How many times have you gotten the “Not supported value type in a shared object” error when trying to put an object into a shared object?
What did you do? Write longer code to turn your innocent object into a shared one, perhaps iterating over all of its properties?
How about transferring a shared object from one shared group to another?
An interesting thing about objects: you already know they can be used as a hash map (a key/value system), but what about gettings all the keys or all the values with a single line of code?
Keep reading, because 4D v18 R3 is out and brings good news!
Have you ever needed to use methods with both thread-safe and not thread-safe calls, with conditions that skip over commands that are not thread-safe? Currently the compiler prevents doing this and an error is thrown, however there’s a flag that lets you disable this check and this blog post shows you how.
Preemptive processes are very useful for performing extensive calculations in parallel, using all of the cores in your CPU, but sometimes you need to delegate tasks to a cooperative process (e.g., using a non-preemptive command). With the New signal command, you can pause a process and wait for the response from another process!
Preemptive processes allow your applications to take full advantage of multi-core computers, resulting in faster responses to simultaneous requests. Beginning with 4D v17 R4, Begin SQL, End SQL, and the associated SQL statements used with an internal database are now preemptive compliant.
Do you need to run complex calculations or exports on the client side? Use preemptive processes and start these actions without slowing down users and their daily work!
4D lets you take full advantage of multi-core computers with preemptive processes. In a previous blog post, we’ve provided you with a list of enhanced commands that can be used in preemptive processes, enabling faster global execution time and more connected users. And in response to your feedback, we increased the list with even more commands.
Preemptive processes are powerful since they allow your applications to take full advantage of multi-core computers; consequently, this results in faster global execution time and more connected users. With 4D v17, we went a step further by providing you 1,000+ thread-safe commands!
Watch the Preemptive multi-threading video from the 4D Summit 2018 Keynote. It features Add Komoncharoensiri, Director of Technical Services (4D USA), explaining how the preemptive support in 4D helps you take advantage of your multi-core machines, as well as the new concepts we shipped along the way to help you boost performance and efficiency.
Preemptive Web Services allow your applications to take full advantage of multi-core computers resulting in faster responses for simultaneous requests. It’s not only powerful, it’s also easy: from the Method Properties of any method, just select the “Can be run in preemptive processes” option and you’re off and running!
A 4D method can be preemptive or not according to compilation options and the use of the language, as explained in this post.
4D v16 R6 introduces a new concept of communication between processes: shared object and shared collection variables! They are a solution so you can avoid using interprocess variables (which can’t be used in multi-thread mode). Thanks to this, you will be able to easily share information between preemptive processes.
A preemptive process is powerful as it allows your application to take full advantage of multi-core computers; consequently, its results in a faster global execution time and more connected users. A 4D process may be preemptive or not according to compilation options and the use of the language, as explained in this post.
In order to benefit from this mode as much as possible we are pleased to announce that more than twenty existing commands have been enhanced and can now be used inside preemptive processes.
Got a question, suggestion or just want to get in touch with the 4D bloggers? Drop us a line!
* Your privacy is very important to us. Please click here to view our Policy