Locking in Standard Item Journal Posting Routines

Does anyone have a problem with posting item journals in a multi-user environment, in Attain? We have many inventory movements recorded by scanners. These are loaded into standard Navision item journal tables by bespoke code units, and then the standard posting functions called to post them. The posting routines run serially. We have several posts happening concurrently in different parts of the factory, and there is a lot of waiting while the posting routines do one post at once. Does anyone have experience of making these functions happen concurrently? We need to stock control to be as on-line as possible.

We had the same problems, but with Sales Invoices. For now the only optimal solution /without having to re-write the whole item posting routines - which is tough and DANGEROUS task/ is to assess the bottlenecks from the hardware point of view. Then upgrade the hardware. This is the only “safe” solution for speeding up the posting.

Navision Attain contains HUGE amount of 'LOCKTABLE’s. There is only one way to allow users to work: [powerfull WORKSTATIONS] + [VERY powerfull SERVER] + [VERY fast network]. Hmmm… But you can try to comment some LOCKTABLEs… Business Applications Programmer Sertified Navision Developer SIA “Sintegra” Latvia

Andy, you may want to consider moving to SQL if you have not already. Properly configured SQL runs faster than C/SIDE as the DB gets larger. You can also take advantage of the multiprocessor box with SQL. Also, you can try putting your server on a 1Gig network concentrator, while leaving the rest of your network running on 10/100. This will eliminate congestion and it is way cheaper than a whole new server… :slight_smile: Alex

Many thanks for all your ideas. I agree that it is not a good idea to start rewriting the journal posting routines. Our software company have suggested buying Navision application server (expensive), but there are other things to try before that. It seems that the best first approach is to make the code which creates the journal lines as efficient as possible, to decrease the number of lines - posting then goes faster. We are already using the SQL database, but the 1Gig network concentrator sounds interesting. What is it, and what does it do? The users posting these journals are on a remote WAN link, connected to our VPN by a 256Kbps link, using Citrix.

A concentrator is a router that has some 100MB ports and one or two 1GB ports. Its purpose is to increase throughput and reduce network wait times. You may also find a 1GB port plug-ins to your existing router, depenging on the manufacturer. Cheap solution may be adding a second 100MB adapter to the server and configuring network parameters, to increase the speed 2x. Alex