## docker 原理与应用实践 张成远 容器系统整体架构 • Namespace • CGroup • Device Mapper • Pull Image • Start Container • Stop Container • Docker Image Storage ## 容器系统整体架构 ### JD.COM 京东 kubernetes Docker Etcd 0efa24/p3_1.jpg) Fleet Channel Libchan lxc Weave linux CGroup Namespace Aufs/Btrfs Device Mapper capabilities netlink netfilter ## Namespace ### JD.COM 京东 • 提供进程级别的资源隔离 • 为进程提供不同的命名空间视图 框架为上层应用提供了丰富的设备映射及 IO 策略方面的支持 • Docker 存储端实现之一使用 DM - thin provision • 上层通过 dmsetup 工具或 libdevmapper 库使用 ## Device Mapper ### JD.COM 京东 

人工智能资料下载，可访问百度：尚硅谷官网 <mapper resource="EmployeeMapper.xml"> mapper> ### 2.4 创建 Mybatis 的 sql 映射文件 ## 1 ) 参考 MyBatis MyBatis 的官方手册 <mapper namespace="suibian"> select id, last_name lastName, email tbl_employee where id = #{id} mapper> ### 2.5 测试 1) 参考 MyBatis 的官方手册 @Test public void test() throws Exception { String resource =

--ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` Where pytorch.org/whl/cpu https:// storage.openvinotoolkit.org/simple/wheels/nightly/" VLLM_TARGET_DEVICE=openvino python -m pip install -v . ``` ## 1.3.4 Performance tips vLLM OpenVINO backend uses https://download.pytorch.org/ →whl/cpu ``` - Finally, build and install vLLM CPU backend: $ VLLM_TARGET_DEVICE=cpu python setup.py install Note: - BF16 is the default data type in the current CPU backend (that

--ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` pytorch.org/whl/cpu https:// storage.openvinotoolkit.org/simple/wheels/nightly/" VLLM_TARGET_DEVICE=openvino python -m pip install -v . ``` ## 1.3.4 Performance tips vLLM OpenVINO backend uses https://download.pytorch.org/ →whl/cpu ``` - Finally, build and install vLLM CPU backend: $ VLLM_TARGET_DEVICE=cpu python setup.py install Note: - BF16 is the default data type in the current CPU backend (that

systems. You can create such a build with the following commands: ```bash $ export VLLM_TARGET_DEVICE=empty $ pip install -e . ``` Tip: Building from source requires quite a lot compilation. If you --ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` OpenVINO backend: $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" VLLM_TARGET_ →DEVICE=openvino python -m pip install -v . ## 1.3.4 Performance tips vLLM OpenVINO backend uses the following

systems. You can create such a build with the following commands: ```bash $ export VLLM_TARGET_DEVICE=empty $ pip install -e . ``` Tip: Building from source requires quite a lot compilation. If you --ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` OpenVINO backend: $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" VLLM_TARGET_ →DEVICE=openvino python -m pip install -v . ## 1.3.4 Performance tips vLLM OpenVINO backend uses the following

systems. You can create such a build with the following commands: ```bash $ export VLLM_TARGET_DEVICE=empty $ pip install -e . ``` Tip: Building from source requires quite a lot compilation. If you --ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` OpenVINO backend: $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" VLLM_TARGET_ →DEVICE=openvino python -m pip install -v . ## 1.3.4 Performance tips vLLM OpenVINO backend uses the following

--ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` pytorch.org/whl/cpu https://storage. →openvinotoolkit.org/simple/wheels/pre-release" VLLM_TARGET_DEVICE=openvino python →-m pip install -v . ``` ## 1.3.4 Performance tips vLLM OpenVINO backend uses https://download.pytorch.org/ →whl/cpu ``` - Finally, build and install vLLM CPU backend: $ VLLM_TARGET_DEVICE=cpu python setup.py install ## Note: - BF16 is the default data type in the current CPU backend

roduction| |4|wireguard-vpn-peer2site-router|On router| |4|wireguard-vpn-peer2site-inside|Inside device| |3|wireguard-vpn-site2site|Site-to-site| |3|wireguard-vpn-defaultgw|Default gateway| |3|wiregua Azure| ||Multipath| |device-mapper-multipathing-introduction|Introduction| |device-mapper-multipathing-configuration|Configuration| |device-mapper-multipathing-setup|Setup| |device-mapper-multipathing-usage-debug|Usage -agents|Pacemaker - fence agents| |ubuntu-ha-pacemaker-fence-agents|Distributed Replicated Block Device (DRBD)| |ubuntu-ha-drbd|Ubuntu HA - Migrate from crmsh to pcs| |ubuntu-ha-migrate-from-crmsh-to-pcs|Databases|

systems. You can create such a build with the following commands: ```bash $ export VLLM_TARGET_DEVICE=empty $ pip install -e . ``` Tip: Building from source requires quite a lot compilation. If you --ipc=host \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --device /dev/kfd \ --device /dev/dri \ -v :/app/model \ vllm-rocm \ bash ``` OpenVINO backend: $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" VLLM_TARGET_ →DEVICE=openvino python -m pip install -v . ## 1.3.4 Performance tips vLLM OpenVINO backend uses the following