PERFORMANCE MATTERS (joint work with Charlie Curtsinger, Grinnell College) emeryberger.com, @emeryberger Emery Berger College of Information and Computer Sciences UMASS AMHERSTA short time ago : un.bmp Ogle is too slow! OGLE’84 is too slow!Transistors (millions) Clock Speed (MHz) Performance used to be easy 0.001 0.01 0.1 1 10 100 1,000 10,000 1970 1975 1980 1985 1990 1995 gle loading… No mojitos for me… Back to the present…Transistors (millions) Clock Speed (MHz) Performance not easy anymore 0.001 0.01 0.1 1 10 100 1,000 10,000 1970 1975 1980 1985 1990 1995

Performance of Apache Ozone on NVMe Wei-Chiu Chuang (jojochuang) Ritesh Shukla (kerneltime) Agenda • Overview of how Ozone and how it scales • Why NVME is important for Ozone for scaling • Benefits Benefits of using NVME • Impala performance results from NVME clusters • Write path improvements results from NVME clusters • Summary • Questions Ozone Architecture Why does Ozone Scale? Separation of HDFS Storage Density 1000’s of nodes at 600TB per node 1000’s of nodes at 100TB per node Scalability 10B Objects 400M Objects Recovery Fast recovery ( < 5 min restart) Slow startup based on size

09.Services Web 19 décembre 2023 Développement web il3 Services web HE-Arc (DGR) 2022 Applications distribuées • Motivation : répartir l’exécution sur plusieurs machines – Principe : Les composants/services abstraction différences – Exemples : RPC, RMI (java), CORBA, DCOM (MS) • Utiliser les technologies du web, comme HTTP et XML : – indépendantes de la plateforme, éprouvées, largement utilisées • Système distribué rvices 1 Service web • 2 visions : – Utiliser les technos web pour développer des applis distribuées – Accès pour une application aux services offerts aux humains • Service web = webapp pour une autre

shared chat / drawing canvas – Game events © Joe Mertz – Mobile to Cloud: Building Distributed Applications • Workarounds have been devised • E.g. Polling – Client continuously polls the server • E the next time he wants to send a letter. © Joe Mertz – Mobile to Cloud: Building Distributed Applications • Provides for true two-way ongoing communication between a client and server. • Each side Some old browsers don't implement WebSockets © Joe Mertz – Mobile to Cloud: Building Distributed Applications // Create a new WebSocket var wSocket = new WebSocket("ws://www.example.com/socketserver") //

future of cloud native applications Mark Russinovich @markrussinovich Open Application Model dapr: Distributed Application Platform Building Cloud Scale, Hybrid Applications Agenda Application Models App developers need to know and code for each infrastructure they deploy to Kubernetes for applications "[Kubernetes] is really hard to get into it and understand how all the parts play together people." – Software Architect @ Crisp Kubernetes focuses on container infrastructure, not on applications Application developers need to be experts in Kubernetes APIs Production use of Kubernetes

1 Quickly Testing Qt Desktop Applications with Approval Tests Clare Macrae (She/her) clare@claremacrae.co.uk 16 September 2020 CppCon (Online)2 Audience: Developers testing Desktop GUIs, including and readable with fixtures • ApprovalTests.Cpp.Qt feedback welcome!80 Quickly Test Qt Desktop Applications • All links from this talk, and more, via: – bit.ly/TestingQt – github.com/claremacrae/talks

《Go Web 编程》 《Go Web 编程》 因为自己对Web开发比较感兴趣，所以最近抽空在写一本开源的书籍《Go Web编程》《Build Web Application with Golang》。写这本书不表示我能力很强，而是我愿意分享，和大家一起分享Go写Web应用的一些东西。 对于从PHP/Python/Ruby转过来的同学了解Go怎么写Web应用开发的 对于从C/C++转过来的同学了解Web到底是怎么运行起来的 对于从C/C++转过来的同学了解Web到底是怎么运行起来的 我一直认为知识是用来分享的，让更多的人分享自己拥有的一切知识这个才是人生最大的快乐。 这本书目前我放在Github上，我现在基本每天晚上抽空会写一些，时间有限、能力有限，所以希望更多的朋友参与到 这个开源项目中来。 撰写方法 撰写方法 文件命名 文件命名 每个章节建立一个md文件，如第11章的第3节，则建立11.3.md 11.3.md。 这样读者就可以把相应的Markdown文件编译成html文件，执行go build build.go，执行生成的文件，就会在底目 录下生成相应的html文件 交流 交流 欢迎大家加入QQ群：259316004 《Go Web编程》专用交流群 大家有问题还可以上德问上一起交流学习：http://www.dewen.org/topic/165 致谢 致谢 首先要感谢Golang-China的QQ群1023198

Being Friendly to Your Hardware Performance Engineering A gentle introduction to hardware for software engineers 2Where does C++ run? 3On an abstract C++ machine 4On an abstract C++ machine? In most practical cases at boot time only Same capacity, different composition => different performance profile From JESD 79-4 DDR4 specificationMemory • Memory system is in the uncore • Cores act Multiple instructions resulting in fewer operations • ISA restrictions may have impact to performance Imaginary ARM mov r20, 0x123456789abcdef0Register renaming 52 Branching Fetch Decode Queue

WHITEPAPER How GitOps Boosts Business Performance: The Facts How GitOps Boosts Business Performance: The Facts 2 INTRODUCTION As cloud-native applications have become more prevalent, the concept competitive advantage with an increase in innovation. This positive effect is not limited to the performance of engineering teams. Technology, in particular cloud native technology like Kubernetes and its together six years of data drawn from over 31,000 technology professionals worldwide. It charts the performance of engineering teams across the world against four key measures: lead time for new features, failure