parallel-for | ParallelFor is PHP class to iterate array in multi process

I'm getting unpredictable errors in my program when running with a lot of threads and am trying to find where the problem sits by inserting critical sections. Now I've got to the following

I try to write 2D Cross correlation in Sycl and OneAPI. The idea is to write a kind of Map skeleton which wraps OneAPI calls hiding hardware targeting issues through some parameter specifying the kind of target (CPU or GPU/Accelerator).

this is my Map Class:

I have a method in an AWS Lambda which receives a completed list of items. It processes those items and then inserts them into a Dynamo DB table. Order of processing or inserting into the DB does not matter.

I'm looking for the most thread-safe, yet easy to understand manner in which to make this run faster; Either by using async await or something other (likely parallel?) operations.

I was thinking Parallel.ForEach() but that seems a bit heavy. Is there a simpler, more obvious way?

I need to fill a lot of very large arrays by opening up thousands of csv files, extracting columns of data, and inserting them into 3D and 4D matrices. I've tried writing this in parallel, but what always happens is that my computer crashes when my memory fills up. I've looked at this question, Parallel `for` loop with an array as output, but I have not gotten those suggestions to work for me. Here's my code (generalized where needed):

I try to analyze the execution time of my function on device. I read this link: https://docs.oneapi.com/versions/latest/dpcpp/iface/event.html but I did not find in the documentation any information about sycl::info::event_profiling, which let me to understand what they correspond exactly. I mean, the command_start, command_end, command_submit. for example: This is a part of my code, kernel,

With the below code, everything usually works fine. However, very rarely (once every month perhaps, when the code is run daily) the .Find() method can't find the item it's looking for.

I am looking to parallelise processing of a task which is dependent on an object (State), which is not thread-safe, and whose construction is time-expensive.

For this reason, I was looking into partition-local variables, but either I am doing it wrong or looking for something else. This more or less represents my current implementation:

I have the following code where I want to retrieve employee information for a given list of employee ids. I have validation in place to throw exception if the count is more than 1 million. In most cases the request will be less than 200K so I am splitting the request into 4 partitions, each partition containing equal number of employee ids. All 4 partitions are executed in parallel and joined together after Task.WhenAll. Can some one please give me some hints on improving this further ? I looked at https://houseofcat.io/tutorials/csharp/async/parallelforeachasync and https://medium.com/@alex.puiu/parallel-foreach-async-in-c-36756f8ebe62 but unable to make it to work. The code mentioned below works but its hard coded to split into 4 partitions. How do I make it more parallel with dynamic partitions with max degree of parallelism set to say 50? If the input is 100K ids, I want to split in 10 partitions and execute all 10 in parallel.

I'm using TBB library like this: