convoy | Redis-backed unique job | Runtime Evironment library

I have a heterogenous list as described in CPDT:

I am following the book Computational Type Theory and Interactive Theorem Proving with Coq, and one of the exercises is for me to write of term of type:

I have a request to execute an orch only when a particular file is present, so for example:

1. My customer will load several XML files.
2. After all XML files were loaded, an special file called "OK.CTRL" will be created by them to point out that the XML filecopy finished.
3. My orch should be listening on a receive port for this "OK.CTRL" file creation. When exists then it should start processing all the XML files.

I've been trying to do it in several ways:

1. With 2 receive ports, the first as activated and the second not, but then I got an error: "you must specify at least one already-initialized correlation set for a non-activation receive that is on a non-selfcorrelating port".
2. Same as 1, but with both on activated, and then there's another error: "an activatable receive must be the first executable statement in a service".
3. Using a correlating set, but then the error "in a sequential convoy the ports must be identical", but I need 2 different ports, since file mask are different.

I tried as well with several combinations, but the results are always one of the 3 stated above.

Any ideas on how can I get it to work?

Is the only difference between starvation and convoy effect that convoy effect is mainly defined on FCFS scheduling algorithms and starvation is on priority based scheduling?

I researched on both effects but couldn't find a comparison. This is based on operating systems theory which I learned for my college degree.

The example I will give is a train. A train has multiple train-cars and in my system each train car will send me a packet of info on a determined interval. For example, I can guarantee I will have at least 1 packet from each train-car every 10 minutes.

What I want to do is show an animation of the train cars proceeding along the map.

The problem I have is how to align all the data to make a train - visualization with each car in it's relative position to each other - when the data is recorded at different times and without any order relevant to the order of the train cars?

What ends up happening is sometimes car2 will look like it's in front of car1! In the example below, if I showed the data reported for this period, car2 looks like it's on top of car1!

For example

I have a Direct to MessageBox send port publishing a message, for which I would like to have the InterchangeID promoted so that I can create a sequential convoy against the current batch of messages. I am using an initializing correlation set, declaring BTS.InterchangeID as the correlated property, and applying it to the Send action.

When I run the code, these messages cause routing errors. Looking at the routing failure report, I can see that the InterchangeID is present but is "Not Promoted".

Why in this particular case is the correlation set not promoting the property when this usually just works?

I have a SQL Server table [hobbies] which looks like this:

I'm writing a toy language where nodes in the AST can have any number of children (Num has 0, Arrow has 2, etc). You might call these operators. Additionally, exactly one node in the AST might be "focused". We index the data type with Z if it has a focus, or H if it doesn't.

I need advice on a few parts of the code. Hopefully it's alright to ask all of these at once, since they're related.

1. How would you define the type of internal nodes with one focus, InternalZ? Right now I say "we have S n children -- n of them are unfocused and one (at some given index) is focused. A slightly more intuitive option (which looks zipper-like) would be InternalZ : forall n m, arityCode (n + 1 + m) -> Vector.t (t H) n -> t Z -> Vector.t (t H) m -> t Z. I know I don't want to deal with that addition, though.

2. Refining types: In both interesting cases in eq I compare the two ns (number of children). If they're the same, I should be able to "coerce" the arityCodes and Vector.ts to have the same type. Right now I hacked this with two lemmas. How should I do this properly? It seems like Adam Chlipala's "convoy pattern" might help but I couldn't work out how.

3. If I uncomment either of the Vector.eqb calls, Coq complains "Cannot guess decreasing argument of fix.". I understand the error but I'm not sure how to circumvent it. The first thing that comes to mind is that I might have to index t by its depth of children.

My code:

We have an application that uses Kestrel to serve HTTP requests. We've had some problems in the past where high load caused the thread pool to spawn thousands of threads, at which point we would get lock convoy symptoms. Most of the time, the threads would start blocking each other at Monitor.Enter() somewhere in our code, causing delays and more contentions until the application became unresponsive with a 100% CPU usage due to context switching. The problem would not go away until we restarted the application.

However, we've eliminated most locks and implemented a throttling mechanism so that we don't allow more than 1000 threads to enter the application. We're using the System.Threading.Semaphore class to allow only a set number of threads to continue. This has solved our lock contention problems, but possible introduced a new problem:

We still get cases of 100% CPU usage and high thread count (500-1000 threads), although this time the threads are not blocked on Monitor.Enter(). Instead, when we do thread dump (using Microsoft.Diagnostics.Runtime.ClrRuntime), we see the following call stack (for hundreds of threads):