DOCSLIB.ORG

Frontera-Crawling the Spanish

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Frontera-Crawling the Spanish Frontera: open source, large scale web crawling framework Alexander Sibiryakov, October 1, 2015 [email protected] Sziasztok résztvevők! • Born in Yekaterinburg, RU • 5 years at Yandex, search quality department: social and QA search, snippets. • 2 years at Avast! antivirus, research team: automatic false positive solving, large scale prediction of malicious download attempts. 2 Task • Crawl Spanish web to gather statistics about hosts and their sizes. • Limit crawl to .es zone. • Breadth-first strategy: first crawl 1-click distance documents, next 2-clicks, and so on, • Finishing condition: absence of hosts with less than 100 crawled documents. • Low costs. 3 Spanish internet (.es) in 2012 • Domain names registered - 1,56М (39% growth per year) • Web server in zone - 283,4K (33,1%) • Hosts - 4,2M (21%) • Spanish web sites in DMOZ catalog - 22043 * - отчет OECD Communications Outlook 2013 4 Solution • Scrapy* - network operations. • Apache Kafka - data bus (offsets, partitioning). • Apache HBase - storage (random access, linear scanning, scalability). • Twisted.Internet - library for async primitives for use in workers. • Snappy - efficient compression algorithm for IO-bounded applications. * - network operations in Scrapy are implemented asynchronously, based on the same Twisted.Internet 5 Architecture Kafka topic Crawling strategy SW workers DB Storage workers 6 1. Big and small hosts problem • When crawler comes to huge number of links from some host, along with usage of simple prioritization models, it turns out queue is flooded with URLs from the same host. • That causes underuse of spider resources. • We adopted additional per- host (optionally per-IP) queue and metering algorithm: URLs from big hosts are cached in memory. 7 3. DDoS DNS service Amazon AWS • Breadth-first strategy assumes first visiting of previously unknown hosts, therefore generating huge amount of DNS request. • Recursive DNS server on each downloading node, with upstream set to Verizon and OpenDNS. • We used dnsmasq. 8 4. Tuning Scrapy thread pool’а for efficient DNS resolution • Scrapy uses a thread pool to resolve DNS name to IP. • When ip is absent in cache, request is sent to DNS server in it’s own thread, which is blocking. • Scrapy reported numerous errors related to DNS name resolution and timeouts. • We added option to Scrapy for thread pool size and timeout adjustment. 9 5. Overloaded HBase region servers during state check • Crawler extracts from document hundreds of links in average. • Before adding this links to queue, they needs to be checked if they weren’t already crawled (to avoid repetitive visiting). • On small volumes SSDs were just fine. After increase of table size, we had to move to HDDs, and response times dramatically grew up. • Host-local fingerprint function for keys in HBase. • Tuning HBase block cache to fit average host states into one block. 10 6. Intensive network traffic from workers to services • We noticed throughput between workers Kafka and HBase up to 1Gbit/s. • Switched to Thrift compact protocol for HBase communication. • Message compression in Kafka using Snappy. 11 7. Further query and traffic optimizations to HBase • State check required lion’s share of requests and network throughput. • Consistency was another requirement. • We created local state cache in strategy worker. • For consistency, spider log was partitioned by host, to avoid cache overlap between workers. 12 State cache • All operations are batched: • If key is absent in cache, it’s requested from HBase, • every ~4K documents cache is flushed to HBase. • When achieving 3M (~1Гб) elements, flush and cleanup happens. • It seems Least-Recently-Used (LRU) algorithm is a good fit there. Spider priority queue (slot) • Cell has an array of: - fingerprint, - Crc32(hostname), - URL, - score • Dequeueing top N. • Such design is prone to huge hosts. • Partially this problem can be solved using scoring model taking into account known document count per host. 14 8. Problem of big and small hosts (strikes back!) • During crawling we’ve found few very huge hosts (>20M docs) • All queue partitions were flooded with pages from few huge hosts, because of queue design and scoring model used. • We made two MapReduce jobs: • queue shuffling, • limiting all hosts to no more than 100 documents. 15 Hardware requirements • Single-thread Scrapy spider gives 1200 pages/min. from about 100 websites in parallel. • Spiders to workers ratio is 4:1 (without content) • 1 Gb of RAM for every SW (state cache, tunable). • Example: • 12 spiders ~ 14.4K pages/min., • 3 SW and 3 DB workers, • Total 18 cores. Software requirements • Apache HBase, • Apache Kafka, CDH (100% Open source Hadoop package) • Python 2.7+, • Scrapy 0.24+, • DNS Service. 17 Maintaining Cloudera Hadoop on Amazon EC2 • CDH is very sensitive to free space on root partition, parcels, and storage of Cloudera Manager. • We’ve moved it using symbolic links to separate EBS partition. • EBS should be at least 30Gb, base IOPS should be enough. • Initial hardware was 3 x m3.xlarge (4 CPU, 15Gb, 2x40 SSD). • After one week of crawling, we ran out of space, and started to move DataNodes to d2.xlarge (4 CPU, 30.5Gb, 3x2Tb HDD). Spanish (.es) internet crawl results • fnac.es, rakuten.es, adidas.es, equiposdefutbol2014.es, druni.es, docentesconeducacion.es - are the biggest websites • 68.7K domains found (~600K expected), • 46.5M crawled pages overall, • 1.5 months, • 22 websites with more than 50M pages where are the rest of web servers?! Bow-tie model A. Broder et al. / Computer Networks 33 (2000) 309-320 Y. Hirate, S. Kato, and H. Yamana, Web Structure in 2005 Graph Structure in the Web — Revisited, Meusel, Vigna, WWW 2014 Main features • Online operation: scheduling of new batch, updating of DB state. • Storage abstraction: write your own backend (sqlalchemy, HBase is included). • Canonical URLs resolution abstraction: each document has many URLs, which to use? • Scrapy ecosystem: good documentation, big community, ease of customization. 24 Main features • Communication layer is Apache Kafka: topic partitioning, offsets mechanism. • Crawling strategy abstraction: crawling goal, url ordering, scoring model is coded in separate module. • Polite by design: each website is downloaded by at most one spider. • Python: workers, spiders. References • Distributed Frontera. https://github.com/ scrapinghub/distributed-frontera • Frontera. https://github.com/scrapinghub/frontera • Documentation: • http://distributed-frontera.readthedocs.org/ • http://frontera.readthedocs.org/ 26 Future plans • Lighter version, without HBase and Kafka. Communicating using sockets. • Revisiting strategy out-of-box. • Watchdog solution: tracking website content changes. • PageRank or HITS strategy. • Own HTML and URL parsers. • Integration into Scrapinghub services. • Testing on larger volumes. 27 Contribute! • Distributed Frontera is a historically first attempt to implement web scale web crawler using Python. • Truly resource-intensive task: CPU, network, disks. • Made in Scrapinghub, a company where Scrapy was created. • A plans to become an Apache Software Foundation project. 28 We’re hiring! http://scrapinghub.com/jobs/ 29 Köszönöm! Thank you! Alexander Sibiryakov, [email protected].
Load more

Recommended publications
  • Cluster Setup Guide
    Frontera Documentation Release 0.8.0.1 ScrapingHub Jul 30, 2018 Contents 1 Introduction 3 1.1 Frontera at a glance...........................................3 1.2 Run modes................................................5 1.3 Quick start single process........................................6 1.4 Quick start distributed mode.......................................8 1.5 Cluster setup guide............................................ 10 2 Using Frontera 15 2.1 Installation Guide............................................ 15 2.2 Crawling strategies............................................ 16 2.3 Frontier objects.............................................. 16 2.4 Middlewares............................................... 19 2.5 Canonical URL Solver.......................................... 22 2.6 Backends................................................. 23 2.7 Message bus............................................... 27 2.8 Writing custom crawling strategy.................................... 30 2.9 Using the Frontier with Scrapy...................................... 35 2.10 Settings.................................................. 37 3 Advanced usage 51 3.1 What is a Crawl Frontier?........................................ 51 3.2 Graph Manager.............................................. 52 3.3 Recording a Scrapy crawl........................................ 58 3.4 Fine tuning of Frontera cluster...................................... 59 3.5 DNS Service............................................... 60 4 Developer documentation 63 4.1
    [Show full text]
  • Frontera Documentation Release 0.4.0
    Frontera Documentation Release 0.4.0 ScrapingHub December 30, 2015 Contents 1 Introduction 3 1.1 Frontera at a glance...........................................3 1.2 Run modes................................................5 1.3 Quick start single process........................................6 1.4 Quick start distributed mode.......................................8 2 Using Frontera 11 2.1 Installation Guide............................................ 11 2.2 Frontier objects.............................................. 11 2.3 Middlewares............................................... 12 2.4 Canonical URL Solver.......................................... 13 2.5 Backends................................................. 14 2.6 Message bus............................................... 16 2.7 Crawling strategy............................................. 17 2.8 Using the Frontier with Scrapy...................................... 17 2.9 Settings.................................................. 20 3 Advanced usage 31 3.1 What is a Crawl Frontier?........................................ 31 3.2 Graph Manager.............................................. 32 3.3 Recording a Scrapy crawl........................................ 38 3.4 Production broad crawling........................................ 40 4 Developer documentation 45 4.1 Architecture overview.......................................... 45 4.2 Frontera API............................................... 47 4.3 Using the Frontier with Requests.................................... 49 4.4
    [Show full text]
  • High Performance Distributed Web-Scraper
    High performance distributed web-scraper Denis Eyzenakh Anton Rameykov Igor Nikiforov Institute of Computer Science and Institute of Computer Science and Institute of Computer Science and Technology Technology Technology Peter the Great St.Petersburg Peter the Great St.Petersburg Peter the Great St.Petersburg Polytechnic University Polytechnic University Polytechnic University Saint – Petersburg, Russian Federation Saint – Petersburg, Russian Federation Saint – Petersburg, Russian Federation [email protected] [email protected] [email protected] Abstract—Over the past decade, the Internet has become the gigantic and richest source of data. The data is used for the II. EXISTING WEB SCRAPING TECHNIQUES extraction of knowledge by performing machine leaning analysis. Typically, web scraping applications imitate a regular web In order to perform data mining of the web-information, the data user. They follow the links and search for the information should be extracted from the source and placed on analytical storage. This is the ETL-process. Different web-sources have they need. The classic web scraper can be classified into two different ways to access their data: either API over HTTP protocol types: web-crawlers and data extractors “Fig. 1”. or HTML source code parsing. The article is devoted to the approach of high-performance data extraction from sources that do not provide an API to access the data. Distinctive features of the proposed approach are: load balancing, two levels of data storage, and separating the process of downloading files from the process of scraping. The approach is implemented in the solution with the following technologies: Docker, Kubernetes, Scrapy, Python, MongoDB, Redis Cluster, and СephFS.
    [Show full text]
  • Scrapy Cluster Documentation Release 1.2
    Scrapy Cluster Documentation Release 1.2 IST Research November 29, 2016 Contents 1 Introduction 3 1.1 Overview.................................................3 1.2 Quick Start................................................5 2 Kafka Monitor 13 2.1 Design.................................................. 13 2.2 Quick Start................................................ 14 2.3 API.................................................... 16 2.4 Plugins.................................................. 31 2.5 Settings.................................................. 34 3 Crawler 37 3.1 Design.................................................. 37 3.2 Quick Start................................................ 42 3.3 Controlling................................................ 44 3.4 Extension................................................. 49 3.5 Settings.................................................. 53 4 Redis Monitor 59 4.1 Design.................................................. 59 4.2 Quick Start................................................ 60 4.3 Plugins.................................................. 61 4.4 Settings.................................................. 64 5 Rest 69 5.1 Design.................................................. 69 5.2 Quick Start................................................ 69 5.3 API.................................................... 73 5.4 Settings.................................................. 77 6 Utilites 81 6.1 Argparse Helper............................................. 81 6.2 Log Factory..............................................
    [Show full text]
  • Frontera-Open Source Large Scale Web Crawling Framework
    Frontera: Large-Scale Open Source Web Crawling Framework Alexander Sibiryakov, 20 July 2015 [email protected] Hola los participantes! • Born in Yekaterinburg, RU • 5 years at Yandex, search quality department: social and QA search, snippets. • 2 years at Avast! antivirus, research team: automatic false positive solving, large scale prediction of malicious download attempts. 2 «A Web crawler starts with a list of URLs to visit, called the seeds. As the crawler visits these URLs, it identifies all the hyperlinks in the page and adds them to the list of URLs to visit, called the crawl frontier.». –Wikipedia: Web Crawler article, July 2015 3 Motivation • Client needed to crawl 1B+ pages/week, and identify frequently changing HUB pages. • Scrapy wasn’t suitable for broad crawling and had no crawl frontier capabilities* • People were tend to favor Apache Nutch instead of Scrapy. Hyperlink-Induced Topic Search, Jon Kleinberg, 1999 4 tophdart.com Frontera: single-threaded and distributed • Frontera is all about knowing what to crawl next and when to stop. • Single-Threaded mode can be used for up to 100 websites (parallel downloading), • for performance broad crawls there is a distributed mode. 6 Main features • Online operation: scheduling of new batch, updating of DB state. • Storage abstraction: write your own backend (sqlalchemy, HBase is included). • Canonical URLs resolution abstraction: each document has many URLs, which to use? • Scrapy ecosystem: good documentation, big community, ease of customization. 7 Single-threaded use cases • Need of URL metadata and content storage, • Need of isolation of URL ordering/queueing logic from the spider • Advanced URL ordering logic (big websites, or revisiting) 8 Single-threaded architecture 9 Frontera and Scrapy • Frontera is implemented as a set of custom scheduler and spider middleware for Scrapy.
    [Show full text]
  • Distributed Training
    Scalable and Distributed DNN Training on Modern HPC Systems: Challenges and Solutions Keynote Talk at SDAS ‘19 by Dhabaleswar K. (DK) Panda The Ohio State University E-mail: [email protected] http://www.cse.ohio-state.edu/~panda Understanding the Deep Learning Resurgence • Deep Learning is a sub-set of Machine Learning – But, it is perhaps the most radical and revolutionary subset – Automatic feature extraction vs. hand-crafted features • Deep Learning – A renewed interest and a lot of hype! – Key success: Deep Neural Networks (DNNs) – Everything was there since the late 80s except the “computability of DNNs” Courtesy: http://www.deeplearningbook.org/contents/intro.html Network Based Computing Laboratory SDAS (June ‘19) 2 Deep Learning Use Cases and Growth Trends Courtesy: https://www.top500.org/news/market-for-artificial-intelligence-projected-to-hit-36-billion-by-2025/ Network Based Computing Laboratory SDAS (June ‘19) 3 Increasing Usage of HPC, Big Data and Deep Learning Big Data HPC (Hadoop, Spark, (MPI, RDMA, HBase, Lustre, etc.) Memcached, etc.) Convergence of HPC, Big Deep Learning Data, and Deep Learning! (Caffe, TensorFlow, BigDL, etc.) Increasing Need to Run these applications on the Cloud!! Network Based Computing Laboratory SDAS (June ‘19) 4 Newer Workflows - Deep Learning over Big Data (DLoBD) • Deep Learning over Big Data (DLoBD) is one of the most efficient analyzing paradigms • More and more deep learning tools or libraries (e.g., Caffe, TensorFlow) start running over big data stacks, such as Apache Hadoop and Spark
    [Show full text]
  • HPC and AI Middleware for Exascale Systems and Clouds
    HPC and AI Middleware for Exascale Systems and Clouds Talk at HPC-AI Advisory Council UK Conference (October ‘20) by Dhabaleswar K. (DK) Panda The Ohio State University E-mail: [email protected] Follow us on http://www.cse.ohio-state.edu/~panda https://twitter.com/mvapich High-End Computing (HEC): PetaFlop to ExaFlop 100 PetaFlops in 415 Peta 2017 Flops in 2020 (Fugaku in Japan with 7.3M cores 1 ExaFlops Expected to have an ExaFlop system in 2021! Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 2 Increasing Usage of HPC, Big Data and Deep/Machine Learning Big Data (Hadoop, Spark, HPC HBase, (MPI, PGAS, etc.) Memcached, etc.) Convergence of HPC, Big Deep/Machine Data, and Deep/Machine Learning Learning! (TensorFlow, PyTorch, BigDL, cuML, etc.) Increasing Need to Run these applications on the Cloud!! Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 3 Converged Middleware for HPC, Big Data and Deep/Machine Learning? Physical Compute Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 4 Converged Middleware for HPC, Big Data and Deep/Machine Learning? Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 5 Converged Middleware for HPC, Big Data and Deep/Machine Learning? Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 6 Converged Middleware for HPC, Big Data and Deep/Machine Learning? Hadoop Job Deep/Machine Learning Job Spark Job Network Based Computing Laboratory HPC-AI-UK (Oct ‘20) 7 Presentation Overview • MVAPICH Project – MPI and PGAS (MVAPICH) Library with CUDA-Awareness • HiBD Project – High-Performance
    [Show full text]
  • Vergleich Aktueller Web-Crawling-Werkzeuge
    Hochschule Wismar University of Applied Sciences Technology, Business and Design Fakult¨at fur¨ Ingenieurwissenschaften, Bereich EuI Bachelor-Thesis Vergleich aktueller Web-Crawling-Werkzeuge Gedruckt am: 30. April 2021 Eingereicht am: von: Christoph Werner Betreuende Professorin: Prof. Dr.-Ing. Antje Raab-Dusterh¨ ¨oft Zweitbetreuer: Prof. Dr.-Ing. Matthias Kreuseler Aufgabenstellung Das Ziel der Bachelor Thesis ist ein Vergleich aktueller Web-Crawling-Werkzeuge. Die Crawler sind hinsichtlich ihrer Funktionsweise und Crawling-Ergebnisse zu ver- gleichen und zu bewerten. Des Weiteren ist ein Tool zu konzeptionieren, welches diese Werkzeuge fur¨ die Nutzung in Laborpraktika verfugbar¨ macht. 3 Kurzreferat Webcrawling im Verbund mit Webscraping ist eine der grundlegenden Technologien, um gezielt und automatisiert Daten aus dem Internet zu sammeln. Zun¨achst wird in dieser Thesis auf die Grundlagen des Crawlings und Scrapings eingegangen. Hierbei soll besonderes Augenmerk auf der Architektur und Funkti- onsweise eines Crawlers, dem Robots Exclusion Protocol (REP), zu bedenkende Sicherheitsaspekte, sowie Anti-Crawling/Scraping-Maßnahmen liegen. Darauf aufbauend werden verschiedenste Crawling Frameworks auf Grundlage ihrer Dokumentation bewertet und gegenubergestellt.¨ Abschließend wird ein Tool mit einer grafischen Benutzeroberfl¨ache (GUI) zum Ver- gleich von verschiedenen Crawling Frameworks entwickelt. 4 Abstract Web crawling in combination with web scraping is the key technology for the targeted and automated collection of data from the World Wide Web. First of all, this thesis deals with the basics of crawling and scraping. Special at- tention will be paid to the architecture and functionality of a crawler, the Ro- bots Exclusion Protocol (REP), security aspects to be considered as well as anti- crawling/scraping measures. Based on this, various crawling frameworks are evaluated and compared on the basis of their documentation.
    [Show full text]
  • Parsl Documentation Release 1.1.0
    Parsl Documentation Release 1.1.0 The Parsl Team Sep 24, 2021 CONTENTS 1 Quickstart 3 1.1 Installation................................................3 1.2 Getting started..............................................4 1.3 Tutorial..................................................4 1.4 Usage Tracking..............................................4 1.5 For Developers..............................................5 2 Parsl tutorial 7 2.1 Configuring Parsl.............................................7 2.2 Python Apps...............................................8 2.3 Bash Apps................................................8 2.4 Passing data...............................................9 2.5 AppFutures................................................9 2.6 DataFutures................................................ 10 2.7 Files................................................... 11 2.8 Remote Files............................................... 11 2.9 Sequential workflow........................................... 12 2.10 Parallel workflow............................................. 12 2.11 Parallel dataflow............................................. 13 2.12 Monte Carlo workflow.......................................... 14 2.13 Local execution with threads....................................... 15 2.14 Local execution with pilot jobs..................................... 15 3 User guide 17 3.1 Overview................................................. 17 3.2 Apps................................................... 23 3.3 Futures.................................................
    [Show full text]
  • Frontera Documentation Release 0.8.0
    Frontera Documentation Release 0.8.0 ScrapingHub Jul 30, 2018 Contents 1 Introduction 3 1.1 Frontera at a glance...........................................3 1.2 Run modes................................................5 1.3 Quick start single process........................................6 1.4 Quick start distributed mode.......................................8 1.5 Cluster setup guide............................................ 10 2 Using Frontera 15 2.1 Installation Guide............................................ 15 2.2 Crawling strategies............................................ 16 2.3 Frontier objects.............................................. 16 2.4 Middlewares............................................... 19 2.5 Canonical URL Solver.......................................... 22 2.6 Backends................................................. 23 2.7 Message bus............................................... 27 2.8 Writing custom crawling strategy.................................... 30 2.9 Using the Frontier with Scrapy...................................... 35 2.10 Settings.................................................. 37 3 Advanced usage 51 3.1 What is a Crawl Frontier?........................................ 51 3.2 Graph Manager.............................................. 52 3.3 Recording a Scrapy crawl........................................ 58 3.4 Fine tuning of Frontera cluster...................................... 59 3.5 DNS Service............................................... 60 4 Developer documentation 63 4.1
    [Show full text]
  • Frontera Documentation 0.6.0
    Frontera Documentation 0.6.0 ScrapingHub 2017 02 13 Contents 1 3 1.1 Frontera at a glance...........................................3 1.2 Run modes................................................5 1.3 Quick start single process........................................7 1.4 Quick start distributed mode.......................................8 1.5 Cluster setup guide............................................9 2 Frontera 13 2.1 Installation Guide............................................ 13 2.2 Frontier objects.............................................. 13 2.3 Middlewares............................................... 14 2.4 Canonical URL Solver.......................................... 15 2.5 Backends................................................. 16 2.6 Message bus............................................... 18 2.7 Crawling strategy............................................. 19 2.8 Using the Frontier with Scrapy...................................... 20 2.9 Settings.................................................. 23 3 33 3.1 What is a Crawl Frontier?........................................ 33 3.2 Graph Manager.............................................. 34 3.3 Recording a Scrapy crawl........................................ 40 3.4 Fine tuning of Frontera cluster...................................... 42 3.5 DNS Service............................................... 43 4 45 4.1 Architecture overview.......................................... 45 4.2 Frontera API............................................... 47 4.3 Using the
    [Show full text]
  • High Performance Distributed Web-Scraper
    Труды ИСП РАН, том 33, вып. 3, 2021 г. // Trudy ISP RAN/Proc. ISP RAS, vol. 33, issue 3, 2021 Eyzenakh D.S., Rameykov A.S., Nikiforov I.V. High performance distributed web-scraper. Trudy ISP RAN/Proc. ISP RAS, vol. 33, issue 3, 2021, pp. 87-100 Для цитирования: Эйзенах Д.С., Рамейков А.С., Никифоров И.В. Высокопроизводительный распределенный веб-скрапер. Труды ИСП РАН, том 33, вып. 3, 2021 г., стр. 87-100 (на английском языке). DOI: 10.15514/ISPRAS–2021–33(3)–7. DOI: 10.15514/ISPRAS-2021-33(3)-7 1. Introduction Due to the rapid development of the network, the World Wide Web has become a carrier of a large High Performance Distributed Web-Scraper amount of information. The data extraction and use of information has become a huge challenge nowadays. Traditional access to the information through browsers like Chrome, Firefox, etc. can provide a comfortable user experience with web pages. Web sites have a lot of information and sometimes haven’t got any instruments to access over the API and preserve it in analytical storage. D.S. Eyzenakh, ORCID: 0000-0003-1584-1745 <[email protected]> The manual collection of data for further analysis can take a lot of time and in the case of semi- A.S. Rameykov, ORCID: 0000-0001-7989-6732 <[email protected]> structured or unstructured data types the collection and analyzing of data can become even more I.V. Nikiforov, ORCID: 0000-0003-0198-1886 <[email protected]> difficult and time-consuming. The person who manually collects data can make mistakes Peter the Great St.Petersburg Polytechnic University (duplication, typos in the text, etc.) as far as the process is error-prone.
    [Show full text]