cross | “ Zero setup ” cross compilation and “ cross testing ” | Build Tool library

If you are using NLog the easiest way to log in you startup.cs is to add private property.

This was a discovery for me but yes, it is now possible to use init containers with docker-compose since version 1.29 as can be seen in the PR you linked in your question.

Meanwhile, while I write those lines, it seems that this feature has not yet found its way to the documentation

You can define a dependency on an other container with a condition being basically "when that other container has successfully finished its job". This leaves the room to define containers running any kind of script and exit when they are done before an other dependent container is launched.

To illustrate, I crafted an example with a pretty common scenario: spin up a db container, make sure the db is up and initialize its data prior to launching the application container.

Note: initializing the db (at least as far as the official mysql image is concerned) does not require an init container so this example is more an illustration than a rock solid typical workflow.

The complete example is available in a public github repo so I will only show the key points in this answer.

Let's start with the compose file

you can use window function first_value:

You can see the cost of instructions on most mainstream architecture here and there. Based on that and assuming you use for example an Intel Skylake processor, you can see that one 32-bit imul instruction can be computed per cycle but with a latency of 3 cycles. In the optimized code, 2 lea instructions (which are very cheap) can be executed per cycle with a 1 cycle latency. The same thing apply for the sal instruction (2 per cycle and 1 cycle of latency).

This means that the optimized version can be executed with only 2 cycle of latency while the first one takes 3 cycle of latency (not taking into account load/store instructions that are the same). Moreover, the second version can be better pipelined since the two instructions can be executed for two different input data in parallel thanks to a superscalar out-of-order execution. Note that two loads can be executed in parallel too although only one store can be executed in parallel per cycle. This means that the execution is bounded by the throughput of store instructions. Overall, only 1 value can only computed per cycle. AFAIK, recent Intel Icelake processors can do two stores in parallel like new AMD Ryzen processors. The second one is expected to be as fast or possibly faster on the chosen use-case (Intel Skylake processors). It should be significantly faster on very recent x86-64 processors.

Note that the lea instruction is very fast because the multiply-add is done on a dedicated CPU unit (hard-wired shifters) and it only supports some specific constant for the multiplication (supported factors are 1, 2, 4 and 8, which mean that lea can be used to multiply an integer by the constants 2, 3, 4, 5, 8 and 9). This is why lea is faster than imul/mul.

I can reproduce the slower execution with -O2 using GCC 11.2 (on Linux with a i5-9600KF processor).

The main source of source of slowdown comes from the higher number of micro-operations (uops) to be executed in the -O2 version certainly combined with the saturation of some execution ports certainly due to a bad micro-operation scheduling.

Here is the assembly of the loop with -Os:

You end up with the error with precision because some of your penalization is too strong for this model, if you check the results, you get 0 for f1 score when C = 0.001 and C = 0.01

You need add a transparent hole in .curtain class:

Use a Parameter Store value with a CustomResource.

This answer has a full Typescript CDK example of cross-region refs.

(I originally posted this as a comment because I thought the question was perhaps a duplicate. But on reflection, I see that the linked question and tags only mention CloudFormation, not the CDK. Seems the community gets the most benefit from keeping this question alive).

Here is the benchmarking for the existing to OP's question (borrow test data from @Marek Fiołka but with n <- 10000)