-
Newly added support for unirec array needs to be tested.
https://github.com/CESNET/Nemea-Framework/tree/unirec_array
-
```
> hyriseBenchmarkTPCDS --verify -r 1
...
Using SQLite's result table as expected result table
- Verification failed: Expected result table is not empty, but Hyrise's actual result is (48 µs 90…
-
Some operator that is mostly used by TPC-H 22 is calling get_segment too often. See the following benchmark:
I used the following benchmarks parameters on nemea: `--scheduler --clients 100 -s 1 -t …
-
Commenting out `_simd_scan_with_iterators` leads to several minutes of saved compile time.
**Before**
TableScanImpl | With SIMD | Without SIMD | SIMD impact
------ | ----------- | -----------…
-
When the tools are being built for the first time and configure scripts are running, the configure script of nemea-supervisor misses libtrap which is included in other part of the project.
```
=== c…
-
When building on opensuse-leap 15, I'm getting the following error:
```
$ make
make all-recursive
make[1]: Vstupuje se do adresáře „/home/krejci/workspace/nemea-modules“
Making all in aggregator…
-
TPC-H 18 (scale 1) drops to 0 iter/s when using 224 cores and 100 clients on nemea, which is caused by many concurrent allocations during the aggregation phase, when constructing the elements in the `…
-
...especially, this means that they only use a single thread to translate and optimize queries and this could be a bottleneck. Join Order Benchmark queries can take non-negligible time to optimize.
…
-
Unirecfilter sometimes gets itself into a bad state, i think it is when it is started at about the same time as all other Nemea modules.
When it gets into this state it keeps answering output ifc n…
-
For a change that does basically nothing, I got this on nemea (sf 0,1, gcc, `numactl -N 1 -m 1`):
```
Benchmark | prev. iter/s | new iter/s | change
-----------+--------------+------------+-----…