Vânia de Mathia Campanhão 

Perhaps you can share some additional information to help answer this one?

• where is your TM stored?  Locally, on a shared drive, locally with a location synched with OneDrive etc.
• how large is your TM?
• what spec. is your computer?

Hi Paul,

Thanks for your answer to my question.

My TM is stored locally (on my 1TB SSD). It has about 300,000 TUs and my computer is a Dell, core i7, 7th. gen, 

I've been thinking if there is some kind of limit for a TM (?), and in this case if I should split it to improve performance. Please let me have your thoughts. Thank you again.

Vânia

Vânia de Mathia Campanhão 

That is a very good question.  We do have a KB on this topic and the recommendation is 100,000 TUs: https://gateway.rws.com/csm?id=kb_article_view&sysparm_article=KB0032215

However, I don't think 300,000 is too large and I'm well aware of many translators working regularly with far larger file based TMs than this.  Having said this the recent versions of Trados Studio do increase the physical size of a TM considerably because of the fragment pairs created during the alignment process when you "upgrade" a TM.  So depending on the type of content you have, the number of tags, the length of the segments etc. the actual size and response time is likely to vary.  So with this in mind it's definitely worth testing out whether splitting it into three separate TMs does make a significant difference for you.

You could also play with the other recommendations and test working with your TM as a cloud-based resource to see if this improves the responsiveness.  You have nothing to lose working with copies and a bit of experimenting may help.

Might also be worth seeing if people like Walter Blaser , Jerzy Czopik , Alison Field can offer any advice as they have been around a while working with Studio and have probably been through this themselves.

Vânia de Mathia Campanhão 

That is a very good question.  We do have a KB on this topic and the recommendation is 100,000 TUs: https://gateway.rws.com/csm?id=kb_article_view&sysparm_article=KB0032215

However, I don't think 300,000 is too large and I'm well aware of many translators working regularly with far larger file based TMs than this.  Having said this the recent versions of Trados Studio do increase the physical size of a TM considerably because of the fragment pairs created during the alignment process when you "upgrade" a TM.  So depending on the type of content you have, the number of tags, the length of the segments etc. the actual size and response time is likely to vary.  So with this in mind it's definitely worth testing out whether splitting it into three separate TMs does make a significant difference for you.

You could also play with the other recommendations and test working with your TM as a cloud-based resource to see if this improves the responsiveness.  You have nothing to lose working with copies and a bit of experimenting may help.

Might also be worth seeing if people like Walter Blaser , Jerzy Czopik , Alison Field can offer any advice as they have been around a while working with Studio and have probably been through this themselves.

Vânia de Mathia Campanhão 

When you say "Trados takes too long to search this specific TM", do you refer to the TM lookup while translating in the Editor or to a manual search in the view "Translation Memories". Can you be a bit more explicit about what you consider as "too long". How long does it take when you confirm a segment until it moves to the next one? 

Does the problem still occur after you ran a reindex/upgrade of this TM?

Knowing more details will increase the chances that we can help you solve this problem. 

Hi Walter, thank you for taking the time to answer my question.

I meant that Trados takes a very long time (about 30 sec.) searching my TM, i.e. much longer than it would take searching other TMs (those provided by an agency, for example).  I'll run reindex/upgrade to see how it works.

Best regards

Hi Paul! I have a similar problem to Vânia. My TM is stored on the desktop (D:/). It has over a million TUs and my computer is a Dell (Intel® Core i5-7400 CPU). I tried to split the TM in several smaller TMs but I cannot manage. I read above that you said something about “working with my TM as a cloud-based resource” and I think that this could be a good solution for me. Could I get more details about it? Any advice is more than welcome.