Salesforce

Sharing Trados Studio Translation Memories and Termbases On A Network

« Go Back

Information

 
TitleSharing Trados Studio Translation Memories and Termbases On A Network
URL Name000005098
SummaryWhat issues may occur when saving file-based translation memories and termbases on a network drive/in a synchronised folder
Scope/EnvironmentTrados Studio 2022
Trados Studio 2021
Trados Studio 2019
Trados Studio 2017
Multiterm 2021
Multiterm 2019
Multiterm 2017
Trados Team
Translation Memories
Termbases
Terminology
Question
How do I share file-based Trados Studio Translation Memories (*.sdltm) and Termbases (*.sdltb) across a network so that other users can access them at the same time?
Answer

Background

The Translation Memory files in the question are individual files with the TM file extension *.sdltm and in this article, we will refer to them as ‘file-based TMs’.
The Terminology files (Termbase) are also individual files with the file extension *.sdltb.

If a user wants to share resources such as Translation Memories (TMs) and/or Termbase terminology at the same time, with a colleague or a small team without deploying a server solution this article provides guidelines on how to do so using an alternative networked environment.  However, this is not a recommended workflow.

It goes on to explain the limitations and disadvantages of sharing file-based TMs or  Termbase and follows on with a list of a number of best practices for TM/ Termbase-resource sharing.
 

Answer

The answer is that the sharing of Trados Studio file-based TMs and Terminology, outside of a server environment is possible with either Studio Freelance or Studio Professional but is not a recommended workflow. Users can provide shared access to file-based TMs and Termbase terminology by using a networked file repository or an external cloud-based file sharing facility.  All that is needed is the setup of a shared folder and an acceptance of an invitation to share, which can be sent out to named users. These individual files can be saved, read, amended, copied, and deleted by any user that is given access to the network.

Newly added translations can be shared instantly as 100% matches are seen by all users of the translation memory or suggestion in the Term Recognition window. That means, if a user adds a new translation unit in the TM or a new term/ entry in the Termbase and another user comes across the same segment, the latter will be able to retrieve it immediately as a 100% match.

Users can access the file at any time but not always simultaneously.  This depends on network integrity and locking issues. To ensure accurate version control, it is recommended that rigid access rules must be adhered to by all users and a process of ‘read/write’ permissions introduced to ensure that the latest, most up to date version of the TM and Termbase is being shared.  Alternatively, a process of having a number of copy TMs or Termbases could be used instead of just one, users could be allowed 'to read' all TMs’ and Termbases but only ‘write to one’, on their own this would help to ensure accurate version control.
 

Limitations of file-based TM / Terminology sharing

1. Performance issues, such as speed, can impact sharing file-based TMs / Termbases, and work could be impacted by network degradation or inconsistent access. Network infrastructure, bandwidth, firewalls, monitoring tools or backup tools can all cause issues outside of the Studio environment.
2. Using external network facilities such as cloud-based sharing facilities, could be subject to security and confidentiality breaches and may impact GDPR ruling.
3. Access to file-based TMs / Termbases over external networks can be open to malpractice, such as copying the entire file for own use when there is no permission-based management of files.
4. Multiple-user access to the same file-based TM / Termbase file can introduce inconsistent versions, and users need to be diligent in TM / Termbase version control.
5. Access to shared file-based TMs / Termbase should ideally be restricted to 2 users, and at a maximum 3 users to avoid creating issues around TM / Termbase maintenance and inconsistent versions.
6. Using one file-based TM / Termbase across several users can cause corruption of the file and also locking problems.
7. Uplift fragment matching technology was introduced with Studio 2017. File-based TMs that have taken advantage of this functionality are now larger and degradation of performance can impact usage of shared file-based TMs, making access much slower.
8. Fuzzy matches cannot be shared instantly in this way, because updating the local fuzzy index for each segment added to the translation memory would affect performance negatively, resulting in unacceptable response times for users. Instead, the local fuzzy index for users is synchronized with new translation units every twenty minutes - from that point, the newly added translation units can also be shared as fuzzy matches.
 

Recommendations for best practices

1. Project managers should refrain from running the Prepare Task Sequence nor any other batch processing activity while other users are simultaneously using the file-based TM / Termbase.
Possible errors that may occur:
a. Translation Provider threw exception...
b. Database file locked...
c. Database is locked
d. An unexpected network error occurred.
2. Translation memory administrators should refrain from performing any TM maintenance work, as typically this would cause a heavy load and introduce performance issues. In addition, any other TM work that adds many transactions in a short time frame should be avoided. All "heavy TM activities" should therefore only be carried out when there are users are not simultaneous using the translation memory.
3. Ensure the network that is used is configured to offer the highest performance ie speed and uses the latest hardware and software versions.
4. Avoid open-source or Linux-based NAS, opt for Windows-based networks.
5. Try to limit shared file-based TMs to 50,000 TUs – this will help with performance and maintenance. 
6. Trados Studio Freelance licenses are desktop licenses and will only run on workgroup-based networks. Studio Freelance will not work on a domain-based network, Studio Professional is the version used on domain-based networks.
7. For more information on network types refer to the KB articles.  If you try to run Studio Freelance on a domain-based network, the following error message will be displayed:  Your current license does not allow you to run Trados Studio when connected to a corporate domain-based network
 

RWS Support

If the user is using external networking facilities or cloud-based sharing on their network or behind their own firewalls, it is the user’s responsibility to ensure that they have the correct security measures in place and that they allocate users with the correct permissions. RWS provides configuration options in its software that enable end users to enable security features but this is the sole responsibility of the end user to ensure they are in place and suitable for their organization. If you are working with file-based TMs / Termbases that are shared across multiple users and have any errors or run into issues, the first step should be to try to reproduce the issue on your own local hardware, away from the network - copy the file-based TM / Terminology onto your local disk. 
 

RWS's recommended route for TM / Termbase terminology Sharing

If you want to share TMs with more than 2 or 3 users or your TM size exceeds 50,000 TUs or a Termbase with more than 50,000 entries, you may like to consider Trados Team / Accelerate / Enterprise solution to share your translation resources.  Trados Team / Accelerate / Enterprise is a SaaS solution.  

The key benefits of using Trados Team / Accelerate / Enterprise are:

1. Sharing and accessing translation resources centrally and in real-time
2. Control and security with role-based permissions giving users unique degrees of authority to work on the same TM / Termbase.
3. Automation of tasks
4. Terminology consistency and TM accuracy across your organization
 

More information about Trados Team / Accelerate / Enterprise can be found here:  https://www.trados.com

Reference
Attachment 1 
Attachment 2 
Attachment 3 
Attachment 4 
Attachment 5 

Powered by