags
cycles in a structure with an index-box-check found cycle report. This argument starts
the utility at one or more specified items and works down the entire substructure. It finds
apparently incorrect indexes and, at the same time, also finds cycles. Conversely, the
find_cycles argument works up the substructure from the specified item; if you specify
only top-level items, it returns no results. Also, the find_cycles argument takes all items
by default.
In general, arguments with structure in the name interpret given item IDs as top-level
items, and perform the relevant action on the entire structure below each item. Options
without structure in the name simply perform the relevant action on each item in isolation.
Therefore, the -check_indexes argument simply checks each item's index but does not
find any cycles; the -check_structure_indexes argument also identifies any cycles.
To check for cycles under a top-level item, enter a command similar to the following
example:
qsearch_process_queue -dump_substructure -item_id=top-level-item-id(s) 2>&1
) | grep " item " | sed -e "s/ *item *//" | sort -u >all_substructure_item_uids_file
qsearch_process_queue -find_cycles -uid=@all_substructure_item_uids_file
The -dump_substructure argument produces an output similar to the following example,
followed by a list of the item UIDs of all items in the structure. The -find_cycles then reads
and processes the list of items.
# descend_structure: at level 1
item gsWFWitwAAgcRA
occThread gBUFWitwAAgcRA
occThread gBfFWitwAAgcRA
# descend_structure: at level 2
item g4SFWitwAAgcRA
item wESFWitwAAgcRA
occThread ARXFWitZAAgcRA
occThread ABRFWi9dAAgcRA
occThread gJZFWitwAAgcRA
occThread gJdFWitwAAgcRA
apn wYXFWivRAAgcRA
# descend_structure: at level 3
item wIaFWitwAAgcRA
item w4aFWitwAAgcRA

Check and generate missing UGPART-BBOX creation form (NX)

1. Run the create_or_update_bbox_and_tso utility in the following format to find all NX datasets
that do not have the UGPART-BBOX form:
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=NXBBOXFORM -mode=query -output_dir=c:\temp\ugpart_bbox.

3-2 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

2. Set the QSEARCH_update_enabled preference to False and then run the

run_tc_publishing_utility.bat utility in the following format to generate the missing forms:
run_tc_publishing_utility.bat -u=%tc_admin_user% -p=%tc_admin_pwd% -g=dba -i="!file!"
-publish_tr=yes -record_pa=yes -log="!file!_refile.txt"

Generate TruShape (TSO) files (NX)

1. In NX, click Assemblies->Site Standards->TruShape tab.

Note
The use of refined search based on TSO files is optional for spatial searches.

2. Set the cell size and cell units, and click Generate Component Shape Representations on
Save.
TSO files are generated upon every save.

Note
The default cell size in NX is 0.0 mm. A cell size of 5.0 mm is recommended for initial
testing. You can define your own cell size that is best for your design practice. Bear in
mind that smaller the cell size, the longer the proximity search takes.

3. Run the run_tc_publishing_utility.bat utility in the following format to generate TSO files for the
existing NX dataset:
run_tc_publishing_utility.bat -u=%tc_admin_user% -p=%tc_admin_pwd% -g=dba -i="!file!"
-publish_tr=yes -record_pa=yes -log="!file!_refile.txt"

Note
The run_tc_publishing_utility.bat utility is supplied with NX and more information is
provided in the Teamcenter Integration for NX documentation in the NX Help.

Creating bounding boxes

The Teamcenter bounding boxes are generated by populating the bounding box table in Teamcenter
with the information in the UGPART-BBOX. The create_or_update_bbox_and_tso utility is used to
generate the Teamcenter bounding boxes. This task is performed once, when you initially populate
the database.

Creating bounding boxes for NX

The figure below illustrates the process of copying the information from the UGPART-BBOX form
to the Teamcenter bounding box table PBOUNDINGBOX table.

PLM00301 12.2 Setting Up Cacheless Search 3-3

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

Before you start creating the bounding boxes, it is recommended that you do the following:
• Set up the environment variable to manage the syslogs, TC_KEEP_SYSTEM_LOG=Y.

• Set up a temporary directory, for example, TC_TMP_DIR=C:\temp.

Query dataset UIDs for creating bounding boxes

1. Run the create_or_update_bbox_and_tso utility in the following format to query the database
and find out how many bounding boxes need to be processed for the entire database.
create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=NXBBOXTOBBOX -mode=query -output_dir=%TC_TMP_DIR%.
This generates the createOrUpdateBBoxAndtso_report{%ID%_0.log file.

2. Check the output directory for the report log file.

Each report has 50,000 UID entries. Till the query is completed, reports are generated and the
report names are incremented by 1.
An ideal report reads like this:
Following NX Datasets need to update Bounding Boxes:
x5D1IcSSoQUWxA
RNI1IcSSoQUWxA
htB1IcSSoQUWxA
xpH1IcSSoQUWxA
h5M1IcSSoQUWxA
RiN1IcSSoQUWxA
hBH1IcSSoQUWxA

Create bounding boxes

1. Set the QSEARCH_update_enabled preference to False.

2. Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD}

-g={GROUP} -translation_mode=NXBBOXTOBBOX -mode=process
-object_list=%TC_TMP_DIR%\createOrUpdateBBoxAndtso_report{%ID%}_0.log command.

3-4 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

Note
You can run the command only if the product has been made indexable when using
product scoped search.

3. Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD}

-g={GROUP} -translation_mode=NXBBOXTOBBOX -mode=query –scope=ALL
-output_dir=%TC_TMP_DIR% command to query all datasets of all indexable products.

4. Validate the bounding boxes by sending an assembly to Structure Manager, adding the Bounding
Boxes column, and then expanding the assembly to a level that has geometry.
The column will be populated with the bounding box information.

Delete bounding boxes

If you delete bounding boxes, the operation removes the bounding boxes, multi-bounding boxes, and
the relations that link them together.
Depending on your requirement, you may run any of these commands:
• Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}
-delete_all_bboxes command to delete all the bounding boxes in the database.

• Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}

-mode=delete -dataset=hBH1IcSSoQUWxA command to delete the bounding boxes for a
specific dataset.

• Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}

-mode=delete -product={PRODUCT_ITEM_ID} command to delete bounding boxes for a
specific product.

Creating bounding boxes and TSO files for JT

To create bounding boxes and TSO files, you must first load the JT file, extract the bounding box
information, add that to the Teamcenter bounding box table, and then create the TSO as a named
reference to the JT dataset. To start with, do the following:
• Set the TC_KEEP_SYSTEM_LOG environment variable to Y to manage the syslog.

• Create a temporary directory, for example, TC_TMP_DIR=C:\temp.

Query dataset UIDs for creating bounding boxes and TSO

Depending on your requirement, run one or all the utilities to query the database to find out how
many UIDs will need to be processed.
1. Run the create_or_update_bbox_and_tso –u={USER_ID} -p={PASSWORD} -g={GROUP}
-translation_mode=JTTOBBOX+JTTOTSO -mode=query -output_dir=%TC_TMP_DIR%
utility to query the database to find out how many UIDs will need to be processed for the complete
database (both bounding boxes and TSO files).

PLM00301 12.2 Setting Up Cacheless Search 3-5

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

2. Run the create_or_update_bbox_and_tso –u={USER_ID} -p={PASSWORD} -g={GROUP}

-translation_mode=JTTOBBOX -mode=query -output_dir=%TC_TMP_DIR% utility to query
the database to find out how many UIDs will need to be processed for the bounding boxes only.

3. Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP

-translation_mode=JTTOTSO -mode=query -output_dir=%TC_TMP_DIR% utility to query the
database to find out how many UIDs will need to be processed for the TSO files only.

Create bounding boxes and TSO files

1. Set the QSEARCH_update_enabled preference to False.

2. Run the create_or_update_bbox_and_tso -u=infodba -p=infodba

-g=dba -translation_mode=JTTOBBOX+JTTOTSO -mode=process
-object_list=%TC_TMP_DIR%\createOrUpdateBBoxAndtso_report{%ID%}_.lo utility to
create both entities, the bounding boxes and the TSO files.

3. Run the create_or_update_bbox_and_tso -u=infodba -p=infodba

-g=dba -translation_mode=JTTOBBOX -mode=process
-object_list=%TC_TMP_DIR%\createOrUpdateBBoxAndtso_report{%ID%}_.log utility
to create only the bounding box.

4. Run the create_or_update_bbox_and_tso -u=infodba -p=infodba

-g=dba -translation_mode=JTTOTSO -mode=process
-object_list=%TC_TMP_DIR%\createOrUpdateBBoxAndtso_report{%ID%}_.log utility
to create only the TSO files.

5. Validate the bounding boxes by sending an assembly to Structure Manager, adding the Bounding
Boxes column, and then expanding the assembly to a level that has geometry.
The column will be populated with the bounding box information.

Delete bounding boxes

If you delete bounding boxes, the operation removes the bounding boxes, multi-bounding boxes, and
the relations that link them together.
Depending on your requirement, you may run any of these commands:
• Run the create_or_update_bbox_and_tso -u=infodba -p=infodba -g=dba -delete_all_bboxes
command to delete all the bounding boxes in the database.

• Run the create_or_update_bbox_and_tso -u=infodba -p=infodba -g=dba -mode=delete

-dataset=hBH1IcSSoQUWxA command to delete the bounding boxes for a specific dataset .

• Run the create_or_update_bbox_and_tso -u={USER_ID} -p={PASSWORD} -g={GROUP}

-mode=delete -product={PRODUCT_ITEM_ID} command to delete bounding boxes for a
specific product.

3-6 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

Creating a spatial index

After generating the bounding boxes from NX or JT, you can generate the index using the
qsearch_process_queue utility. This provides the option to create, update, or delete the index boxes.

Populating the queue

After you create the bounding boxes you populate the queue. The queue entries are Very Large
Arrays (VLA) and contain the item, the occurrence thread, or the meappearance path node that is
related to the particular CAD geometry that the bounding box represents. Each VLA has a maximum
of 5,000 object items (depends on the QSEARCH_chunk_size preference value: 5,000 is the
default). If the object items exceeds this value, another entry is created in the update table.
Before you process any updates, ensure the following:
• TC_KEEP_SYSTEM_LOG environment variable is Yes.

• QSEARCH_update_enabled preference is True.

• QSEARCH_foreground_processing_halted preference is True.

• QSEARCH_force_simplify_to_one_box preference is True.

• Set a temporary directory, for example, TC_TMP_DIR=C:\temp\create_bbox_uids.

Note
Populating the queue generates entries into the PQSEARCHINDEXUPDATEENTRY
table for each bounding box.

Processing all updates

This processes all updates of all bounding box UIDs, including the ones that already have index
boxes). Run the qsearch_process_queue command once in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-force_queue_all_possible_updates.

Processing necessary updates

When you want to process only the necessary updates (where index boxes are not generated
yet) or only add the bounding box UIDs of those that have not been processed, run the
qsearch_process_queue command in the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP}
-force_queue_all_necessary_updates.

Listing the queue

After you populate a queue, list the queue by running the qsearch_process_queue utility in the
following format:

PLM00301 12.2 Setting Up Cacheless Search 3-7

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -list_queue –verbose

2> %TC_TMP_DIR%\list_queue.txt
The output is similar to this:
Queue contains 7 objects in 1 entry entry = 00007445 = gQM14MFEoQUWxA object = 00007446
= BpF1IcSSoQUWxA, type = 1 (itemType), date = 18-Mar-2011 10:33:40, box-delta = 0,
state = 0 (unprocessed), session = 00007444 = AII14MFEoQUWxA, is-run-time = FALSE Item
"15259388-BELT ASM-R/SEAT CTR BKL" object = 00007447 = BpH1IcSSoQUWxA, type = 1 (itemType),
date = 18-Mar-2011 10:33:40, box-delta = 0, state = 0 (unprocessed), session = 00007444 =
AII14MFEoQUWxA, is-run-time = FALSE Item "11561331-BOLT ASM"

Processing the queue

When the queue is populated, the next step is to generate the index. This populates the
PQSEARCHINDEX table with the spatial index information for items, occurrence threads, and
appearance path nodes that are impacted by the bounding boxes. The spatial index is the rolled
up bounding box of all children and all their revisions as described in Understanding spatial search
indexes.
In the example below if we start from the bottom and take the bounding box of the dataset
(piece part), its transform then rolls up to its subassembly. We then continue rolling up the
assembly structure and populating the information for each item. The following example uses the
QSEARCH_force_simplify_to_one_box method.

The entries of the PBOUNDINGBOX table are as follows:

The entries of the PQSEARCHINDEX table are as follows:

3-8 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

After processing the queue, it is recommended that you generate the database again to improve
search performance.

Process all updates

• Run the qsearch_process_queue utility in the following format to process all the entries in
the queue:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue.

Process by UID
• Run the qsearch_process_queue utility in the following format to process all multiple UIDs
using a comma-separated list:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue
-uid=0$0F56myQyYmUC.

Process by item id list

• Run the qsearch_process_queue utility in the following format to process multiple Item ID’s
using a comma-separated list:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -process_queue
-item_id=@list_of_items.txt.

Deleting all indexes

If you want to delete all indexes from your database, use the qsearch_process_queue command in
the following format:
qsearch_process_queue -u={USER_ID} -p={PASSWORD} -g={GROUP} -clear_all_indexes.

Setting up live background updates

The live background updates can be done by using Dispatcher, cron jobs, or using a Windows service.
If you use Dispatcher, you must set up to manage the daily updates of the spatial index. Set its timeout
period to a duration long enough for the application to process QSearchProcessQueue requests.

Setting up Dispatcher to manage updates (NX)

To set up Dispatcher for managing updates, ensure that you have:
• Installed the QSearchProcessQueue Dispatcher translator, using TEM as shown in the following
figure.

PLM00301 12.2 Setting Up Cacheless Search 3-9

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

• Set the QSEARCH_update_enabled preference to True.

• Set the QS_BBOX_GENERATION_FROM_NX_ENABLED preference to True.

The following figure illustrates the process of how an NX dataset is created, saved, and the spatial
index, updated with Dispatcher setup.
1. When a geometry is saved, triggers are automatically set for creating bounding boxes and
TSO within the NX application. It also triggers a structure update to roll up the bounding box
information to its parent structure.

2. BBOX is created or updated.

3. TSO is created or updated.

4. The creation of spatial indexes is triggered in the form of a Dispatcher request.

Note
If you would like to configure the QsearchProcessQueue Dispatcher translator as a
recurring request, set the following preferences as required:
• QSEARCH_queue_recurring_background_updates

• QSEARCH_recurring_background_update_survival_time

• QSEARCH_recurring_background_update_interval

3-10 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

If you want to reduce the objects getting added to the queue and also reduce the number of Dispatcher
requests, specify the item types you want to avoid in the QSEARCH_types_to_avoid_processing
preference.

Setting up Dispatcher to manage updates (JT)

To set up Dispatcher to manage updates, ensure that you have:
• Installed the following Dispatcher translators, using TEM, as shown in the following figure:
o JtToBboxAndTso

o QSearchProcessQueue

Note
It is recommended that you use only one translator for the spatial search indexer.

• Set the QSEARCH_update_enabled preference to True if you want to generate TSO files.

The following figure illustrates the process of how a JT dataset is created, saved, and the spatial
index updated with Dispatcher setup.
1. When a geometry is saved, triggers are automatically set for creation of a JT file.

2. The JT file is created and it requests the Dispatcher to generate the bounding boxes and TSO
files.

3. The JT is loaded and the bounding box information is populated in the bounding box table. The
tables in the database are populated with the bounding box information.

4. TSO files are generated and attached as a named reference.

5. A Dispatcher request is triggered, and a spatial index is created.

PLM00301 12.2 Setting Up Cacheless Search 3-11

 Chapter
Chapter 3: 3: Working
Working withdatabase
with full full database cacheless
cacheless search
search

Note
If you would like to configure the QsearchProcessQueue Dispatcher translator as a
recurring request, set the following preferences as required:
• QSEARCH_queue_recurring_background_updates

• QSEARCH_recurring_background_update_survival_time

• QSEARCH_recurring_background_update_interval

If you want to reduce the objects getting added to the queue and also reduce the
number of dispatcher requests, specify the item types you want to avoid in the
QSEARCH_types_to_avoid_processing preference.

Setting up cron jobs

If you want to generate the spatial indexes using a cron job script, set the
QSEARCH_dispatcher_not_available preference to True.

Configuring cacheless search in a Multi-Site Collaboration

environment
If you work in a Multi-Site Collaboration environment, you can make cacheless spatial and attribute
searches of products or programs that are owned or replicated at remote sites. Where appropriate,
local and remote objects are incorporated in the search results.
If you enable this feature:
• Teamcenter checks access permissions to the remote BOM lines. If you do not have the
necessary permissions, you cannot import the affected remote lines. If appropriate, you can
review the search results and import only the lines to which you have access permissions.

3-12 Setting Up Cacheless Search PLM00301 12.2

 Working with full database cacheless search

• Identical revision and variant rules must be defined at all sites or searches may return inconsistent
results. Also, the content of these revision rules and variant rules must always be consistent
across all the sites.

• The content of saved queries must be consistent across all sites or searches may return
inconsistent results.

• You can only initiate the search in Design Context, not in Structure Manager or Multi-Structure
Manager.

To configure this feature, you must set the following preferences:

• QS_remote_master_site
This preference lists the remote sites and the product or program that can be searched at each of
those sites. It must be defined in a format similar to the following example:
DEFAULT:site1
Product2:master_site2
Product3:master_site3
The default site is always searched.

• QS_remote_master_site_override
This preference allows the user to define a preferred remote site to make cacheless searches. It
must be defined in a format similar to the following example:
Product3:master_site3
Whenever you add a new product or assembly to remotely search in Design Context, the
administrator must update this preference with the new product and its owning master site at all
sites in the Multi-Site Collaboration network.

You do not need to manually configure IDSM to use remote cacheless searches. When you install
or upgrade the system using Teamcenter Environment Manager (TEM), it installs the callbacks
necessary for remote cacheless search. Each site participating in remote cacheless search must
install or upgrade the database using the same version of TEM.

PLM00301 12.2 Setting Up Cacheless Search 3-13

Appendix A: Oracle Indexes

Ensure that the following Oracle indexes are created for the best performance of cacheless search.
• PVARIANTEXPRESSION(POPERATOR,PFORMULA_STATE)

• PSOccurrenceNotes PPNOTE_TEXTS(UPPER(PVALU_0))

• PPSOCCURRENCE(RCHILD_ITEMU,ROCC_THREADU)

• PPSOCCURRENCE(RALTERNATE_ETC_REFU,ROCC_THREADU)

• PITEMREVISION(PUID,RITEMS_TAGU)

• PPOM_APPLICATION_OBJECT(ROWNING_GROUPU,PUID)

• PPEXPRESSIONS(PUID)

• PVARIANTEXPRESSION(POPERATOR, PUID)

PLM00301 12.2 Setting Up Cacheless Search A-1

Appendix B: Troubleshooting

Some common issues and how to troubleshoot them are included here for your reference.

How do I validate a spatial index?

Validate the index by running the qsearch_process_queue utility with the check_structure_index
option.

Why is qsearch_process_queue taking so long?

A recurring request to the qsearch_process_queue, or any request to qsearch_process_queue
may take longer than the dispatcher timeout period as it might have a lot of data to process.
The dispatcher assumes that the old request was finished, and launches a new
qsearch_process_queue request for any new incoming request. This causes more than one
qsearch_process_queue request to be processed at the same time, leading to locking contention,
and subsequent problems with spatial index updates.
To avoid this, it is advised to set the timeout period for dispatcher to be high value.

How do I improve search engine performance?

If you find that the search engine performance is not the best, consider the following:
• If you have a small product structure (one thousand or less occurrence threads), you may
improve performance by switching to a top down search.

• The lower the variation in geometry between different revisions of the same item, the better
the search engine performs.

• The greater the correlation between the product structure and the geometry breakdown, the
better the search engine performs.

• The accuracy of the search depends on the cell size you configure in NX or the voxel size in JT. A
small cell or voxel size provides more accurate search results, but smaller cell sizes result in
larger JT or TruShape files and response time takes more time.

• Avoid accumulating large number of entries in the indexing queue. The faster the updates are
processes, the sooner live updates are available.

• During the deployment process, monitor the size of the queue by periodically running the
qsearch_process_queue utility with the -list_queue option. If the queue search is increasing
steadily, you might have to reconsider your deployment strategy.

PLM00301 12.2 Setting Up Cacheless Search B-1

 Appendix
Appendix B: B: Troubleshooting
Troubleshooting

Why is my search slow?

If you have a consistently slow search, set the following environment variables and execute a search.
• TC_JOURNAL=SUMMARY

• TC_KEEP_SYSTEM_LOG=1

You can then examine the tcserver system logs, which show a system status summary, a list of
low-level call stacks, and the time consumed by each call. You can identify operations that consume
considerable time and report them to get help from us. If the search is slow because of database
issues, then get help from the database administrator.

B-2 Setting Up Cacheless Search PLM00301 12.2

Siemens Industry Software

Headquarters
Europe
Granite Park One
Stephenson House
5800 Granite Parkway
Sir William Siemens Square
Suite 600
Frimley, Camberley
Plano, TX 75024
Surrey, GU16 8QD
USA
+44 (0) 1276 413200
+1 972 987 3000

Asia-Pacific
Americas
Suites 4301-4302, 43/F
Granite Park One
AIA Kowloon Tower, Landmark East
5800 Granite Parkway
100 How Ming Street
Suite 600
Kwun Tong, Kowloon
Plano, TX 75024
Hong Kong
USA
+852 2230 3308
+1 314 264 8499

About Siemens PLM Software

© 2019 Siemens Product Lifecycle Management

Siemens PLM Software, a business unit of the Siemens
Software Inc. Siemens and the Siemens logo are
Industry Automation Division, is a leading global provider
registered trademarks of Siemens AG. D-Cubed,
of product lifecycle management (PLM) software and
Femap, Geolus, GO PLM, I-deas, Insight, JT, NX,
services with 7 million licensed seats and 71,000 customers
Parasolid, Solid Edge, Teamcenter, Tecnomatix and
worldwide. Headquartered in Plano, Texas, Siemens
Velocity Series are trademarks or registered trademarks
PLM Software works collaboratively with companies
of Siemens Product Lifecycle Management Software
to deliver open solutions that help them turn more
Inc. or its subsidiaries in the United States and in other
ideas into successful products. For more information
countries. All other trademarks, registered trademarks
on Siemens PLM Software products and services, visit
or service marks belong to their respective holders.
www.siemens.com/plm.