Skip to contentThis article is reprinted from Zhihu Column: 11. RDMA Shared Receive Queue, Author: Savir. The IB protocol significantly reduces the memory capacity requirements on the receiving end through the SRQ mechanism. This article mainly introduces the principles of SRQ and the similarities and differences with RQ.
This article is reprinted from Zhihu Column: 10. RDMA and Completion Queue, Author: Savir. CQ and QP are interdependent and serve as the medium for hardware to "report task status" to software. This article provides analysis and explanation of most of the content related to CQ in the protocol.
This article is reprinted from Zhihu Column: 9. RDMA Queue Pair, Author: Savir. QP is the most critical concept in RDMA technology, serving as the medium for software to "issue commands" to hardware. This article analyzes and explains most of the content related to QP in the protocol.
This article discusses how to use Workbox to add PWA functionality to Hugo static websites, enhancing loading speed and user experience through Service Worker. The advantages of PWA include fast loading, offline access, push notifications, and installation to the home screen. The article provides a detailed introduction to registering Service Worker, using Workbox's caching strategies, and offers detailed configuration steps and example code.
This article focuses on introducing the basic concepts of Ollama and its notable advantages, including its open-source and free nature, ease of use, rich models, and low resource consumption. It then provides a detailed installation and usage guide, covering detailed installation steps for different operating systems and Docker environments, as well as how to download and run models. Additionally, the article introduces how to deploy Ollama on HPC clusters and how to create a local code completion assistant by integrating IDE plugins.
This article is reprinted from Zhihu column: 8. RDMA Address Handle, author: Savir. AH stands for Address Handle. In service types such as UD, the local end uses AH to specify the remote node.
This article is reprinted from Zhihu Column: 7. RDMA Protection Domain, author: Savir. To isolate various RDMA resources, the protocol introduces the concept of PD. PD stands for Protection Domain, which "groups" related software resources together, and resources between different protection domains cannot access each other.
This article is reprinted from Zhihu Column: 6. RDMA Memory Region, Author: Savir. The essence of network communication is the migration of data in storage media, and RDMA manages memory through MR. MR is a special memory region, and this article introduces its concept and function.
OneAPI is an open and standards-based specification that supports multiple architecture types, including but not limited to GPU, CPU, and FPGA. The specification has direct programming and API-based programming paradigms. Intel® OneAPI is a software suite that includes various tools and libraries to support the OneAPI specification. This article documents the process of installing Intel® OneAPI-2024.0 on CentOS 7.6.
Pure is a new programming model and runtime system designed to fully leverage shared memory within nodes in environments based on the Message Passing Interface (enhancing the utilization of idle core capabilities through tasks). Pure utilizes shared memory in two ways: (1) allowing ranks to steal work from each other while waiting for messages to arrive; (2) enabling high-performance message passing and collective operations between processes within a node using efficient lock-free data structures. Researchers evaluated the key message passing and collective features of Pure through micro benchmark tests and demonstrated that in CoMD molecular dynamics and miniAMR adaptive mesh refinement applications, Pure can achieve up to 2.1x application acceleration when scaling to 4096 ranks.
