-
### Steps to reproduce
Run 15000 queries/second on a database with multiple connection pools. CPU will be abnormally high due to contention on a lock to get connection pools.
### The issue
On o…
-
It [looks like](https://github.com/rampantpixels/rpmalloc) `rpmalloc` has benchmarks showing it surpasses lockless performance, but I'm not sure if it's the same lockless in your comparisons.
I lik…
-
Is it possible to convert this radix tree implementation to a lockless radix tree(such as using with RCU)?
-
Notably `AbstractNodeJSEnv` and `ExternalJSEnv` can probably be improved. Notably use less inheritance.
-
It would be helpful if you would publish a PDF of the output. To not pollute the master branch where you keep the sources you could publish that build output in another branch on this repository.
-
I'm running tensorflow on a machine with 64 CPUs and no GPUs. I noticed that although tf detects the number of CPU's correctly (
I tensorflow/core/common_runtime/local_device.cc:40] Local device intra…
-
`MemoryManager` causes a performance regression and should be replaced by a faster algorithm or, if we fail to find/implement one, just `malloc` and `free`, which still provides better performance tha…
-
I am scratching my head over this issue for some hours now, but I am not able to figure out why spacemacs does not properly ident my file with an 8 character indent.
Running `clang-format my-file-n…
ghost updated
6 years ago
-
Current threading requirement is that only a single native thread can call into Duktape and another thread may enter only when the previous one exits.
When implementing blocking native calls it'd b…
-
Hello,
I have an odd issue. I have an application that is using `NR_FORWARD` on *all* packets so it is connecting the host stack transparently.
With the ixgbe driver `admode=1`, iperf to this ho…