diff options
| author | Christian Krinitsin <mail@krinitsin.com> | 2025-07-03 07:27:52 +0000 |
|---|---|---|
| committer | Christian Krinitsin <mail@krinitsin.com> | 2025-07-03 07:27:52 +0000 |
| commit | d0c85e36e4de67af628d54e9ab577cc3fad7796a (patch) | |
| tree | f8f784b0f04343b90516a338d6df81df3a85dfa2 /results/classifier/gemma3:12b/performance/1272252 | |
| parent | 7f4364274750eb8cb39a3e7493132fca1c01232e (diff) | |
| download | qemu-analysis-d0c85e36e4de67af628d54e9ab577cc3fad7796a.tar.gz qemu-analysis-d0c85e36e4de67af628d54e9ab577cc3fad7796a.zip | |
add deepseek and gemma results
Diffstat (limited to 'results/classifier/gemma3:12b/performance/1272252')
| -rw-r--r-- | results/classifier/gemma3:12b/performance/1272252 | 25 |
1 files changed, 25 insertions, 0 deletions
diff --git a/results/classifier/gemma3:12b/performance/1272252 b/results/classifier/gemma3:12b/performance/1272252 new file mode 100644 index 000000000..ed580396a --- /dev/null +++ b/results/classifier/gemma3:12b/performance/1272252 @@ -0,0 +1,25 @@ + +qemu-img ftp/http convert + +Converting images with ftp or http as source could be done a lot faster. The way it works now (qemu 1.7.50) is significantly slower than the optimal way. + +FTP - how it works now +1. Connect and login to ftp-server. Ask for size of file. +2. Get a chunk of data using rest+retr +3. Goto step 1 again in a loop until all data is retrieved + +FTP - better solution +1. Connect and login to ftp-server. Dont ask for size of file. +2. Retrieve all remaining data +3. Goto step 1 again if disconnected/io error (max NN errors etc) + + +Http - how it works now +1. Connect to webserver and ask for size of file / http HEAD. +2. Get a chunk of data using http Range. +3. Goto step 1 again in a loop until all data is retrieved. + +Http - better solution +1. Connect to webserver. +2. Retrieve all remaining data. +3. Goto step 1 again if disconnected/io error (max NN errors). \ No newline at end of file |