Pager 4 - Cuterwrite's Blog

👋 Welcome to Cuterwrite 's Blog

Research Related

Notes: Pure - Improving Message Passing to Better Utilize Intra-Node Shared Memory

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.

Research Related

Scientific Research Chart Drawing

In scientific research, precise and expressive data visualization is a crucial step. This article aims to explore and recommend a series of professional and widely used scientific charting tools, which have become indispensable work partners for researchers due to their powerful features and flexibility.

High Performance Computing

RDMA Basic Service Types

This article is reprinted from Zhihu Column: 5. Basic Service Types of RDMA, Author: Savir. Just like the UDP and TCP in the transport layer of the TCP/IP protocol stack, RDMA technology also specifies service types for different scenarios, with each type emphasizing different aspects of reliability and speed. This article introduces the characteristics and application scenarios of 4 main service types.

High Performance Computing

RDMA Operation Types

This article is reprinted from Zhihu column: 4. RDMA Operation Types, author: Savir. There are various communication models/modes in RDMA. Based on the CPU's involvement during data transmission and reception, they can be divided into two-sided operations requiring the participation of both sending and receiving ends, and one-sided operations requiring only the initiator's involvement. This article introduces the data transfer process in software and hardware for different operation types.

High Performance Computing Comprehensive Tech Stack

Setting Up the Xuantie 900 Series Toolchain and xuantie-qemu Environment

With the rapid development of domestic RISC-V chip technology, Alibaba Pingtouge's Xuantie 900 series processors have attracted much attention for their high performance and low power consumption. To better support the development and debugging work based on the Xuantie 900 series chips, it is crucial to set up a complete toolchain and simulation environment. This article will provide a detailed introduction on how to set up the Xuantie 900 series toolchain and xuantie-qemu environment.