本文目标

        使用Akamai Cloud VPU资源完成一个视频转码任务。具体为将一个1080P的H.264格式视频转出成H.265格式。以此让读者熟悉基于VPU资源构建转码服务的基本步骤。

测试结论

资源	转码时长	转码后体积
VPU	2.898s	14MB
CPU	1m25.862s	8.2MB

相关资源

实例规格：NETINT Quadra T1U x1 Small
操作系统：ubuntu 24.04lts

VPU软件包位置：https://docs.netint.com/vpu/quadra/
当前版本：Release: v5.4.0（2025年11月7日）

源视频位置：https://github.com/chthomos/video-media-samples#

准备工作

创建VPU实例

系统更新

sudo apt update && sudo apt upgrade -y && sudo apt install unzip nvme-cli

下载安装包

wget https://releases.netint.com/quadra/POFCG6D02HW7XXX//Quadra_V5.4.0.zip
unzip Quadra_V5.4.0.zip
cd Quadra_V5.4.0/

安装软件

bash quadra_quick_installer.sh

进入安装步骤

设置系统参数

设置权限

安装依赖包

安装Libxcoder

安装ffmpeg软件，具体版本可以根据需求进行选择

下载样例视频

git clone https://github.com/chthomos/video-media-samples.git

转码测试

使用VPU转码

time ffmpeg -y -hide_banner -nostdin -vsync 0 -c:v h264_ni_quadra_dec -i /root/video-media-samples/big-buck-bunny-1080p-30sec.mp4 -c:v h265_ni_quadra_enc /root/output.h265.mp4
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/root/video-media-samples/big-buck-bunny-1080p-30sec.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf56.25.101
  Duration: 00:00:30.02, start: 0.000000, bitrate: 6053 kb/s
    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], 5860 kb/s, 24 fps, 24 tbr, 12288 tbn, 48 tbc (default)
    Metadata:
      handler_name    : VideoHandler
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, 5.1, fltp, 192 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
[SID=7e3f]|| ni_set_custom_dec_template height 1080 width 1920 fps_number 24 fps_denominator 1
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (h264_ni_quadra_dec) -> hevc (h265_ni_quadra_enc))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))
[h265_ni_quadra_enc @ 0x13de3400] Session state: 0 allocate frame fifo.
[h265_ni_quadra_enc @ 0x13de3400] pix_fmt is 0, sw_pix_fmt is -1 resolution 1920x1080
[h265_ni_quadra_enc @ 0x13de3400] sw_pix_fmt assigned to pix_fmt was 0, is now -1
[h265_ni_quadra_enc @ 0x13de3400] dts offset: 3, gop_offset_count: 0
[SID=2249]|| Encoder_complete_info:session_id 0x2249, total frames input:0 buffered: 0 completed: 0 output: 0 dropped: 0 error: 0
[h265_ni_quadra_enc @ 0x13de3400] xcoder_setup_encoder sw_pix_fmt 0 ori_pix_fmt 0
Output #0, mp4, to '/root/output.h265.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.45.100
    Stream #0:0(und): Video: hevc (h265_ni_quadra_enc) (hvc1 / 0x31637668), yuv420p(tv), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 24 fps, 12288 tbn, 24 tbc (default)
    Metadata:
      handler_name    : VideoHandler
      encoder         : Lavc58.91.100 h265_ni_quadra_enc
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, 5.1, fltp, 341 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
      encoder         : Lavc58.91.100 aac
frame=  720 fps=252 q=27.0 Lsize=   14005kB time=00:00:30.01 bitrate=3822.4kbits/s speed=10.5x    
video:12729kB audio:1255kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.152426%
[SID=7e3f]|| Decoder_complete_info:session_id 0x7e3f, total frames input:720 buffered: 0 completed: 720 output: 720 dropped: 0 error: 0 corrupted: 0
[SID=d277]|| Encoder_complete_info:session_id 0xd277, total frames input:720 buffered: 0 completed: 720 output: 720 dropped: 0 error: 0
[aac @ 0x13daf580] Qavg: 3155.695

real    0m2.898s
user    0m0.658s
sys     0m0.468s

使用CPU转码

time ffmpeg -y -hide_banner -nostdin -vsync 0 -c:v h264 -i /root/video-media-samples/big-buck-bunny-1080p-30sec.mp4 -c:v hevc /root/output-libx265.h265.mp4
-vsync is deprecated. Use -fps_mode
Passing a number to -vsync is deprecated, use a string argument as described in the manual.
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/root/video-media-samples/big-buck-bunny-1080p-30sec.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf56.25.101
  Duration: 00:00:30.00, start: 0.000000, bitrate: 6058 kb/s
  Stream #0:0[0x1](und): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 1920x1080 [SAR 1:1 DAR 16:9], 5860 kb/s, 24 fps, 24 tbr, 12288 tbn (default)
    Metadata:
      handler_name    : VideoHandler
      vendor_id       : [0][0][0][0]
  Stream #0:1[0x2](und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, 5.1, fltp, 192 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
      vendor_id       : [0][0][0][0]
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> hevc (libx265))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))
x265 [info]: HEVC encoder version 3.5+1-f0c1022b6
x265 [info]: build info [Linux][GCC 13.2.0][64 bit] 8bit+10bit+12bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [info]: Main profile, Level-4 (Main tier)
x265 [info]: Thread pool created using 2 threads
x265 [info]: Slices                              : 1
x265 [info]: frame threads / pool features       : 1 / wpp(17 rows)
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge         : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias  : 24 / 250 / 40 / 5.00 
x265 [info]: Lookahead / bframes / badapt        : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb       : 1 / 1 / 0
x265 [info]: References / ref-limit  cu / depth  : 3 / off / on
x265 [info]: AQ: mode / str / qg-size / cu-tree  : 2 / 1.0 / 32 / 1
x265 [info]: Rate Control / qCompress            : CRF-28.0 / 0.60
x265 [info]: tools: rd=3 psy-rd=2.00 early-skip rskip mode=1 signhide tmvp
x265 [info]: tools: b-intra strong-intra-smoothing lslices=6 deblock sao
Output #0, mp4, to '/root/output-libx265.h265.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf60.16.100
  Stream #0:0(und): Video: hevc (hev1 / 0x31766568), yuv420p(progressive), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 24 fps, 12288 tbn (default)
    Metadata:
      handler_name    : VideoHandler
      vendor_id       : [0][0][0][0]
      encoder         : Lavc60.31.102 libx265
    Side data:
      cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: N/A
  Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, 5.1, fltp, 341 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
      vendor_id       : [0][0][0][0]
      encoder         : Lavc60.31.102 aac
[out#0/mp4 @ 0x595320c85340] video:7079kB audio:1253kB subtitle:0kB other streams:0kB global headers:2kB muxing overhead: 0.293141%
frame=  720 fps=8.4 q=35.9 Lsize=    8357kB time=00:00:29.99 bitrate=2282.4kbits/s speed=0.35x    
x265 [info]: frame I:      6, Avg QP:26.25  kb/s: 29779.26
x265 [info]: frame P:    248, Avg QP:27.78  kb/s: 4515.13 
x265 [info]: frame B:    466, Avg QP:34.80  kb/s: 199.29  
x265 [info]: Weighted P-Frames: Y:11.3% UV:3.6%
x265 [info]: consecutive B-frames: 26.8% 15.0% 9.8% 44.9% 3.5% 

encoded 720 frames in 85.63s (8.41 fps), 1932.36 kb/s, Avg QP:32.31
[aac @ 0x595320c95840] Qavg: 451.761

real    1m25.862s
user    2m47.321s
sys     0m0.812s

Akamai Cloud VPU 产品简介

        Akamai Linode 提供的 VPU 实例（Video Processing Unit） 基于 NETINT Quadra T1U 硬件卡，专为高密度视频编解码和 AI 视频分析场景设计。该卡采用 SoC 架构，集成硬件级编码、解码、AI 推理引擎（NNP）以及 NVMe 接口，具备低功耗、高吞吐的优势。