Submit Search
Upload
Using CDN to improve performance
•
Download as PPT, PDF
•
44 likes
•
19,788 views
G
Gea-Suan Lin
Follow
Technology
Slideshow view
Report
Share
Slideshow view
Report
Share
1 of 178
Download now
Recommended
"Kafka Connect is an ideal tool for building data pipelines. It is both reliable and scalable, with a pluggable interface that lets you flow data between Kafka and any system you need. A Connect pipeline is made up of many different components, and understanding how each of these interact together is essential, even for the simplest setup. In this talk we will introduce the Connect components, from connectors, to transformations to the runtime itself. We will also share some of the new capabilities and best practices that you should be aware of to help you run and manage connectors effectively. Finally we will talk about some different open source projects that have been built on top of Connect that can help you get the most out of the framework."
Getting up to speed with Kafka Connect: from the basics to the latest feature...
Getting up to speed with Kafka Connect: from the basics to the latest feature...
HostedbyConfluent
Data is a critical infrastructure for building machine learning systems. From ensuring accurate ETAs to predicting optimal traffic routes, providing safe, seamless transportation and delivery experiences on the Uber platform requires reliable, performant large-scale data storage and analysis. In 2016, Uber developed Apache Hudi, an incremental processing framework, to power business critical data pipelines at low latency and high efficiency, and helps distributed organizations build and manage petabyte-scale data lakes. In this talk, I will describe what is APache Hudi and its architectural design, and then deep dive to improving data operations by providing features such as data versioning, time travel. We will also go over how Hudi brings kappa architecture to big data systems and enables efficient incremental processing for near real time use cases. Speaker: Satish Kotha (Uber) Apache Hudi committer and Engineer at Uber. Previously, he worked on building real time distributed storage systems like Twitter MetricsDB and BlobStore. website: https://www.aicamp.ai/event/eventdetails/W2021043010
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudi
Bill Liu
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022 Apache Kafka without Zookeeper is now production ready! This talk is about how you can run without ZooKeeper, and why you should.
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022
HostedbyConfluent
This talk covers Kafka cluster sizing, instance type selections, scaling operations, replication throttling and more. Don’t forget to check out the Kafka-Kit repository. https://www.youtube.com/watch?time_continue=2613&v=7uN-Vlf7W5E
Thoughts on kafka capacity planning
Thoughts on kafka capacity planning
JamieAlquiza
SF Bay Area ClickHouse Meetup talk on integrating GraphQL with ClickHouse by Aleksey Studnev of Bitquery.io
Bitquery GraphQL for Analytics on ClickHouse
Bitquery GraphQL for Analytics on ClickHouse
Altinity Ltd
In this session, we walk through the Amazon VPC network presentation and describe the problems we were trying to solve when we created it. Next, we walk through how these problems are traditionally solved, and why those solutions are not scalable, inexpensive, or secure enough for AWS. Finally, we provide an overview of the solution that we've implemented and discuss some of the unique mechanisms that we use to ensure customer isolation, get packets into and out of the network, and support new features like VPC endpoints.
(NET403) Another Day, Another Billion Packets
(NET403) Another Day, Another Billion Packets
Amazon Web Services
2017/06/30 - 07/01 にかけて開催された,db analytics_show_case の講演資料です.
クラウド上のデータ活用デザインパターン
クラウド上のデータ活用デザインパターン
Amazon Web Services Japan
Apache Hudi is an open data lake platform, designed around the streaming data model. At its core, Hudi provides a transactions, upserts, deletes on data lake storage, while also enabling CDC capabilities. Hudi also provides a coherent set of table services, which can clean, compact, cluster and optimize storage layout for better query performance. Finally, Hudi's data services provide out-of-box support for streaming data from event systems into lake storage in near real-time. In this talk, we will walk through an end-end use case for change data capture from a relational database, starting with capture changes using the Pulsar CDC connector and then demonstrate how you can use the Hudi deltastreamer tool to then apply these changes into a table on the data lake. We will discuss various tips to operationalizing and monitoring such pipelines. We will conclude with some guidance on future integrations between the two projects including a native Hudi/Pulsar connector and Hudi tiered storage.
Change Data Capture to Data Lakes Using Apache Pulsar and Apache Hudi - Pulsa...
Change Data Capture to Data Lakes Using Apache Pulsar and Apache Hudi - Pulsa...
StreamNative
Recommended
"Kafka Connect is an ideal tool for building data pipelines. It is both reliable and scalable, with a pluggable interface that lets you flow data between Kafka and any system you need. A Connect pipeline is made up of many different components, and understanding how each of these interact together is essential, even for the simplest setup. In this talk we will introduce the Connect components, from connectors, to transformations to the runtime itself. We will also share some of the new capabilities and best practices that you should be aware of to help you run and manage connectors effectively. Finally we will talk about some different open source projects that have been built on top of Connect that can help you get the most out of the framework."
Getting up to speed with Kafka Connect: from the basics to the latest feature...
Getting up to speed with Kafka Connect: from the basics to the latest feature...
HostedbyConfluent
Data is a critical infrastructure for building machine learning systems. From ensuring accurate ETAs to predicting optimal traffic routes, providing safe, seamless transportation and delivery experiences on the Uber platform requires reliable, performant large-scale data storage and analysis. In 2016, Uber developed Apache Hudi, an incremental processing framework, to power business critical data pipelines at low latency and high efficiency, and helps distributed organizations build and manage petabyte-scale data lakes. In this talk, I will describe what is APache Hudi and its architectural design, and then deep dive to improving data operations by providing features such as data versioning, time travel. We will also go over how Hudi brings kappa architecture to big data systems and enables efficient incremental processing for near real time use cases. Speaker: Satish Kotha (Uber) Apache Hudi committer and Engineer at Uber. Previously, he worked on building real time distributed storage systems like Twitter MetricsDB and BlobStore. website: https://www.aicamp.ai/event/eventdetails/W2021043010
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudi
Bill Liu
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022 Apache Kafka without Zookeeper is now production ready! This talk is about how you can run without ZooKeeper, and why you should.
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022
Introducing KRaft: Kafka Without Zookeeper With Colin McCabe | Current 2022
HostedbyConfluent
This talk covers Kafka cluster sizing, instance type selections, scaling operations, replication throttling and more. Don’t forget to check out the Kafka-Kit repository. https://www.youtube.com/watch?time_continue=2613&v=7uN-Vlf7W5E
Thoughts on kafka capacity planning
Thoughts on kafka capacity planning
JamieAlquiza
SF Bay Area ClickHouse Meetup talk on integrating GraphQL with ClickHouse by Aleksey Studnev of Bitquery.io
Bitquery GraphQL for Analytics on ClickHouse
Bitquery GraphQL for Analytics on ClickHouse
Altinity Ltd
In this session, we walk through the Amazon VPC network presentation and describe the problems we were trying to solve when we created it. Next, we walk through how these problems are traditionally solved, and why those solutions are not scalable, inexpensive, or secure enough for AWS. Finally, we provide an overview of the solution that we've implemented and discuss some of the unique mechanisms that we use to ensure customer isolation, get packets into and out of the network, and support new features like VPC endpoints.
(NET403) Another Day, Another Billion Packets
(NET403) Another Day, Another Billion Packets
Amazon Web Services
2017/06/30 - 07/01 にかけて開催された,db analytics_show_case の講演資料です.
クラウド上のデータ活用デザインパターン
クラウド上のデータ活用デザインパターン
Amazon Web Services Japan
Apache Hudi is an open data lake platform, designed around the streaming data model. At its core, Hudi provides a transactions, upserts, deletes on data lake storage, while also enabling CDC capabilities. Hudi also provides a coherent set of table services, which can clean, compact, cluster and optimize storage layout for better query performance. Finally, Hudi's data services provide out-of-box support for streaming data from event systems into lake storage in near real-time. In this talk, we will walk through an end-end use case for change data capture from a relational database, starting with capture changes using the Pulsar CDC connector and then demonstrate how you can use the Hudi deltastreamer tool to then apply these changes into a table on the data lake. We will discuss various tips to operationalizing and monitoring such pipelines. We will conclude with some guidance on future integrations between the two projects including a native Hudi/Pulsar connector and Hudi tiered storage.
Change Data Capture to Data Lakes Using Apache Pulsar and Apache Hudi - Pulsa...
Change Data Capture to Data Lakes Using Apache Pulsar and Apache Hudi - Pulsa...
StreamNative
Jupil Hwang Senior Solutions Engineer Confluent
ksqlDB로 시작하는 스트림 프로세싱
ksqlDB로 시작하는 스트림 프로세싱
confluent
This session will focus on how leveraging Fargate and its serverless approach to deploying and managing containers will help increase operational efficiencies and reduce the time to ramp up your operations to run production containerized workloads. Datree will share their journey to adopt containers and the steps they were able to accelerate and avoid by using Fargate as well do a demo.
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Amazon Web Services
새벽 배송과 함께 신선한 먹거리를 제공하는 핫 스타트업인 Kurly에서는 급증하는 서비스 요구사항을 충족하기 위하여, AWS 를 적극적으로 활용하고 있습니다. Kurly가 진행하였던, AWS 상에서의 안정적인 서비스 운영 구축 경험과 최신 인공지능/기계학습 기술 활용을 통한, Next Generation Retail의 미래를 보여드리고자 합니다.
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Amazon Web Services Korea
February 2015 Hive User Group meetup at LinkedIn http://www.meetup.com/Hive-User-Group-Meeting/events/219794523/ Presentation about physical join strategies employed used by Apache Hive and how they may be employed to optimize workflows.
Hive join optimizations
Hive join optimizations
Szehon Ho
msr_以前のアーキテクチャ
msr_以前のアーキテクチャ
msr_以前のアーキテクチャ
default Takakuni
Organizations need to perform increasingly complex analysis on data — streaming analytics, ad-hoc querying, and predictive analytics — in order to get better customer insights and actionable business intelligence. Apache Spark has recently emerged as the framework of choice to address many of these challenges. In this session, we show you how to use Apache Spark on AWS to implement and scale common big data use cases such as real-time data processing, interactive data science, predictive analytics, and more. We will talk about common architectures, best practices to quickly create Spark clusters using Amazon EMR, and ways to integrate Spark with other big data services in AWS. Learning Objectives: • Learn why Spark is great for ad-hoc interactive analysis and real-time stream processing. • How to deploy and tune scalable clusters running Spark on Amazon EMR. • How to use EMR File System (EMRFS) with Spark to query data directly in Amazon S3. • Common architectures to leverage Spark with Amazon DynamoDB, Amazon Redshift, Amazon Kinesis, and more.
Best Practices for Using Apache Spark on AWS
Best Practices for Using Apache Spark on AWS
Amazon Web Services
Qlik sense- Technical Seminar
Qlik sense- Technical Seminar
Sanjana Gondane
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 김용우 솔루션즈 아키텍트, AWS 기업 고객들이 클라우드를 도입할 때 자사의 데이터 센터와 클라우드상의 자원들과의 안전한 연동은 매우 중요합니다. AWS Direct Connect는 일관된 네트워크 성능을 기반으로 AWS자원을 보다 빠르고 안전하게 사용하실수 있는 방법을 제공합니다. 본 세션에서는 AWS Direct Connect(DX)와 VPN의 특징 및 구성 방법을 다루고, 하이브리드 환경의 아키텍쳐를 위한 DX 서비스 신청부터 구성까지의 모든 단계를 살펴봅니다. 특히 가용성을 높이기 위한 DX 회선 이중화 디자인 및 VPN을 이용한 DX 회선 Backup 방법 그리고 Direct Connect gateway 를 통한 글로벌 서비스 네트워크 구성 방법 까지 다양하게 살펴보도록 하겠습니다.
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
Amazon Web Services Korea
2016년 4월 27일 DB Day 에서 김기완 솔루션즈 아키텍트 께서 발표하신 “Amazon Aurora Technical Deep Dive “ 발표자료입니다.
Amazon Aurora Deep Dive (김기완) - AWS DB Day
Amazon Aurora Deep Dive (김기완) - AWS DB Day
Amazon Web Services Korea
2018/01/13に開催された「第16回 クラウド女子会 〜新年だよ!2018年を貴女はどんな年にしたい?〜」に登壇させて頂いたときの資料です。 ※公開にあたり一部加筆修正しています。
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
Mamoru Ohashi
Mark Teehan, Principal Solutions Engineer, Confluent Use the Debezium CDC connector to capture database changes from a Postgres database - or MySQL or Oracle; streaming into Kafka topics and onwards to an external data store. Examine how to setup this pipeline using Docker Compose and Confluent Cloud; and how to use various payload formats, such as avro, protobuf and json-schema. https://www.meetup.com/Singapore-Kafka-Meetup/events/276822852/
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
confluent
Ⅰ. Percona Server for MySQL 1. Introduction 2. Storage Engine 3. Percona Server’s History 4. DBMS Comparison 5. Benchmark Ⅱ. Percona Server 5.7 1. New feature 2. Performance 3. Scalability 4. Flexibility 5. Reliability 6. Management 7. Diagnostics Ⅲ. Percona Server 구축사례 1. Reference
Percona server for MySQL 제품 소개
Percona server for MySQL 제품 소개
NeoClova
2018/02/13 開催分 動画配信 on AWS
AWS Black Belt Online Seminar 2018 動画配信 on AWS
AWS Black Belt Online Seminar 2018 動画配信 on AWS
Amazon Web Services Japan
Hadoop Summit 2105
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
DataWorks Summit
Linux, Windows環境での基幹系システム実現に最適なミッションクリティカルサーバー、"HPE Integrity Suprdome X" の紹介資料です。
HPE Superdome X ご紹介資料
HPE Superdome X ご紹介資料
日本ヒューレット・パッカード株式会社
Apache Hadoop? need to choose Hive or Imapla? Understand the differences?
Hive vs. Impala
Hive vs. Impala
Omid Vahdaty
Integrating Apache Kafka with other systems in a reliable and scalable way is often a key part of a streaming platform. Fortunately, Apache Kafka includes the Connect API that enables streaming integration both in and out of Kafka. Like any technology, understanding its architecture and deployment patterns is key to successful use, as is knowing where to go looking when things aren’t working.
From Zero to Hero with Kafka Connect
From Zero to Hero with Kafka Connect
Databricks
En esta presentacion vemos los conceptos de Apache Spark par análisis de datos.
Análisis de datos con Apache Spark
Análisis de datos con Apache Spark
Eduardo Castro
Learn best practices for building a real-time streaming data architecture on AWS with Spark Streaming, Amazon Kinesis, and Amazon Elastic MapReduce (EMR). Get a closer look at how to ingest streaming data scalably and durably from data producers like mobile devices, servers, and even web browsers, and design a stream processing application with minimal data duplication and exactly-once processing. Presented by: Guy Ernest, Principal Business Development Manager, Amazon Web Services Customer Guest: Harry Koch, Solutions Architecture, Philips
Real-Time Streaming Data on AWS
Real-Time Streaming Data on AWS
Amazon Web Services
Performance troubleshooting of distributed data processing systems is a complex task. Apache Spark comes to rescue with a large set of metrics and instrumentation that you can use to understand and improve the performance of your Spark-based applications. You will learn about the available metric-based instrumentation in Apache Spark: executor task metrics and the Dropwizard-based metrics system. The talk will cover how Hadoop and Spark service at CERN is using Apache Spark metrics for troubleshooting performance and measuring production workloads. Notably, the talk will cover how to deploy a performance dashboard for Spark workloads and will cover the use of sparkMeasure, a tool based on the Spark Listener interface. The speaker will discuss the lessons learned so far and what improvements you can expect in this area in Apache Spark 3.0.
Performance Troubleshooting Using Apache Spark Metrics
Performance Troubleshooting Using Apache Spark Metrics
Databricks
In MicroMarketMonitor’s recent report, it was noted that the North American content delivery network market is expected to grow from $1.95 billion in 2013 to $7.83 billion in 2019. One significant factor driving this growth is end user interaction with online content. The interaction between a user and online content is far more complex today than it was a few years ago. Today’s users are much more likely to be streaming a longer video from a mobile phone or accessing a SaaS portal when working from home. These are far more complex experiences that did not exist five or so years ago. Given the expected growth of the CDN market in the coming years, this post will define exactly what a CDN is and how it operates. A content delivery network, also called a CDN, improves the performance, security and reliability of a website. Since the start of the internet, websites have evolved in how they deliver content to the end user. In the example below, you will see the CDNetworks’ Twitter stream. Highlighted in red, is what is known as static content. The CDNetworks logo and profile descriptions are basic graphics and HTML text that rarely change. Highlighted in yellow, is what is known as dynamic content. The Twitter stream represents content that is always changing and moving.
How Content Delivery Networks Work
How Content Delivery Networks Work
CDNetworks
Ahmet Ozalp is the VP of International Products and Strategy for the Media (Sola) Division of Akamai. In this role Ozalp leads the product management and strategy initiatives for Akamai’s video, software and application delivery, storage and analytics products in EMEA and Asia Pacific regions. He also leads and supports key partnership initiatives with major Telecommunications companies in the region. Prior to Akamai, Ozalp was the CEO of Telenity, a global provider of mobile advertising and social network solutions. Prior to Telenity, Ozalp was a Partner at Atlas Venture, where he led investments in digital media, mobile and online advertising. While at Atlas, Ozalp was the lead investor and board member of Extend Media (acquired by CSCO) and also had board level involvement with Ellacoya Networks (acquired by Arbor Networks), Isilon (IPO, Nasdaq:ISLN) and Gotuit Media. Before joining Atlas, Ozalp held executive positions in marketing and product management at Narad Networks (acquired by Nasdaq:CIEN) and was also part of the founding team of Newnet, a telecom software startup acquired by ADC (Tyco Electronics). In the early part of his career, Ozalp was also a management consultant with Bain & Company’s technology and media practice. Ozalp holds an MBA from the Wharton School of University of Pennsylvania, an MS in EE from Columbia University and two US patents.
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
IDATE DigiWorld
More Related Content
What's hot
Jupil Hwang Senior Solutions Engineer Confluent
ksqlDB로 시작하는 스트림 프로세싱
ksqlDB로 시작하는 스트림 프로세싱
confluent
This session will focus on how leveraging Fargate and its serverless approach to deploying and managing containers will help increase operational efficiencies and reduce the time to ramp up your operations to run production containerized workloads. Datree will share their journey to adopt containers and the steps they were able to accelerate and avoid by using Fargate as well do a demo.
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Amazon Web Services
새벽 배송과 함께 신선한 먹거리를 제공하는 핫 스타트업인 Kurly에서는 급증하는 서비스 요구사항을 충족하기 위하여, AWS 를 적극적으로 활용하고 있습니다. Kurly가 진행하였던, AWS 상에서의 안정적인 서비스 운영 구축 경험과 최신 인공지능/기계학습 기술 활용을 통한, Next Generation Retail의 미래를 보여드리고자 합니다.
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Amazon Web Services Korea
February 2015 Hive User Group meetup at LinkedIn http://www.meetup.com/Hive-User-Group-Meeting/events/219794523/ Presentation about physical join strategies employed used by Apache Hive and how they may be employed to optimize workflows.
Hive join optimizations
Hive join optimizations
Szehon Ho
msr_以前のアーキテクチャ
msr_以前のアーキテクチャ
msr_以前のアーキテクチャ
default Takakuni
Organizations need to perform increasingly complex analysis on data — streaming analytics, ad-hoc querying, and predictive analytics — in order to get better customer insights and actionable business intelligence. Apache Spark has recently emerged as the framework of choice to address many of these challenges. In this session, we show you how to use Apache Spark on AWS to implement and scale common big data use cases such as real-time data processing, interactive data science, predictive analytics, and more. We will talk about common architectures, best practices to quickly create Spark clusters using Amazon EMR, and ways to integrate Spark with other big data services in AWS. Learning Objectives: • Learn why Spark is great for ad-hoc interactive analysis and real-time stream processing. • How to deploy and tune scalable clusters running Spark on Amazon EMR. • How to use EMR File System (EMRFS) with Spark to query data directly in Amazon S3. • Common architectures to leverage Spark with Amazon DynamoDB, Amazon Redshift, Amazon Kinesis, and more.
Best Practices for Using Apache Spark on AWS
Best Practices for Using Apache Spark on AWS
Amazon Web Services
Qlik sense- Technical Seminar
Qlik sense- Technical Seminar
Sanjana Gondane
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 김용우 솔루션즈 아키텍트, AWS 기업 고객들이 클라우드를 도입할 때 자사의 데이터 센터와 클라우드상의 자원들과의 안전한 연동은 매우 중요합니다. AWS Direct Connect는 일관된 네트워크 성능을 기반으로 AWS자원을 보다 빠르고 안전하게 사용하실수 있는 방법을 제공합니다. 본 세션에서는 AWS Direct Connect(DX)와 VPN의 특징 및 구성 방법을 다루고, 하이브리드 환경의 아키텍쳐를 위한 DX 서비스 신청부터 구성까지의 모든 단계를 살펴봅니다. 특히 가용성을 높이기 위한 DX 회선 이중화 디자인 및 VPN을 이용한 DX 회선 Backup 방법 그리고 Direct Connect gateway 를 통한 글로벌 서비스 네트워크 구성 방법 까지 다양하게 살펴보도록 하겠습니다.
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
Amazon Web Services Korea
2016년 4월 27일 DB Day 에서 김기완 솔루션즈 아키텍트 께서 발표하신 “Amazon Aurora Technical Deep Dive “ 발표자료입니다.
Amazon Aurora Deep Dive (김기완) - AWS DB Day
Amazon Aurora Deep Dive (김기완) - AWS DB Day
Amazon Web Services Korea
2018/01/13に開催された「第16回 クラウド女子会 〜新年だよ!2018年を貴女はどんな年にしたい?〜」に登壇させて頂いたときの資料です。 ※公開にあたり一部加筆修正しています。
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
Mamoru Ohashi
Mark Teehan, Principal Solutions Engineer, Confluent Use the Debezium CDC connector to capture database changes from a Postgres database - or MySQL or Oracle; streaming into Kafka topics and onwards to an external data store. Examine how to setup this pipeline using Docker Compose and Confluent Cloud; and how to use various payload formats, such as avro, protobuf and json-schema. https://www.meetup.com/Singapore-Kafka-Meetup/events/276822852/
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
confluent
Ⅰ. Percona Server for MySQL 1. Introduction 2. Storage Engine 3. Percona Server’s History 4. DBMS Comparison 5. Benchmark Ⅱ. Percona Server 5.7 1. New feature 2. Performance 3. Scalability 4. Flexibility 5. Reliability 6. Management 7. Diagnostics Ⅲ. Percona Server 구축사례 1. Reference
Percona server for MySQL 제품 소개
Percona server for MySQL 제품 소개
NeoClova
2018/02/13 開催分 動画配信 on AWS
AWS Black Belt Online Seminar 2018 動画配信 on AWS
AWS Black Belt Online Seminar 2018 動画配信 on AWS
Amazon Web Services Japan
Hadoop Summit 2105
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
DataWorks Summit
Linux, Windows環境での基幹系システム実現に最適なミッションクリティカルサーバー、"HPE Integrity Suprdome X" の紹介資料です。
HPE Superdome X ご紹介資料
HPE Superdome X ご紹介資料
日本ヒューレット・パッカード株式会社
Apache Hadoop? need to choose Hive or Imapla? Understand the differences?
Hive vs. Impala
Hive vs. Impala
Omid Vahdaty
Integrating Apache Kafka with other systems in a reliable and scalable way is often a key part of a streaming platform. Fortunately, Apache Kafka includes the Connect API that enables streaming integration both in and out of Kafka. Like any technology, understanding its architecture and deployment patterns is key to successful use, as is knowing where to go looking when things aren’t working.
From Zero to Hero with Kafka Connect
From Zero to Hero with Kafka Connect
Databricks
En esta presentacion vemos los conceptos de Apache Spark par análisis de datos.
Análisis de datos con Apache Spark
Análisis de datos con Apache Spark
Eduardo Castro
Learn best practices for building a real-time streaming data architecture on AWS with Spark Streaming, Amazon Kinesis, and Amazon Elastic MapReduce (EMR). Get a closer look at how to ingest streaming data scalably and durably from data producers like mobile devices, servers, and even web browsers, and design a stream processing application with minimal data duplication and exactly-once processing. Presented by: Guy Ernest, Principal Business Development Manager, Amazon Web Services Customer Guest: Harry Koch, Solutions Architecture, Philips
Real-Time Streaming Data on AWS
Real-Time Streaming Data on AWS
Amazon Web Services
Performance troubleshooting of distributed data processing systems is a complex task. Apache Spark comes to rescue with a large set of metrics and instrumentation that you can use to understand and improve the performance of your Spark-based applications. You will learn about the available metric-based instrumentation in Apache Spark: executor task metrics and the Dropwizard-based metrics system. The talk will cover how Hadoop and Spark service at CERN is using Apache Spark metrics for troubleshooting performance and measuring production workloads. Notably, the talk will cover how to deploy a performance dashboard for Spark workloads and will cover the use of sparkMeasure, a tool based on the Spark Listener interface. The speaker will discuss the lessons learned so far and what improvements you can expect in this area in Apache Spark 3.0.
Performance Troubleshooting Using Apache Spark Metrics
Performance Troubleshooting Using Apache Spark Metrics
Databricks
What's hot
(20)
ksqlDB로 시작하는 스트림 프로세싱
ksqlDB로 시작하는 스트림 프로세싱
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Operational Excellence with Containerized Workloads Using AWS Fargate (CON320...
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Kurly는 AWS를 어떻게 사용하고 있을까? - 성공적 리테일 디지털 트랜스포메이션 사례 - 박경표 AWS 솔루션즈 아키텍트 / 임상석...
Hive join optimizations
Hive join optimizations
msr_以前のアーキテクチャ
msr_以前のアーキテクチャ
Best Practices for Using Apache Spark on AWS
Best Practices for Using Apache Spark on AWS
Qlik sense- Technical Seminar
Qlik sense- Technical Seminar
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
AWS Direct Connect 를 통한 하이브리드 클라우드 아키텍쳐 설계 - 김용우 솔루션즈 아키텍트, AWS :: AWS Summit...
Amazon Aurora Deep Dive (김기완) - AWS DB Day
Amazon Aurora Deep Dive (김기완) - AWS DB Day
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
ぼくらのアカウント戦略〜マルチアカウントでのガバナンスと権限管理の全て〜
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
From Postgres to Event-Driven: using docker-compose to build CDC pipelines in...
Percona server for MySQL 제품 소개
Percona server for MySQL 제품 소개
AWS Black Belt Online Seminar 2018 動画配信 on AWS
AWS Black Belt Online Seminar 2018 動画配信 on AWS
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
Unify Stream and Batch Processing using Dataflow, a Portable Programmable Mod...
HPE Superdome X ご紹介資料
HPE Superdome X ご紹介資料
Hive vs. Impala
Hive vs. Impala
From Zero to Hero with Kafka Connect
From Zero to Hero with Kafka Connect
Análisis de datos con Apache Spark
Análisis de datos con Apache Spark
Real-Time Streaming Data on AWS
Real-Time Streaming Data on AWS
Performance Troubleshooting Using Apache Spark Metrics
Performance Troubleshooting Using Apache Spark Metrics
Viewers also liked
In MicroMarketMonitor’s recent report, it was noted that the North American content delivery network market is expected to grow from $1.95 billion in 2013 to $7.83 billion in 2019. One significant factor driving this growth is end user interaction with online content. The interaction between a user and online content is far more complex today than it was a few years ago. Today’s users are much more likely to be streaming a longer video from a mobile phone or accessing a SaaS portal when working from home. These are far more complex experiences that did not exist five or so years ago. Given the expected growth of the CDN market in the coming years, this post will define exactly what a CDN is and how it operates. A content delivery network, also called a CDN, improves the performance, security and reliability of a website. Since the start of the internet, websites have evolved in how they deliver content to the end user. In the example below, you will see the CDNetworks’ Twitter stream. Highlighted in red, is what is known as static content. The CDNetworks logo and profile descriptions are basic graphics and HTML text that rarely change. Highlighted in yellow, is what is known as dynamic content. The Twitter stream represents content that is always changing and moving.
How Content Delivery Networks Work
How Content Delivery Networks Work
CDNetworks
Ahmet Ozalp is the VP of International Products and Strategy for the Media (Sola) Division of Akamai. In this role Ozalp leads the product management and strategy initiatives for Akamai’s video, software and application delivery, storage and analytics products in EMEA and Asia Pacific regions. He also leads and supports key partnership initiatives with major Telecommunications companies in the region. Prior to Akamai, Ozalp was the CEO of Telenity, a global provider of mobile advertising and social network solutions. Prior to Telenity, Ozalp was a Partner at Atlas Venture, where he led investments in digital media, mobile and online advertising. While at Atlas, Ozalp was the lead investor and board member of Extend Media (acquired by CSCO) and also had board level involvement with Ellacoya Networks (acquired by Arbor Networks), Isilon (IPO, Nasdaq:ISLN) and Gotuit Media. Before joining Atlas, Ozalp held executive positions in marketing and product management at Narad Networks (acquired by Nasdaq:CIEN) and was also part of the founding team of Newnet, a telecom software startup acquired by ADC (Tyco Electronics). In the early part of his career, Ozalp was also a management consultant with Bain & Company’s technology and media practice. Ozalp holds an MBA from the Wharton School of University of Pennsylvania, an MS in EE from Columbia University and two US patents.
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
IDATE DigiWorld
Explain What is CDN and why it is used . What will be its Future.
Content Delivery Network
Content Delivery Network
Shiv Pandey
Netflix CDN and Open Source
Netflix CDN and Open Source
Gleb Smirnoff
Content Delivery Networks are on the front lines of application performance. ThousandEyes provides deep insights into CDN performance, including geographic load balancing, latency and availability. We'll help you detect and diagnose issues with your CDN providers and track down problems at the origin or the edge. In these slides, we'll share how to: 1. Measure and baseline CDN performance. 2. Diagnose issues with specific files, caches or edge locations. 3. Share data with your CDN to resolve problems.
Monitoring CDN Performance
Monitoring CDN Performance
ThousandEyes
Content caching is one of the most effective ways to dramatically improve the performance of a web site. In this webinar, we’ll deep-dive into NGINX’s caching abilities and investigate the architecture used, debugging techniques and advanced configuration. By the end of the webinar, you’ll be well equipped to configure NGINX to cache content exactly as you need. View full webinar on demand at http://nginx.com/resources/webinars/content-caching-nginx/
NGINX High-performance Caching
NGINX High-performance Caching
NGINX, Inc.
Jifty by Audrey Tang in OSDC.tw 2007
Jifty
Jifty
Gea-Suan Lin
Angular 2 is the next version of Google’s massively popular MVC framework for building complex single page applications in the browser (and beyond).
Introduction to Angular2
Introduction to Angular2
Knoldus Inc.
CONTENT DELIVERY NETWORK
CONTENT DELIVERY NETWORK
CONTENT DELIVERY NETWORK
Saif Muttair
As digital consumption of rich media content explodes and with audience expectations at its peak, media providers have been challenged with not only delivering high-quality audience experiences but also the audience analytics in realtime to enable actionable insights for content publishers. Arkena, one of Europe’s leading media services organizations chose to power it’s analytical platform with Hortonworks Data Platform to cost effectively store and analyze over 3.5 terabytes of data per day. Join Hortonworks and Arkena as they share the industry challenges faced, the solution created which enables real-time and better analytics for their customers.
A Better Rich Media Experience & Video Analytics at Arkena with Apache Hadoop
A Better Rich Media Experience & Video Analytics at Arkena with Apache Hadoop
Reda Benzair
Simple thought about next generation of CDN
CDN 2.0
CDN 2.0
Jedi Kim
Introduction of China cache IDC Services 201611
China cache IDC Services 201611
China cache IDC Services 201611
adam cheng
Introduction to chinacache services
Introduction to chinacache services
Lillian shao
Nginx pronounced as "Engine X" is an open source high performance web and reverse proxy server which supports protocols like HTTP, HTTPS, SMTP, IMAP. It can also be used for load balancing and HTTP caching.
Introduction to Nginx
Introduction to Nginx
Knoldus Inc.
This presentation summarises the development of a Cisco ACI device package for NGINX as a Load-Balancer, made as a proof-of-concept during an internship at Cisco. Want to see the device package and its source code? Check out these Github repositories: https://github.com/FServais/NGINX-Device-Package https://github.com/FServais/NGINX-Agent
Development of a Cisco ACI device package for NGINX as a Load-Balancer
Development of a Cisco ACI device package for NGINX as a Load-Balancer
Fabrice Servais
Integrating content delivery networks into your application infrastructure can offer many benefits, including major performance improvements for your applications. So understanding how CDNs perform — especially for your specific use cases — is vital. However, testing for measurement is complicated and nuanced, and results in metric overload and confusion. It's becoming increasingly important to understand measurement techniques, what they're telling you, and how to apply them to your actual content. In this session, we'll examine the challenges around measuring CDN performance and focus on the different methods for measurement. We'll discuss what to measure, important metrics to focus on, and different ways that numbers may mislead you. More specifically, we'll cover: Different techniques for measuring CDN performance Differentiating between network footprint and object delivery performance Choosing the right content to test Core metrics to focus on and how each impacts real traffic Understanding cache hit ratio, why it can be misleading, and how to measure for it
Measuring CDN performance and why you're doing it wrong
Measuring CDN performance and why you're doing it wrong
Fastly
Caching has been an essential strategy for greater performance in computing since the beginning of the field. Nearly all applications have data access patterns that make caching an attractive technique, but caching also has hidden trade-offs related to concurrency, memory usage, and latency. As we build larger distributed systems, caching continues to be a critical technique for building scalable, high-throughput, low-latency applications. Large systems tend to magnify the caching trade-offs and have created new approaches to distributed caching. There are unique challenges in testing systems like these as well. Ehcache and Terracotta provide a unique way to start with simple caching for a small system and grow that system over time with a consistent API while maintaining low-latency, high-throughput caching.
Scaling Your Cache And Caching At Scale
Scaling Your Cache And Caching At Scale
Alex Miller
MySQL and SSD
MySQL and SSD
Gea-Suan Lin
Hadoop summit 2010 frameworks panel elephant bird
Hadoop summit 2010 frameworks panel elephant bird
Kevin Weil
IXP Gaurab SANOG
Gaurab Ixp Tutorial
Gaurab Ixp Tutorial
Tariq Mustafa
Viewers also liked
(20)
How Content Delivery Networks Work
How Content Delivery Networks Work
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
Future of CDN - Next 10 Years - Ahmet Ozalp, Akamai Technologies - DigiWorld ...
Content Delivery Network
Content Delivery Network
Netflix CDN and Open Source
Netflix CDN and Open Source
Monitoring CDN Performance
Monitoring CDN Performance
NGINX High-performance Caching
NGINX High-performance Caching
Jifty
Jifty
Introduction to Angular2
Introduction to Angular2
CONTENT DELIVERY NETWORK
CONTENT DELIVERY NETWORK
A Better Rich Media Experience & Video Analytics at Arkena with Apache Hadoop
A Better Rich Media Experience & Video Analytics at Arkena with Apache Hadoop
CDN 2.0
CDN 2.0
China cache IDC Services 201611
China cache IDC Services 201611
Introduction to chinacache services
Introduction to chinacache services
Introduction to Nginx
Introduction to Nginx
Development of a Cisco ACI device package for NGINX as a Load-Balancer
Development of a Cisco ACI device package for NGINX as a Load-Balancer
Measuring CDN performance and why you're doing it wrong
Measuring CDN performance and why you're doing it wrong
Scaling Your Cache And Caching At Scale
Scaling Your Cache And Caching At Scale
MySQL and SSD
MySQL and SSD
Hadoop summit 2010 frameworks panel elephant bird
Hadoop summit 2010 frameworks panel elephant bird
Gaurab Ixp Tutorial
Gaurab Ixp Tutorial
Similar to Using CDN to improve performance
首先,简单介绍淘宝网的系统规模和增长速度,以及对软件基础设施带来的挑战;接着,回顾淘宝图片存储与CDN系统的发展历史,如何从商用系统一步一步走到完全自主的系统,描述自主系统的主要架构与设计思想、性能指标和现有的部署规模,并总结一些经验来指导系统研发;然后,描述淘宝在软件基础设施上的规划,并一一阐述当前主要项目的要点与进展状况,这包括TFS、TAIR、千亿级别的分布式表格系统OceanBase、MySQL优化、面向Java环境的专用计算平台、服务器平台、Linux内核定制与优化、组通讯夸父、CDN和低功耗服务器平台等;最后,总结一下软件基础设施研发的原则和经验。
稳定、高效、低碳 -淘宝软件基础设施构建实践
稳定、高效、低碳 -淘宝软件基础设施构建实践
Wensong Zhang
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
Wensong Zhang
taobao,cdn,tfs,cache,lvs,分布式,基础设施,淘宝,阿里巴巴,图片存储,海量数据,集群,架构
Taobao base
Taobao base
mysqlops
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
lovingprince58
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
drewz lin
20110821 Web Development on Cloud Platform - PIXNET
20110821 Web Development on Cloud Platform - PIXNET
Jui-Nan Lin
本报告先简要介绍淘宝网的业务和背后网站系统的主要组成部分,描述所使用的主要软件和当前的系统规模,以及未来的技术挑战,当前的淘宝网站系统是完全采用开源软件和自主开发。 接着,叙述淘宝系统的主要发展过程,在系统规模较小时主要采用商用软件,随着规模越来越大商用软件逐步无法满足需求,不得不采用开源软件结合自主开发,一方面通过使用开源软件提高系统开发的效率,另一方面通过自主开发可以按需求在架构上取舍将系统性能做到极致,这里会举三个例子说明,一是淘宝如何逐步建设世界上最大的面向图片的CDN系统,采用高性价比的混合存储系统,目前有1000Gbps流量的承载能力,今年会发展到约2400Gbps,二是根据网页使用图片的特定抛弃目录空间实现了高可扩展的对象存储系统TFS,目前部署了6PB多的存储空间,存放了4PB多的图片,今年会建设到12PB的存储空间,三是将淘宝的核心数据库从基于IBM小型机、Oracle商用数据库和EMC高端存储的解决方案迁移基于PC服务器加MySQL数据库,结合高速非易失的PCI-E存储设备和多层次的系统优化,实现近百倍的性能提升,非常轻松地应付双11和双12的大促活动。这个过程可以总结出商用软件不能满足大规模系统的需求,采用开源软件与自主开发相结合,有更好的可控性,更高的可扩展性,规模效应使得研发投入都有更高的产出。 然后,描述淘宝的开源策略,淘宝系统中使用大量的开源软件,并在开源软件的基础上进行改进和定制,并把工作成果回馈给上流的开源社区,例如淘宝维护了自己的Linux内核树并不断地向Linux社区贡献patch,目前淘宝在对Linux内核贡献补丁数的公司排名为115,开源了基于Nginx的Tengine服务器,以及向Oracle回馈了JVM的补丁和MySQL的补丁,向Apache回馈hadoop和hbase的补丁等等。淘宝建设了淘蝌蚪开源平台,并开源多年开发主力的基础软件,如分布式存储系统TFS、分布式缓存和K/V系统TAIR、海量数据库OceanBase、分布式数据库中间件 TDDL、手机自动化测试框架Athrun等等。淘宝针对数据密集型应用的特点定制了低功耗服务器,先应用在CDN系统中,淘宝创建了greencompute.org网站,并开源了低功耗服务器的硬件规格和应用测试数据,阿里云开源了直流服务器,和业界一起推进绿色计算。 淘宝是开源系统的受益者,并积极参与开源生态系统的建设,促进开源生态系统的发展,积累更好的口碑,凝聚人才,迎接未来更大的技术挑战。
开源+自主开发 - 淘宝软件基础设施构建实践
开源+自主开发 - 淘宝软件基础设施构建实践
Wensong Zhang
淘宝对象存储与Cdn系统到服务
淘宝对象存储与Cdn系统到服务
drewz lin
首先,描述淘宝网对图片等多媒体对象存储和发送需求,介绍淘宝对象存储与发送系统的全貌和现有规模;接着详细描述存储淘宝图片的分布式文件系统TFS,它的架构与设计思想、性能指标和现有的部署规模;然后阐述淘宝CDN系统的总体结构,一级Cache和二级Cache的规划和基于开源的系统实现和优化,使用SSD/SAS/SATA混合存储系统,并根据访问热度变化将对象在不同的存储介质上迁移;最后,描述从淘宝对象存储与CDN系统走向基础设施服务,并总结这一过程中的经验和经济规律模型。
Taobao图片存储与cdn系统到服务
Taobao图片存储与cdn系统到服务
Wensong Zhang
大型视频网站单点分析与可用性提升-Qcon2011
大型视频网站单点分析与可用性提升-Qcon2011
Yiwei Ma
Dreaming Infrastructure
Dreaming Infrastructure
kyhpudding
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
智杰 付
淘宝图片存储与Cdn系统
淘宝图片存储与Cdn系统
Dai Jun
首先,介绍淘宝图片系统在整个网站中的重要性和发展历史;接着详细描述存储淘宝图片的分布式文件系统,以及它的架构、性能指标和现有的部署规模;然后介绍基于Nginx的Image Server和Cache系统;再阐述淘宝CDN系统的总体结构,一级Cache和二级Cache的规划和基于开源的系统实现;最后,总结一下淘宝图片存储和发送系统设计的原则和经验。
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
Wensong Zhang
Taobao 海量图片存储与CDN系统02
Taobao 海量图片存储与CDN系统02
lovingprince58
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
Michael Zhang
章文嵩:Taobao海量图片存储与cdn系统 v2-系统架构师
章文嵩:Taobao海量图片存储与cdn系统 v2-系统架构师
Enlight Chen
首先,介绍淘宝图片系统在整个网站中的重要性和发展历史;接着详细描述存储淘宝图片的分布式文件系统,以及它的架构与设计思想、性能指标和现有的部署规模;然后介绍基于Nginx的Image Server和Cache系统;再阐述淘宝CDN系统的总体结构,一级Cache和二级Cache的规划和基于开源的系统实现,使用SSD/SAS/SATA混合存储系统,并根据访问热度变化将对象在不同的存储介质上迁移;最后,总结一下淘宝图片存储和发送系统设计的原则和经验。
Taobao海量图片存储与cdn系统 v2-系统架构师
Taobao海量图片存储与cdn系统 v2-系统架构师
Wensong Zhang
视频Cdn架构浅淡 守住每一天
视频Cdn架构浅淡 守住每一天
liuyu105
北京地区PHP爱好者2010聚会[12月19日胜利举办,CU、ThinkinginLamp联办],黑夜路人主讲《构建基于Lamp的中型网站架构》,详情见详情goo.gl/xCCxy或goo.gl/j89N9
构建基于Lamp的中型网站架构
构建基于Lamp的中型网站架构
HonestQiao
Similar to Using CDN to improve performance
(20)
稳定、高效、低碳 -淘宝软件基础设施构建实践
稳定、高效、低碳 -淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
Taobao base
Taobao base
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
淘宝软件基础设施构建实践
20110821 Web Development on Cloud Platform - PIXNET
20110821 Web Development on Cloud Platform - PIXNET
开源+自主开发 - 淘宝软件基础设施构建实践
开源+自主开发 - 淘宝软件基础设施构建实践
淘宝对象存储与Cdn系统到服务
淘宝对象存储与Cdn系统到服务
Taobao图片存储与cdn系统到服务
Taobao图片存储与cdn系统到服务
大型视频网站单点分析与可用性提升-Qcon2011
大型视频网站单点分析与可用性提升-Qcon2011
Dreaming Infrastructure
Dreaming Infrastructure
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
淘宝图片存储与Cdn系统
淘宝图片存储与Cdn系统
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
Taobao 海量图片存储与CDN系统02
Taobao 海量图片存储与CDN系统02
Taobao海量图片存储与cdn系统
Taobao海量图片存储与cdn系统
章文嵩:Taobao海量图片存储与cdn系统 v2-系统架构师
章文嵩:Taobao海量图片存储与cdn系统 v2-系统架构师
Taobao海量图片存储与cdn系统 v2-系统架构师
Taobao海量图片存储与cdn系统 v2-系统架构师
视频Cdn架构浅淡 守住每一天
视频Cdn架构浅淡 守住每一天
构建基于Lamp的中型网站架构
构建基于Lamp的中型网站架构
Using CDN to improve performance
1.
Using CDN to
improve performance Gea-Suan Lin [email_address]
2.
如果有問題…
3.
請不要舉手,
4.
請直接打斷!
5.
亂發問才會亂爆八卦
6.
我不能回答的,
7.
有人可以回答。
8.
9.
不過因為投影片超過一百張
10.
所以請節制…
11.
Anyway…
12.
開始。
13.
從小網站開始…
14.
一台 Server
15.
一台 Web
一台 Database
16.
…
17.
不小心 把網站搞大了
18.
Tuning
19.
很多台 Web
20.
DNS Round Robin
21.
每一台 Web
都有 Public IP
22.
…
23.
前面架一台 Load
balancer
24.
兩個禮拜後…
25.
Load balancer
凌晨三點當掉
26.
總經理還沒睡
27.
把換燈管的叫起來
28.
換燈管的把我叫起來
29.
…
30.
買硬體的 Load
balancer
31.
出事就找人罰站
32.
繼續 Tuning
33.
用 Apache
提供 css ?
34.
一個 css
3KB 佔用一個 50MB 的 httpd…
35.
動態頁面 與 靜態檔案
分離
36.
靜態檔案
37.
獨立網域
38.
static.bubble.com
39.
nginx 或
lighttpd
40.
為什麼用 nginx
?
41.
CSS/JavaScript + gzip
on-the-fly
42.
一台 nginx
43.
兩台 nginx
( 兩份檔案 )
44.
出事了…
45.
客服殺人…
46.
IE6 + gzip
地雷
47.
IE6 時
gzip 關掉
48.
繼續成長…
49.
N 台
nginx (N 份檔案 )
50.
…
51.
政治不正確
52.
Reverse Proxy Cache
53.
Squid
54.
Invalidate 問題
55.
/main.css
56.
/main.css?v=[ 修改時間 ]
57.
/main.css?v=[ CRC32 ]
58.
講到這裡,有沒有問題?
59.
網站繼續大下去…
60.
「拓展海外市場」
61.
需要改善非台灣的瀏覽速度
62.
光速是固定的
63.
台北 – 美西
120+ms
64.
台北 – 美東
180+ms
65.
在美國放伺服器
66.
用 GeoDNS
分配流量
67.
寫 Health
Check 當國外當掉時改用台灣的伺服器
68.
ZzZz…
69.
除了美洲還有歐洲
70.
自己管理的成本過高
71.
外包給專業的…
72.
CDN
73.
( 終於進入主題了 )
74.
Content Delivery Network
75.
內容傳遞網路
76.
( 屁 )
77.
等於沒解釋…
78.
多個伺服器群
79.
多個地點
80.
提供相同的內容
81.
所以…
82.
83.
加快 TCP
handshake 速度
84.
加快 TCP
下載速度
85.
2. 提供高可靠度
86.
不同地點互相備援
87.
3. DoS
防禦
88.
「由專業公司防禦 DoS
」
89.
「死道友不死貧道」
90.
4. 成本
91.
CDN 頻寬比較貴?
92.
台灣頻寬比 CDN
貴!
93.
一般頻寬計算方法: 95%
94.
五分鐘取樣一次
95.
一個月 30
天共 8640 次
96.
刪除最高的 5%
(432 次 )
97.
取剩下最高的
98.
From http://en.wikipedia.org/wiki/Burstable_billing
99.
CDN 頻寬計算方式:總量
100.
總量 =
平均流量 ( 等價 )
101.
1Mbps ~ 10GB/day
102.
換算?
103.
依照網站的 MRTG
Pattern
104.
From http://en.wikipedia.org/wiki/Burstable_billing
105.
95% 2.62M
與平均 1.27M
106.
大約 2.06
倍
107.
5. 克服瀏覽器壓縮
bug
108.
用 CDN
後總量反而下降
109.
講到這裡,有沒有問題?
110.
CDN 的分類
111.
Latency 、 Thoughput
、 Cost
112.
CSS 、 JavaScript
113.
低 Latency
、初期 Thoughput 要高、 Cost 不是主要考量
114.
亞洲要有直接連線的 PoPs
115.
尤其是 HiNet
與 TANet
116.
影音
117.
Latency 不重要,
Thoughput 足夠就好, Cost 要低
118.
封包到美西再到香港
119.
所以…
120.
( 如果有時間的話 )
121.
總是要講一下實際案例?
122.
Akamai
123.
很大,非常大,第一大
124.
效能是最好的
125.
台灣唯一有 PoPs
的 CDN
126.
台灣有經銷商
127.
價錢…
128.
Limelight Networks
129.
第二大?
130.
YouTube 被
Google 買之前的 CDN
131.
有香港與日本的 PoPs
132.
但是都是導到美西 PoPs
133.
第二大不一定比其他的好
134.
CDNetworks
135.
第三大?
136.
二月買下 Panther
Express
137.
討論 Panther
Express
138.
有香港與日本的 PoPs
139.
HiNet 與
TANet 到香港都是 20+ms
140.
HiNet 調整
routing 時會自動跳到日本,約 30+ms
141.
PIXNET 的
CSS/JavaScript
142.
EdgeCast
143.
第 N
大 ( 不重要了 )
144.
有香港與日本的 PoPs
145.
會導到香港的 PoP
,但是會先到美西
146.
但是速度沒問題
147.
PIXNET 的影音
148.
Voxel
149.
就是一家 CDN
業者…
150.
價錢公佈在網站上
151.
沒有亞洲的 PoPs
152.
Amazon CloudFront
153.
Because it’s Amazon…
154.
Don’t waste your
time
155.
Amazon S3
應該找其他 CDN 搭配用
156.
SimpleCDN
157.
完全是拼價錢的 CDN
158.
目前沒有亞洲的 PoPs
159.
不吃 file.css?v=[
時間 ]
160.
就…
161.
實際案例
162.
( 如果還有時間的話 )
163.
download.microsoft.com
164.
Akamai + Limelight
165.
js.microsoft.com
166.
Akamai
167.
www.barackobama.com
168.
Panther Express (g2)
169.
US-only PoPs
170.
public.slideshare.net
171.
Panther Express (g1)
172.
US + EU
PoPs
173.
cdn.slideshare.net
174.
Panther Express (l1)
175.
價錢導向的 CDN
Level
176.
cdn1.badongo.com cdn2.badongo.com cdn3.badongo.com
cdn4.badongo.com
177.
Amazon CloudFront Akamai
EdgeCast ( 目前是這個 ) EdgeCast
178.
Thanks
Download now