- 博客(187)
- 资源 (2)
- 收藏
- 关注
RSS订阅转载 Difference Between Wrk and Wrk2
Source fromhttps://github.com/giltene/wrk2Wrk model, which is similar to the model found in many current load generators, computes the latency for a given request as the time from the sending the first byte of the quest to the time complete response was
2023-05-19 11:43:59
200
原创 Memtiter-benchmark源码解析5 client,connection_manager
client 为 connections_manager的一个子类
2023-04-18 15:34:39
81
原创 Memtiter-benchmark源码解析4memtier_benchmark.cpp源代码解读
m_thread 是 pthread 结构体，将cg_thread (this)本身传进去。client.cpp 通过 MAIN_CONNECTION 连接服务器完成IP连接。一个client_group 拥有一个m_base。
2023-04-18 15:28:10
439
原创 Memtiter-benchmark源码解析4--clients.cpp, run_stats.cpp, shard_connections联合统计latency
Memtiter-benchmark源码解析4--clients.cpp, run_stats.cpp, shard_connections联合统计latency
2023-03-06 10:44:40
102
原创 multilate源代码解读3 AgentStats.h ConnectionStats.h LogHistorgramSampler.h 三者联合起来统计latency
mulitilate
2023-03-05 10:31:36
358
原创 机器学习新方法
Cost-Sensitive Multi-Class ClassificationCost-sensitive multi-class classification is a problem related to multi-class classification, in which instead of there being one or more “correct” labels for each observation, there is an associated vector of co.
2023-02-24 14:53:36
143
原创 Memtiter-benchmark源码解析2--shard_connection
memtitier-benmark--shard_connection
2022-12-28 09:43:02
509
原创 docker swarm
集群管理docker swarm join --token SWMTKN-1-55g0lk1oofqdmf6adpn4idy7wxvql65t6rlhmq0khwlmwq5t6j-bkijx6trp01n8fn3hwhf5h7ap 10.10.1.195:2377servicedocker service create --name busybox busybox:latest sh -c “while true; do sleep 3600; done”docker service lsd
2022-04-06 20:18:31
2375
原创 数学统计知识
Coefficient of determinationhttps://en.wikipedia.org/wiki/Coefficient_of_determinationIn the best case, the modeled values exactly match the observed values, which results in {\displaystyle SS_{\text{res}}=0}{\displaystyle SS_{\text{res}}=0} and {\displa
2022-03-30 14:35:20
250
原创 pytorch学习总结
torch.zeroshttps://pytorch.org/docs/stable/generated/torch.zeros.htmltorch.repeathttps://pytorch.org/docs/stable/generated/torch.Tensor.repeat.htmltorch.unsqueezehttps://pytorch.org/docs/stable/generated/torch.unsqueeze.html?highlight=unsqueeze
2022-03-28 11:09:20
802
原创 linux工具总结
测试系统性能 sysbench Sysbench is an open-source and multi-purpose benchmark utility that evaluates the parameter features tests for CPU, memory, I/O, and database (MySQL) performance.stress and stress-ng It is a simple workload generator for POSIX systems
2022-03-22 20:20:37
429
原创 linux常用命令
常用命令top那么首先需要获取这个进程的PID:ps -ef|grep [process name]获取某个进程名的进程号ps -ef |grep mesos | grep -v grep | awk ‘{print $2}’ps -ef |grep YarnChild | awk ‘{print $2}’| xargs kill -9然后查看该进程的CPU:top -p [PID]查看这个进程的各个线程的CPU:top -H -p [PID]1.CPU占用最多的前10个进
2022-03-15 15:49:12
3599
原创 shell脚本语法
BASH for 循环小结#!/bin/bash# test.sh MAX=10 for (( i = 0; i < MAX ; i ++ ))do echo $idone 有趣的是，这里的 MAX 前面没有 '$'，当然，你使用 $MAX 也是可以的。难道 bash 就那么宽容吗？听说，类似C的这种语法，只有bash独有哦～顺便提一下，在bash中使用expr进行简单的数学计算，如下： MAX=10 for (( i = 0; i < MAX ; i
2022-03-14 17:04:56
1739
原创 Protean:VM Allocation Service at Scale
{OSDI 2020}（Protean:VM Allocation Service at Scale）{Year}, {Authors}, {Journal Name}{引用格式}We describe the design and implemetation of Protean- the Microsoft Azure service responsible for allocating Virtual Machines (VMs) to millions of servers of around
2022-02-08 10:51:36
146
原创 Shenango NSDI 2019Achieving High CPU efficiency for Latency-sensitive Datacenter workloads
附 1. 我的论文笔记框架（Markdown格式）{Title}（文章标题）{Year}, {Authors}, {Journal Name}{引用格式}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)作者的研究目标是什么？Background / Problem StatementCore allocations imp
2022-01-17 11:38:30
201
原创 Sinan:ML-Based and QoS-Aware Resource Management for Cloud Microservices
Sinan:ML-Based and QoS-Aware Resource Management for Cloud Microservices{2021}, {Yanqi Zhang}, {ASPLOS}{引用格式}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)Microservice complicate resource
2022-01-13 09:41:58
245
原创 isca_2021_papers_list
Confidential Serverless Made Efficient with Plug-In Enclaves{Title}（文章标题）{2021}, {Mingyu Li, Yubin Xia, Haibo Chen}, {Journal Name}SummaryServerless computing is becoming the next generation of cloud computing.Real-world serverless characterization re
2021-12-21 11:48:40
427
原创 ATORI : Efficient and Fair Resource Partitioning by Sacrificing Short-Term Benefits for Long-Term
{ATORI : Efficient and Fair Resource Partitioning by Sacrificing Short-Term Benefits for Long-Term}（文章标题）{2021}, {Rohan Basu Roy}, {ISCA}{引用格式}SummarySATORI is a novel strategy to partition multi-core architectural resources to achieve two conflicting
2021-12-21 11:39:03
148
原创 Morpheus: Towards Automated SLOs for Enterprise Clusters
{Morpheus: Towards Automated SLOs for Enterprise Clusters}{2016}, {}, {OSDI}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)Background / Problem StatementMethod(s)The second part of our in
2021-11-19 09:41:04
226
原创 FIRM: An intelligent Fine-Grained Resource Management Framework for SLO-Ooritented Microservices
{FIRM: An intelligent Fine-Grained Resource Management Framework for SLO-Ooritented Microservices}{2020}, {Haoran Qiu}, {OSDI}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)User-facing late
2021-08-24 09:42:17
479
原创 Quasar:Resource-Efficient and Qos-Aware Cluster Management
我的论文笔记框架（Markdown格式）Quasar:Resource-Efficient and Qos-Aware Cluster Management{2014}, {Christina}, {ASPLOS}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)Background / Problem Statement:Me
2021-07-21 10:18:38
310
原创 PerfIso: Performance Isolatin forCommercial Latency-Sensitive Services
我的论文笔记框架（Markdown格式）PerfIso: Performance Isolatin forCommercial Latency-Sensitive Services{2018}, {Iorgulescu}, {ATC}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)The average load is far
2021-07-16 10:14:01
150
原创 Morpheus: Towards Automated SLOs for Enterprise Clusters
{Title}（文章标题）{2016}, {Sangeeth Abdu Jyothi}, {OSDI}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)Modern resource management frameworks for large-scale analytics leave unresolved the proble
2021-07-15 10:10:24
109
原创 SmartHarvest: Harvesting Idle CPUs Safely and Efficiently in the cloud
SmartHarvest: Harvesting Idle CPUs Safely and Efficiently in the cloud{2021}, {Yawen Wang}, {EuroSys}@inproceedings{wang2021smartharvest,title={SmartHarvest: harvesting idle CPUs safely and efficiently in the cloud},author={Wang, Yawen and Arya, Kapil
2021-07-14 10:04:13
366
原创 我的论文笔记框架（Markdown格式）
我的论文笔记框架（Markdown格式）{Title}（文章标题）{Year}, {Authors}, {Journal Name}{引用格式}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)作者的研究目标是什么？Background / Problem Statement研究的背景以及问题陈述：作者需要解决的问题是什么？
2021-07-14 09:22:07
556
原创 paper reading:PARTIES:Qos-Aware Resource Partitioning for Multiple Interative Services
PARTIES:Qos-Aware Resource Partitioning for Multiple Interative Services2019, Shuang Chen,Chritistia Delimitrou, ASPOLOS{引用格式}Summary写完笔记之后最后填，概述文章的内容，以后查阅笔记的时候先看这一段。注：写文章summary切记需要通过自己的思考，用自己的语言描述。忌讳直接Ctrl + c原文。Research Objective(s)We present PAR
2021-07-12 15:53:09
477
原创 Anaconda 常用命令
conda -Vconda infoconda list --revisionsconda install geopandasconda uninstall protobufconda config --show default_channelsconda config --show channelsconda config --set channel_priority flexibleconda config --set report_errors falseconda .
2021-05-28 16:21:14
189
原创 Class Inheritance
Inheritance: An Is-a RelationshipA banana is a fruit, but a lunch has a banana.The Need for Virtual DestructorsIf the destructors are not not virtual, then just the destructor corresponding to the pointer type is called.class Brass{virtual ~Brass(){}
2020-12-17 20:18:29
228
原创 C++ Runtime Type Identification(RTTI)
What is RTTI for?The intent of RTTI is to provide a standard way for a program to determine the type of object during runtime.How Does RTTI Work?The dynamic_cast operator generates a pointer to a derived type from a pointer to a base type, if possible.
2020-12-15 20:53:06
142
原创 STL Iterators
Understanding iterator is the key to understanding the STL. Js as templates make algorithms independent of the type of data stored, iterators make the algorithms independent of the container used.The begin() returns an iterator to the first element, and t
2020-12-10 11:40:34
264
1
原创 Rust-Fearless Concurrency -Thread
use std::thread;fn main(){let v = vec![1,2,3];let handle = thread::spawn(move || { println!("Here's a vector: {:?}",v);});drop(v); // oh no!handle.join().unwrap();}using std::sync::mpsc;use std::thread;fn main(){ let (tx,rx) = mpsc:
2020-12-04 11:08:02
131
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人