Computer recommendations for running large iMatch databases?

[MelissaLaurino]

I am purchasing a new computer solely for running iMatch databases for marine mammal research. Are there any recommendations when purchasing a computer to make this program run more efficiently? Ex: The type of operating system, processor, or memory capacity to look for? Any help or suggestions would be greatly appreciated! Thank you!

[sinus]

I am purchasing a new computer solely for running iMatch databases for marine mammal research. Are there any recommendations when purchasing a computer to make this program run more efficiently? Ex: The type of operating system, processor, or memory capacity to look for? Any help or suggestions would be greatly appreciated! Thank you!

https://www.photools.com/community/index.php?topic=4879.msg38076#msg38076
https://www.photools.com/community/index.php?topic=5677.msg39178#msg39178

Good luck.

[Mario]

In general, current computers are pretty fast. You did not mention how large your database is, or what you do mostly with IMatch, so I can only give some general tips:

IMatch, like any database system, depends on the performance of the disk and how the disk is connected.
If you buy a modern PC with a 6 GB/s SATA interface and a high-speed SSD (pretty much standard) you will have a very fast disk. IMatch loves that.

IMatch internally splits many operations and divides the work among all processor cores installed in your computer ('parallel processing'). If you buy a modern PC, it usually has 4, 6  or even 8 processor cores (there are high-end workstations from companies like Dell or HP with up to 32 cores - yay!).
The more cores, the faster IMatch can process things like background category updates, data-driven categories, collections, filters, file and metadata import  etc. Up to a limit, when the hard disk again becomes the limiting factor.

A modern graphic card with 2 or 4 GB RAM and Windows 10.

IMatch is a 32-bit application and can use up to 3.5 GB of RAM, which is plenty. If your computer has 8 or 16 GB RAM, there will be sufficient room for IMatch and other applications you run (even when you run Photoshop).

If you are an institutional client your IT folks will know what to do with this info. Institutions often buy "workstation"-type computers which are finely tuned, super-reliable and come with support for 3 years or more (Dell, HP, Fujitsu).

If you are buying a PC for yourself, use the above info for checking feature lists. You don't need a high-end gaming class PC or anything, The disk and processor cores are the most important components for performance.

And, please note: There are limits, even for IMatch.

If you ask one of the large vendors (Canto, FotoWare, Extensis, Widen, AssetBank, ...)  to setup a DAM which can handle hundreds of thousands of files, they will start by recommending a dedicated server (farm) or maybe a cloud-based exclusive solution...and the corresponding licensing and pricing structure.
For the price of a full IMatch license you get roughly one hour of consulting time from these companies 

IMatch can handle several hundred thousand of images per database. But of course the more images you add, the slower certain operations will become. If IMatch has to analyze 200,000 files instead of 100,000 to rebuild a data-driven category, it will take twice as long.

IMatch has features that allow you to deal with that, e.g. by switching rarely used data-driven categories to manual updates. To disable automatic write-back and run it only at convenient times. To bulk load massed of new files when you don't need your PC for other things, etc.

The new IMatch Anywhere^TM software

see http://www.photools.com/imatch-anywhere/

makes it much easier to distribute an IMatch database within your institute or company. IMatch Anywhere WebServices^TM have been designed to be fast even on slower hardware. Users then can access and browse your IMatch database using any web browser, from any device. And since IMatch WebServices does not have a user interface, it can concentrate on handling the database very fast, even large databases.