Author: Min Wang

In controlled circumstances you can pass the shared pointer by constant reference. Be sure that nobody is concurrently deleting the object, though this shouldn’t be too hard if you’re careful about…

Background/Use case sometimes we need want to based on the destimation/domain to use different STMP server ( instead of relying on automatically search mx record) the transport_maps is a good…

https://www.furygpu.com/blog/hello   https://eater.net/8bit https://digilent.com/shop/arty-z7-zynq-7000-soc-development-board/   https://www.amd.com/en/products/system-on-modules/kria.html  

Amazon DynamoDB: redis/key-pais store, no sql Amazon EMR (Elastic MapReduce): Apache Hadoop, Apache Spark, Apache Hive, and Presto Amazon Redshift: is based on PostgreSQL Amazon Athena: analysis data in S3…

The pybind11 make python calling any c++ function easier https://pybind11.readthedocs.io/en/stable/basics.html   Also we can invoke python script from c++ as well: https://pybind11.readthedocs.io/en/stable/advanced/smart_ptrs.html   Other alternatives: boost python https://www.boost.org/doc/libs/1_84_0/libs/python/doc/html/index.html  

modules concept ( can put some limit/restrict on the template type ) template <template-parameter-list> concept concept-name = constraint-expression; https://ggulgulia.medium.com/c-20-concepts-part-1-the-basics-40f051c72776 #include <iostream> #include <vector>//first define a concept for print //lets call it…

Docker image: https://hub.docker.com/r/siutin/stable-diffusion-webui-docker docker pull siutin/stable-diffusion-webui-docker:latest-cpu   https://github.com/siutin/stable-diffusion-webui-docker Run it: docker run -it –name sdw –network host -v $(pwd)/models:/app/stable-diffusion-webui/models -v $(pwd)/outputs:/app/stable-diffusion-webui/outputs –rm siutin/stable-diffusion-webui-docker:latest-cpu bash webui.sh –skip-torch-cuda-test –no-half –use-cpu all –share…

(1) use container https://github.com/abetlen/llama-cpp-python/pkgs/container/llama-cpp-python (2) mount k8s storage as /models export MODEL point to the right llama-model.gguf (3) expose 8000 to loadbalancer to outside (4) browse to ip:8000/docs for API…

Ollama is a front-end written in go, and wrap-up the back-end of llama.cpp Here is the steps for setup ollama in k8s cluster (1) write a k8s yaml file, where…

HW Purpose: Add old laptop  to existing k8s HW: Add Laptop with only Wifi ( no ethernet port) Software debian 12 (1) k8s apt-get install bridge-utils apt-get install nfs-common sudo…