PFS存储引擎10/17/22 2 Why ●为了减少使用cpu做内存copy，减少系统调用 ●发挥某些被操作系统屏蔽的功能，例如nvme write zero ●根据阿里《When Cloud Storage Meets RDMA》的说法 ●在100Gbps网络带宽时，内存带宽成为瓶颈 ●Intel Memory Latency Checker (MLC)测试得到的CPU内存带宽是 61Gbps10/17/22 const struct iovec *iov, int iovcnt, off_t offset); ●IO vector的接口主要是为了与brpc的iobuf对接，iobuf由若干地址不连 续的block组成，一次IO提交可以提高效率。10/17/22 8 PFS+SPDK 的部分读写的实现 ●某些盘只支持4k单位读写，但是CurveBS支持512字节读写 ●可能存在部分写的并发冲突 ●引入并发的range

integrate to Cloud NativeCurve intro • Curve is an distributed storage system • High performance • Easiser operation • Cloud native • Curve is composed of two parts • Curve Block Storage (CurveBS) high performance cloud native distributed block storage • Curve File System (CurveFS) • CurveFS: a high performance cloud native file system (Based on CurveBS / S3 compatible Storage)Operator capability (by Helm) patch and minor version upgrads supported FULL LIFECYCLE Plan to Support app lifecycle, storage lifecycle(backup, failure, recovery) DEEP INSIGHTS Plan to Support metrics, alerts, log processing

LINE’s Private Cloud Yutaro Hayakawa October 28, 2020 • Messaging & many family services • 185 million global MAU • 3Tbps+ network traffic in total LINE Verda: LINE’s Private Cloud Service IaaS LB LB NAT … PaaS FaaS … Verda and XDP Based L4 Load Balancer Service • Part of our private cloud service since 2017 • 5100 private, 760 public VIPs • k8s CCM integration (Type: LoadBalancer) L4LB

杭州飞致云信息科技有限公司 软件用起来才有价值，才有改进的机会 1 FIT2CLOUD CloudExplorer 云管理平台产品白皮书 版本(v1.6) 杭州飞致云信息科技有限公司 2022 年 5 月 杭州飞致云信息科技有限公司 软件用起来才有价值，才有改进的机会 2 版权申明 版权所有©杭州飞致云信息科技有限公司 2022。保留一切权利。 非经本公司书面许可，任何单 非经本公司书面许可，任何单位和个人不得擅自摘抄、复制本文档内容的部分或全部。 商标说明 FIT2CLOUD 商标、飞致云商标均为杭州飞致云信息科技有限公司的商标。 本文档提 及的其他所有商标或注册商标，由各自的所有人拥有。 注意 您购买的产品、服务或特性等受杭州飞致云信息科技有限公司商业合同和条款的约束， 本文档中描述的全部或部分产品、服务或特性可能不在您的购买或使用范围之内。除非合同 另有约定，杭州飞致云信 所有陈述、信息和建议不构成任何明示或暗示的担保。 公司简介 FIT2CLOUD 飞致云创立于 2014 年，是多云时代技术领先的企业级软件提供商。 FIT2CLOUD 以“Fit to Cloud”为使命，致力于帮助企业在多云时代更好地测试、运行、 管理和保护其 IT 基础设施和应用，实现数字化转型的目标。 FIT2CLOUD 的产品与解决方案涵盖软件测试、云原生运行时、多云管理、安全合规、

technical scalability to your datacentre, public or private. Whether you want to deploy an OpenStack cloud, a Kubernetes cluster or a 50,000-node render farm, Ubuntu Server delivers the best value scale-out the server installer 3 install/step-by-step Screen-by-screen installer guide 3 install/storage Configuring storage 2 Next steps 3 install/subscription Attach your Ubuntu Pro subscription 3 install/reporting-problems backups-introduction Backups 0 1 reference Reference 2 Cloud Images 3 cloud-images/introduction Introduction 3 cloud-images/amazon-ec2 Amazon EC2 3 cloud-images/google-cloud-engine Google Compute Engine (GCE) 3 f

................. 344 3. Cloud images and uvtool ......................................................................................................... 346 4. Ubuntu Cloud ......................... writing it only supports amd64 processors and does not support LVM or RAID or other more sophisticated storage options, nor does it support reusing existing partitions on the disks of the system you are installing installer offers the choice to install the system as a vanilla Ubuntu server, a MAAS5 bare-metal cloud rack controller or a MAAS6 region controller. If you select one of the MAAS options you will be asked

dashboard-style overview • reports • high-level (business) view of monitored resources Historical data storage 6 • data stored in a database • configurable history • built-in housekeeping procedure Easy central repository in which all configuration, statistical and operational data are stored. Database storage All configuration information as well as the data gathered by Zabbix is stored in a database. Web differentiate hosts when using active agent auto-registration. In some cases (for example, Amazon cloud nodes) it would be great to keep the original hostname while also use other information sent by the

dashboard-style overview • reports • high-level (business) view of monitored resources Historical data storage 5 • data stored in a database • configurable history • built-in housekeeping procedure Easy central repository in which all configuration, statistical and operational data are stored. Database storage All configuration information as well as the data gathered by Zabbix is stored in a database. Web zabbix_server and zabbix_proxy configuration now also applies to Post- greSQL databases in addition to IBM DB2. If this parameter is left undefinied, a default of public is used as the name of the DBSchema

dashboard-style overview • reports • high-level (business) view of monitored resources Historical data storage 5 • data stored in a database • configurable history • built-in housekeeping procedure Easy central repository in which all configuration, statistical and operational data are stored. Database storage All configuration information as well as the data gathered by Zabbix is stored in a database. Web (Brazil) 6.2 Daemon improvements 6.2.1 VMware monitoring optimisations The VMware data processing and storage was changed, resulting in less time VMware cache is locked and also reducing the size of cached

health and integrity of servers, virtual machines, applications, services, databases, websites, the cloud and more. Zabbix uses a flexible notification mechanism that allows users to configure e-mail based dashboard-style overview • reports • high-level (business) view of monitored resources Historical data storage 6 • data stored in a database • configurable history • built-in housekeeping procedure Easy central repository in which all configuration, statistical and operational data are stored. Database storage All configuration information as well as the data gathered by Zabbix is stored in a database. Web