All Things Open : Crash Course in Open Source Cloud Computing
•Télécharger en tant que PPTX, PDF•
7 j'aime•2,767 vues
Very few trends in IT have generated as much buzz as cloud computing. This session will cut through the hype and quickly clarify the ontology for cloud computing. The bulk of the conversation will focus on the open source software that can be used to build compute clouds (infrastructure-as-a-service) and the complimentary open source management tools that can be combined to automate the management of cloud computing environments. The session will appeal to anyone who has a good grasp of traditional data center infrastructure but is struggling with the benefits and migration path to a cloud computing environment. Systems administrators and IT generalists will leave the discussion with a general overview of the options at their disposal to effectively build and manage their own cloud computing environments using free and open source software.
Recommandé
Recommandé
Contenu connexe
Tendances
Tendances (20)
En vedette
En vedette (16)
Similaire à All Things Open : Crash Course in Open Source Cloud Computing
Similaire à All Things Open : Crash Course in Open Source Cloud Computing (20)
Plus de Mark Hinkle
Plus de Mark Hinkle (18)
Dernier
Dernier (20)
All Things Open : Crash Course in Open Source Cloud Computing
- 1. All Things Open 2014 Crash Course in Open Source Cloud Computing Mark Hinkle Senior Director, Open Source Solutions Citrix Inc. mark.hinkle@citrix.com mrhinkle@gmail.com @mrhinkle
- 2. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com ABOUT ME I Help Build Open Source Ecosystems Open Source Experience • Manage Citrix Open Source Business Office • Apache CloudStack Committer and PMC Member • Advisory boards Gluster and Xen Project • Joined Citrix via Cloud.com acquisition July 2011 • Zenoss Core open source project to 100,000 users, 1.5 million downloads • Former LinuxWorld Magazine Editor-in-Chief • Open Management Consortium organizer • Author - “Windows to Linux Business Desktop Migration” – Thomson • NetDirector Project - Open Source Configuration Management All Things Open 2014 - Open Source Cloud Computing
- 3. http://www.slideshare.net/socializedsoftware Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Slides Available on Slideshare: Creative Commons Attributions-ShareAlike 4.0 International Share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the material for any purpose, even commercially. The licensor cannot revoke these freedoms as long as you follow the license terms. All Things Open 2014 - Open Source Cloud Computing
- 4. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com AGENDA • Vetting Open Source Cloud Projects • Virtualization • Infrastructure-as-a-Service • Platform-as-a-Service • SDN • Open Source for Amazon Web Services All Things Open 2014 - Open Source Cloud Computing
- 5. VETTING OPEN SOURCE PROJECTS How can you tell if they’re Legit • Code Velocity • Committers • Committer Reputation • User-driven or Vendor-Driven Innovation • User Activity • Corporate Support* • Reputation of Foundation* By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing
- 6. …the future of technological innovation is not stealing limited resources away from one another, but creating new resources — and new opportunities to create new resources — together in a rich ecosystem.” By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPEN SOURCE ISN’T A ZERO-SUM GAME Allison Randal Open Source Hacker Former OSCON Program Chair @allisonrandal All Things Open 2014 - Open Source Cloud Computing
- 7. http://www.openhub.net http://activity.openstack.org By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPEN SOURCE ANALYSIS Visualizing Community Activity All Things Open 2014 - Open Source Cloud Computing
- 8. DevOps Toolchain By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPEN SOURCE CLOUD STACK Platform-as-a-Service (PaaS) ? ? Infrastructure-as-a-Service (IaaS) Orchestration ? Compute Storage Networking (Networking-as-a-Service) All Things Open 2014 - Open Source Cloud Computing Orchestration Configuration Management Monitoring
- 9. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com VIRTUALIZATION Carving up compute resources OPEN SOURCE • Xen Project • Citrix XenServer • KVM • VirtualBox • OpenVZ • LXC • libcontainer PROPRIETARY • VMware • Microsoft Hyper-V • OracleVM (Based on Xen Project) All Things Open 2014 - Open Source Cloud Computing
- 10. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com HYPERVISORS AND CONTAINERS Differences in virtualization Type 1 Hypervisors VMware, Xen Project, Hyper-V Type 2 Hypervisors KVM, VirtualBox All Things Open 2014 - Open Source Cloud Computing Containers LXC, libcontainer
- 11. • Different file formats for virtual machines (VMware uses vmdk file format, Xen and Hyper-V use VHD, KVM uses Raw or QCOW2) • Guest images may be “processor architecture” • VMware and Xen can manage SCSI devices, but • KVM and Xen can use virtio drivers but not • VMware uses a proprietary agent inside the guest OS (VMware tools) which does not work with Xen or KVM By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com THE PORTABILITY PROBLEM Containers compared to Hardware Virtualization bound KVM cannot VMware • Yada, Yada, Yada All Things Open 2014 - Open Source Cloud Computing
- 12. • Lets your run a Linux system within • A container is a group of processes on a Linux box, put together the provide an isolated environment • From the inside, it looks like a VM • Externally it looks like normal processes • “chroot on steroids” By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com LINUX CONTAINERS “Lightweight” Linux Virtualization another Linux system All Things Open 2014 - Open Source Cloud Computing
- 13. • Code – Application is stored • Build – Code is built (Jenkins) • Test – Unit tests are By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com CONTINUOUS INTEGRATION Rebuild Applications on any Cloud and/or Virtualized Infrastructure in a repository (Subversion,Git) automated (Jenkins) • Deploy – Deploy code to server various ways Code Build Test Deploy Thoughtworks Go – Open Source Continuous Deliver System All Things Open 2014 - Open Source Cloud Computing
- 14. Docker is an open-source project to easily create lightweight, portable, self-sufficient containers from any application. The same container that a developer builds and tests on a laptop can run at scale, in production, on VMs, bare metal, public clouds and more. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com DOCKER CONTAINER PACKAGING Open source LXC Packaging Engine To learn more please visit: www.docker.io All Things Open 2014 - Open Source Cloud Computing
- 15. • Compliment to LXC not a replacement • Managed daemonized processes on Linux • Create ability to re-use and manage similar • Content agnostic • Hardware agnostic • Easy to automate • Integrated with other tools: Chef, OpenShift, By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com WHAT IS DOCKER System for Managing and Deploying LXC Containers using LXC libcontainer applications Puppet, VMware, etc. All Things Open 2014 - Open Source Cloud Computing
- 16. DOCKER’S GROWING ECOSYSTEM By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing
- 17. Kubernetes builds on top of Docker to construct a clustered container scheduling service. Kubernetes enables users to ask a cluster to run a set of containers. The system will automatically pick worker nodes to run those containers on, which we think of more as "scheduling" than "orchestration” To learn more please visit: https://github.com/GoogleCloudPlatform/kubernetes Greek for Shipmaster By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com KUBERNETES Container Cluster Management – Scheduler All Things Open 2014 - Open Source Cloud Computing
- 18. DOCKER RELATED PROJECTS • Fig -Fast, isolated development environments • Flynn - Next-generation application platform • Panamax – Drag-and-Drop Docker Containerization • Project Atomic – JEOS designed to run Docker containers • SocketPlane – Docker Networking (coming soon) • Weave – Docker Networking • 13,000+ Docker-related repos on Github By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing
- 19. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com $141 Billion Market Cap $363 Billion Market Cap All Things Open 2014 - Open Source Cloud Computing $356 Billion Market Cap PUBLIC CLOUD
- 20. Project Year Started License Virtualization By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com MINIMUM VIABLE CLOUD Infrastructure-as-a-Service | IaaS |Compute Orchestration All Things Open 2014 - Open Source Cloud Computing Technologies Apache CloudStack 2008 Apache (Bare Metal), Xenserver, KVM, LXC VMware Hyper- V Eucalyptus 2006 GPL Xen, KVM, VMware (commercial version) OpenNebula 2005 Apache Xen, KVM, VMware OpenStack 2010 (Developed by NASA by Anso Labs previously) Apache VMware ESX and ESXi, , Xen, XenServer, KVM, LXC, QEMU and Virtual Box
- 21. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPENSTACK The Boy Band of the Open Source Cloud All Things Open 2014 - Open Source Cloud Computing
- 22. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPENSTACK SHARED SERVICES Span Compute, Storage and Networking IDENTITY SERVICE IMAGE SERVICE All Things Open 2014 - Open Source Cloud Computing TELEMETRY SERVICE ORCHESTRATION SERVICE
- 23. EVEN MORE OPENSTACK PROJECTS Span Compute, Storage and Networking • Trove Database Service • Ironic Bare Metal (Ironic) • Marconi Queue Service By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com • Cinder Block Storage Service • Ceilometer Metering/Monitoring • Heat Orchestration All Things Open 2014 - Open Source Cloud Computing
- 24. OPENSTACK SOLUTION PROVIDERS If you can’t do it yourself “OpenStack is not a product. If you are building a large infrastructure, it’s more like a tool kit. It gives you a lot of technologies that do take a lot of effort to integrate.” Chris Kemp, OpenStack Board Member and Co-Founder By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing CEO of Piston Computing
- 25. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com CLOUD APIS Everything (should) have an API in the Cloud All Things Open 2014 - Open Source Cloud Computing • Deltacloud(ruby) • Daisein(java) • Jclouds(java) • Libcloud(python) • Fog(ruby)
- 26. Project Description Ceph Distributed file storage system developed by DreamHost -> GlusterFS Scale Out NAS system aggregating storage over Ethernet or Riak CS Riak CS is open source software designed to provide simple, available, distributed cloud storage at any scale. Riak CS is S3- API compatible and supports per-tenant reporting for billing and metering use cases. (object) Sheepdog Distributed storage for KVM hypervisors, distributed iSCSI By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com CLOUD STORAGE Virtualized, Distributed usually on Commodity Hardware InkTank -> Red Hat (block, object, file) Infiniband (file) OpenStack Storage Long-term object storage system (object) All Things Open 2014 - Open Source Cloud Computing
- 27. CLOUD AUTOMATION TOOLS One to many tools for managing large numbers of devices Ansible Ansible's SSH-key based access allows contributors to the Fedora Project to assist in automating infrastructure while having access limited appropriately. (Originally authored Func) Capistrano Utility and framework for executing commands in parallel on multiple remote machines, via SSH. It uses a simple DSL that allows you to define tasks, which may be applied to machines in certain roles RunDeck Rundeck is an open-source process automation and command orchestration tool with a web Func Func provides a two-way authenticated system for generically executing tasks, integrations with MCollective The Marionette Collective AKA MCollective is a framework to build server orchestration or Salt Execute arbitrary shell commands or choose from dozens of pre-built modules of common (or Scalr Provide scaling across multiple cloud computing platforms, integrates with Chef. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Project Description console. puppet and cobbler. parallel job execution systems. complex) commands. All Things Open 2014 - Open Source Cloud Computing
- 28. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing
- 29. Project Sponsors Languages/Frameworks By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com PLATFORM-AS-A-SERVICE Abstracted Cloud-Scale Run-Time Environments CloudFoundry VMware -> Pivotal -> CloudFoundry Foundation All Things Open 2014 - Open Source Cloud Computing Spring for Java, Ruby for Rails and Sinatra, node.js, Grails, Scala on Lift and more via partners (e.g. Python, PHP) Cloudify Gigaspaces [Groovy for deployment recipes] OpenShift Origin Red Hat Java, Ruby, PHP, Perl and Python Apache Stratos WSO2 - >Apache Stratus PHP, Tomcat, MySQL “cartridges”
- 30. Apache Mesos is a cluster manager that simplifies the complexity of running applications on a shared pool of servers. Largely supported by Twitter, used by LinkedIn, AirBNB too. Features • Fault-tolerant replicated master using ZooKeeper • Scalability to 10,000s of nodes • Isolation between tasks with Linux Containers • Multi-resource scheduling (memory and CPU aware) • Java, Python and C++ APIs for developing new By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com APACHE MESOS One to many tools for managing large numbers of devices parallel applications • Web UI for viewing cluster state To learn more please visit: http://mesos.apache.org/ All Things Open 2014 - Open Source Cloud Computing
- 31. Decoupling of the control and data planes of the network to improve efficiency. Communication from a SDN controller via a protocol to network devices both physical and virtual. Abstractions allow for programmable networks. Network can be changed quickly via a controller Network offerings can match virtualization offerings for finer grained security in a highly volatile compute landscape. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com SOFTWARE DEFINED VNirtuEalizTatiWon mOeetRs thKe neItwNorkG(SDN) Automation Dynamic Networks Security Heterogeneous Management Single control point for various devices. All Things Open 2014 - Open Source Cloud Computing
- 32. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Business Applications SDN OVERVIEW All Things Open 2014 - Open Source Cloud Computing Network Services SDN Control Software API API Network Devices Network Devices Network Devices Network Devices Network Devices Network Devices Application Layer Control Layer Infrastructure Layer Control Data Plane Interface (e.g. OpenFlow)
- 33. BENEFITS OF SDN Network Virtualization is the final frontier of Software Defined Datacenter By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com • Dynamically update networks • Automate network functionality • “Program” security into the network • Centrally apply policies to network and services • Optimize networks All Things Open 2014 - Open Source Cloud Computing
- 34. OpenFlow enables networks to evolve, by giving a remote controller the power to modify the behavior of network devices, through a well-defined "forwarding instruction set". The growing OpenFlow ecosystem now includes routers, switches, virtual switches, and access points from a range of vendors. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPENFLOW Virtualization meets the network All Things Open 2014 - Open Source Cloud Computing
- 35. OPEN SOURCE SDN Software Defined Network Controllers and more Floodlight The Floodlight Open SDN Controller is an enterprise-class, Apache-licensed, Java-based OpenFlow Controller. It is supported by a community of developers including a number of engineers from Big Switch Networks. - See more at: http://www.projectfloodlight.org/floodlight/#sthash.9IhA1Ih5.dpuf Indigo Indigo is an open source project aimed at enabling support for OpenFlow on physical and hypervisor switches. Big Switch has helped numerous companies OpenFlow enable their equipment, and we provide firmware for a number of popular switches. Indigo is the basis of Switch Light by Big Switch Networks. - See more at: http://www.projectfloodlight.org/indigo/#sthash.K7LiHcqc.dpuf Lincx LINCX is a pure OpenFlow software switch written in Erlang. It runs within a separate domain under Xen Nox NOX is the original OpenFlow controller, and facilitates development of fast C++ controllers on Linux. Open Daylight Linux Foundation Collaborative Project based on Cisco One Controller and plugins from numerous Open vSwitch Open vSwitch is a open source (ASL 2.0), multilayer virtual switch designed to enable massive network automation through programmatic extension, while still supporting standard management interfaces and protocols (e.g. NetFlow, sFlow, SPAN, RSPAN, CLI, LACP, 802.1ag). By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Project Description hypervisor using LING (erlangonxen.org). vendors in development. E.g IBM DOVE All Things Open 2014 - Open Source Cloud Computing
- 36. Open vSwitch is a production quality, multilayer virtual switch licensed under the open source Apache 2.0 license. It is designed to enable massive network automation through programmatic extension, while still supporting standard management interfaces and protocols (e.g. NetFlow, sFlow, SPAN, RSPAN, CLI, LACP, 802.1ag). To learn more please visit our website: http://openvswitch.org/ By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPEN VSWITCH All Things Open 2014 - Open Source Cloud Computing
- 37. DevOps Toolchain By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com OPEN SOURCE CLOUD STACK CloudFoundry, OpenShift, Gigaspaces Docker Platform-as-a-Service Mesos Kubernetes Infrastructure-as-a-Service | IaaS | Orchestration (OpenStack, Apache CloudStack, Eucalyptus) Compute (Containers, KVM, Xen) Storage (Ceph, Gluster) Networking (OpenDaylight, All Things Open 2014 - Open Source Cloud Computing Contrail) Orchestration - Ansible/SaltStack/Scalr* Configuration Management (CFengine/Chef/Puppet) Monitoring (logstash,graphite,)
- 38. EUREKA PRIAM SIMIAN ARMY By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com ASGARD ASTYANAX EDDA All Things Open 2014 - Open Source Cloud Computing 38 http://netflix.github.com NETFLIX AWS TOOLBAG Tools developed by a super Amazon Web Services Power User
- 39. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com CONTACT ME Happy to Chat about Open Source, Cloud or Pittsburgh Sports Professional: mark.hinkle@citrix.com Personal: mrhinkle@gmail.com Phone: 919.228.8049 Professional: http://open.citrix.com Personal: http://www.socializedsoftware.com Twitter: @mrhinkle All Things Open 2014 - Open Source Cloud Computing
- 40. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com APPENDIX A Additional Links to related stuff All Things Open 2014 - Open Source Cloud Computing
- 41. ADDITIONAL LINKS • Devops Toolchains Group • Software Defined Networking: The New Norm for Networks (Whitepaper) • DevOps Wikipedia Page • NoSQL-Database.org – Ultimate Guide to the Non-Relational Universe • Open Cloud Initiative • NIST Cloud Computing Platform • Open Virtualization Format Specs • Clouderati Twitter Account • Planet DevOps • Nicira Whitepaper – It’s Time to Virtualize the Network • Why Open vSwitch FAQ • Stanford Seminar - Software-Defined Networking at the Crossroads By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com All Things Open 2014 - Open Source Cloud Computing
- 42. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com ADDITIONAL LINKS (CONT’D) • SDN, NFV, and open source: The Operator’s View • Puppet Labs: Build a Toolbox for Continuous Delivery All Things Open 2014 - Open Source Cloud Computing
- 43. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com APPENDIX B Stuff I’d liked to have talked about but didn’t have time All Things Open 2014 - Open Source Cloud Computing
- 44. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com 60 SECOND CLOUD DEFINITION Just because Software Marketing Guys Think it’s the Internet 5 CHARACTERISTICS OF CLOUD 1. On-Demand Self-Service 2. Broad Network Access 3. Resource Pooling 4. Rapid Elasticity 5. Measured Service User Cloud a.k.a. SOFTWARE-AS-A-SERVICE Developer Cloud a.k.a. PLATFORM-AS-A-SERVICE Systems Cloud a.k.a. INFRASTRUCTURE-AS-A-SERVICE All Things Open 2014 - Open Source Cloud Computing
- 45. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com SCALE-UP SCALE OUT Elasticity and the cloud Vertical Scaling (Scale-Up) Allocate additional resources to VMs, requires a reboot, no need for distributed app logic, single-point of OS failure Horizontal Scaling (Scale-Out) Application needs logic to work in distributed fashion (e.g. HA-Proxy and Apache Hadoop) All Things Open 2014 - Open Source Cloud Computing
- 46. Bitnami BitNami provides free, ready to run environments for your favorite open source web applications and frameworks, including Drupal, Joomla!, Wordpress, PHP, Rails, Django and many more. Boxgrinder BoxGrinder is a set of projects that help you grind out appliances for multiple Oz Command-line tool that has the ability to create images for common Linux SUSE Studio SUSE Studio supports building and deploying directly to cloud services such as By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com SOURCING CLOUD APPLIANCES Packaging Engines for VMs Tool/Project What you can do with them virtualization and Cloud providers distributions to run on KVM Amazon EC2. All Things Open 2014 - Open Source Cloud Computing
- 47. PACKER MULTIPLATFORM VM CREATION Packer is easy to use and automates the creation of any type of machine image. It embraces modern configuration management by encouraging you to use automated scripts to install and configure the software within your Packer-made images. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com To learn more please visit: www.packer.io Open source Automation for VMs All Things Open 2014 - Open Source Cloud Computing
- 48. CONFIGURATION MANAGEMENT TOOLS Tools with features for configuring cloud infrastructure Project Year Started Language License Client/Server Chef 2009 Ruby Apache Chef Solo – No By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com CFengine 1993 C Apache Yes All Things Open 2014 - Open Source Cloud Computing Chef Server - Yes Puppet 2004 Ruby GPL Yes & standalone Salt 2011 Python Apache yes Hitchhiker’s Guide to the Open Cloud by @mrhinkle 48
- 49. CLOUD MONITORING TOOLS Tools with features for monitoring cloud infrastructure Project Type of Monitoring Collection Methods Cacti / RRDTool Performance SNMP, syslog Nagios Availability SNMP,TCP, ICMP, IPMI, By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Graphite Performance Agent All Things Open 2014 - Open Source Cloud Computing syslog Sensu Availability Agent Zabbix Availability/ Performance and more SNMP, TCP/ICMP, IPMI, Synthetic Transactions Zenoss Availability, Performance, Event Management SNMP, ICMP, SSH, syslog, WMI Hitchhiker’s Guide to the Open Cloud by @mrhinkle 49
- 50. CLOUD PROVISIONING TOOLS Packaging Engines for VMs Can provision 10s to 1000s of machines on various clouds. Cobbler Distributed virtual infrastructure using koan (kickstart of a network to PXE boot VMs) for Red Hat, OpenSUSE Fedora, Debian, Ubuntu VMs Salt Cloud Tool to provision “salted” VMs that can then be updated by a central server By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com Project Installation Targets Apache Provisionr (incubating) Crowbar (Bare metal provisioning) JuJu Public Clouds - Amazon Web Services HP Cloud, Private OpenStack clouds, Bare Metal via MAAS. via ZeroMQ All Things Open 2014 - Open Source Cloud Computing Hitchhiker’s Guide to the Open Cloud by @mrhinkle 50
- 51. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com BIG DATA All Things Open 2014 - Open Source Cloud Computing
- 52. API: many » Query Method: MapReduce, Replicaton: , Written in: Java, Concurrency: eventually consistent , Misc: like "Big-Table on Amazon Dynamo alike", initiated by Facebook CouchDB Document Store API: Memcached API+protocol (binary and ASCII) , most languages, Protocol: Memcached REST interface for cluster conf + management, Written in: C/C++ + Erlang (clustering), Replication: Peer to Peer, fully consistent, Misc: Transparent topology changes during operation, provides memcached-compatible caching buckets API: Java / any writer, Protocol: any write call, Query Method: MapReduce Java / any exec, Replication: HDFS Replication, Written in: Java PI: Thrift (Java, PHP, Perl, Python, Ruby, etc.), Protocol: Thrift, Query Method: HQL, native Thrift API, Replication: HDFS Replication, Concurrency: MVCC, Consistency Model: Fully consistent Misc: High performance C++ implementation of Google's Bigtable. MongoDB Document Store API: BSON, Protocol: C, Query Method: dynamic object-based language & MapReduce, Replication: Redis Key Value/ Tuple Store API: Tons of languages, Written in: C, Concurrency: in memory and saves asynchronous disk after a defined time. Append only mode available. Different kinds of fsync policies. Replication: Master / Slave, Misc: also lists, sets, sorted sets, hashes, queues. Riak Key Value / Tuple Store API: JSON, Protocol: REST, Query Method: MapReduce term matching , Scaling: Multiple Masters; Written in: Erlang, Concurrency: eventually consistent (stronger then MVCC via Vector Clocks) By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com NOSQL DATABASES Horizontally scalable unstructured data retrieval Name Type Description Apache Wide Column Cassandra Store/Families HBase Wide Column Store/Families Hypertable Wide Column Store/Families Master Slave & Auto-Sharding, Written in: C++,Concurrency All Things Open 2014 - Open Source Cloud Computing
- 53. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com MAP REDUCE Algorithm for Parallelized Data Set Processing Problem Data Master Node All Things Open 2014 - Open Source Cloud Computing Worker Node 1 Worker Node 2 Worker Node 3 Solution Data Map Reduce
- 54. By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com APACHE HADOOP Apache Project for Parallelized Data Set Processing Overview • Handles large amounts of data • Stores data in native format • Delivers linear scalability at low cost • Resilient in case of infrastructure failures • Transparent application scalability All Things Open 2014 - Open Source Cloud Computing Features • Handles large amounts of data • Stores data in native format • Delivers linear scalability at low cost • Resilient in case of infrastructure failures • Transparent application scalability
- 55. Machine Learning By Mark R. Hinkle @mrhinkle mrhinkle@gmail.com APACHE HADOOP ECOSYSTEM Non-Relational DB Hadoop Hadoop Common HDFS Distributes & replicates data across machines All Things Open 2014 - Open Source Cloud Computing MapReduce Distributes & monitors tasks Hive Data warehouse that provides SQL interface. Ad hoc projection of data structure to unstructured MapReduce • Parallel programming • Handles large data blocks HBase Column-oriented schema-less distributed DB modeled after Google’s BigTable Random real time read/write. Scripting Pig Platform for manipulating and analyzing large data sets. Scripting language for analysts. Mahout Machine learning libraries for recommendations , clustering, classifications and item sets. Chuckwa Zookeeper
Notes de l'éditeur
- Dashboard of Performance Openhub has a good graphical representation of code velocity and listing of developers – www.openhub.com Bitgeria Bitgeria does number of dashboards.
- Top choices for Cloud Computing are Xen and KVM. OpenVZ, container virtualization for Linux, is an interesting option as it has a very minimal overhead to scale application space similar to containers like BSD Jails. Advantage is that memory allocation is soft and unutilized memory can be used by other applications.
- Type 1 Hypervisor – VMware, Xen Project, XenServer, Hyper-V Type 1 (or native, bare metal) hypervisors run directly on the host's hardware to control the hardware and to manage guest operating systems. A guest operating-system thus runs on another level above the hypervisor. Type 2 Hypervisor – VM Type 2 (or hosted) hypervisors run within a conventional operating-system environment. With the hypervisor layer as a distinct second software level, guest operating-systems run at the third level above the hardware. VMware Workstation and VirtualBox exemplify Type 2 hypervisors.
- Image portability across hypervisors https://www.ibm.com/developerworks/community/blogs/9e696bfa-94af-4f5a-ab50-c955cca76fd0/entry/image_portability_across_hypervisors1?lang=en
- Martin Fowler - Continuous Integration http://www.martinfowler.com/articles/continuousIntegration.html
- What is Docker: http://www.docker.com/whatisdocker/ Common use cases for Docker include: Automating the packaging and deployment of applications Creation of lightweight, private PAAS environments Automated testing and continuous integration/deployment Deploying and scaling web apps, databases and backend services
- OpenStack Shared Services - https://www.openstack.org/software/openstack-shared-services/ Identity Service OpenStack Identity provides a central directory of users mapped to the OpenStack services they can access. It acts as a common authentication system across the cloud operating system and can integrate with existing backend directory services like LDAP. It supports multiple forms of authentication including standard username and password credentials, token-based systems and AWS-style logins. Image Service The OpenStack Image Service provides discovery, registration and delivery services for disk and server images. The ability to copy or snapshot a server image and immediately store it away is a powerful capability of the OpenStack cloud operating system. Stored images can be used as a template to get new servers up and running quickly and more consistently if you are provisioning multiple servers than installing a server operating system and individually configuring additional services. It can also be used to store and catalog an unlimited number of backups. Telemetry Service The OpenStack Telemetry service aggregates usage and performance data across the services deployed in an OpenStack cloud. This powerful capability provides visibility and insight into the usage of the cloud across dozens of data points and allows cloud operators to view metrics globally or by individual deployed resources. Orchestration Service OpenStack Orchestration is a template-driven engine that allows application developers to describe and automate the deployment of infrastructure. The flexible template language can specify compute, storage and networking configurations as well as detailed post-deployment activity to automate the full provisioning of infrastructure as well as services and applications. Through integration with the Telemetry service, the Orchestration engine can also perform auto-scaling of certain infrastructure elements.
- OpenStack Shared Services https://www.openstack.org/software/openstack-shared-services/ Identity Service OpenStack Identity provides a central directory of users mapped to the OpenStack services they can access. It acts as a common authentication system across the cloud operating system and can integrate with existing backend directory services like LDAP. It supports multiple forms of authentication including standard username and password credentials, token-based systems and AWS-style logins. Image Service The OpenStack Image Service provides discovery, registration and delivery services for disk and server images. The ability to copy or snapshot a server image and immediately store it away is a powerful capability of the OpenStack cloud operating system. Stored images can be used as a template to get new servers up and running quickly and more consistently if you are provisioning multiple servers than installing a server operating system and individually configuring additional services. It can also be used to store and catalog an unlimited number of backups. Telemetry Service The OpenStack Telemetry service aggregates usage and performance data across the services deployed in an OpenStack cloud. This powerful capability provides visibility and insight into the usage of the cloud across dozens of data points and allows cloud operators to view metrics globally or by individual deployed resources. Orchestration Service OpenStack Orchestration is a template-driven engine that allows application developers to describe and automate the deployment of infrastructure. The flexible template language can specify compute, storage and networking configurations as well as detailed post-deployment activity to automate the full provisioning of infrastructure as well as services and applications. Through integration with the Telemetry service, the Orchestration engine can also perform auto-scaling of certain infrastructure elements.
- Debate: How Many Open Source Platforms Are Enough? http://www.enterprisetech.com/2014/06/23/debate-many-open-source-platforms-enough/ OpenStack came in for the most criticism for issues such cost of deployment and maintenance. “OpenStack is not a product,” Kemp responded. “If you are building a large infrastructure, it’s more like a tool kit. It gives you a lot of technologies that do take a lot of effort to integrate.” The tool kit is used to create a product, Kemp stressed. OpenStack Vendors Canonical Ubuntu OpenStack - http://www.ubuntu.com/cloud/tools/openstack CloudScaling – Elastic Cloud Infrastructure - http://www.cloudscaling.com/ Elastic Cloud Infrastructure – built on OpenStack – enables any IT group to deploy cloud services comparable to the capabilities of the world’s largest and most successful public clouds. Cloudscaling solutions allow your organization to rapidly scale resources, achieve new levels of agility and improve market responsiveness. All with full control and governance in the privacy of your on-premise data center. HP Cloud OS - http://www8.hp.com/us/en/business-solutions/solution.html?compURI=1421776#.UzoD3K1dVDo Based on OpenStack technology, HP Cloud OS provides the foundation for the HP Cloud common architecture across private, public, and hybrid cloud delivery. Piston Cloud Computing - http://www.pistoncloud.com/openstack-cloud-software/ Piston OpenStack is a software product that uses advanced systems intelligence to orchestrate an entire private cloud environment using commodity hardware. Starting with an extremely lightweight custom Linux OS called Iocane Micro-OS™, and using an advanced high-availability system called Moxie Runtime Environment™, Piston keeps your cloud running no matter what – through hardware failure, operator error, upgrades, and power outages. Red Hat Distribution of OpenStack - http://openstack.redhat.com/Main_Page RDO is a community of people using and deploying OpenStack on Red Hat Enterprise Linux, Fedora and distributions derived from these (such as CentOS, Scientific Linux and others). We have documentation to help get started, forums where you can connect with other users, and community-supported packages of the most up-to-date OpenStack releases available for download. Rackspace Private Cloud powered by OpenStack - http://www.rackspace.com/cloud/private/
- Types of Tasks Accomplished by an API Provisioning (creating, re-creating, moving, or deleting components e.g. virtual machines, vlans) Configuration (assigning or changing attributes of the architecture such as security and network settings) Cloud Providers Daisein - Jclouds – java API Abstraction Libcloud – started by CloudKick (now Rackspace) to abstract clouds, Apache incubator project Deltacloud – started by Red Hat to abstract clouds, Apache incubator project Fog - provider and abstraction level API across compute and storage, written in Ruby
- Ansible Ansible's SSH-key based access allows contributors to the Fedora Project to assist in automating infrastructure while having access limited appropriately. Ansible is also used to roll out and manage clusters of machines and ISV software, such as Basho's flagship key-value store Riak. Capistrano Capistrano is a developer tool for deploying web applications. It is typically installed on a workstation, and used to deploy code from your source code management (SCM) to one, or more servers. Capistrano recently added classes capabilities that match cobbler. RunDeck RunDeck is cross-platform open source software that helps you automate ad-hoc and routine procedures in data center or cloud environments. RunDeck allows you to run tasks on any number of nodes from a web-based or command-line interface. RunDeck also includes other features that make it easy to scale up your scripting efforts including: access control, workflow building, scheduling, logging, and integration with external sources for node and option data. Func Func allows for running commands on remote systems in a secure way, like SSH, but offers several improvements. Func allows you to manage an arbitrary group of machines all at once. Func automatically distributes certificates to all "slave" machines. There's almost nothing to configure. Func comes with a command line for sending remote commands and gathering data. There are lots of modules already provided for common tasks. Anyone can write their own modules using the simple Python module API. Everything that can be done with the command line can be done with the Python client API. The hack potential is unlimited. You'll never have to use "expect" or other ugly hacks to automate your workflow. It's really simple under the covers. Func works over XMLRPC and SSL. Since func uses certmaster, any program can use func certificates, latch on to them, and take advantage of secure master-to-slave communication. There are no databases or crazy stuff to install and configure. Again, certificate distribution is automatic too. Mcollective The Marionette Collective AKA mcollective is a framework to build server orchestration or parallel job execution systems. Mcollective is used as a means of programmatic execution of Systems Administration actions on clusters of servers. MCollective use modern tools like Publish Subscribe Middleware and modern philosophies like real time discovery of network resources using meta data and not hostnames. Delivering a very scalable and very fast parallel execution environment. Scalr Scalr is a pretty darn good open source cloud management tool. It provides both an automation framework (do Foo when Bar) and a web interface (where is this volume mounted) for managing infrastructure on the cloud, like EC2. FEATURES * Integrated into Opscode Chef, for configuration management. * Pre-automated software, such as nginx, mysql, redis, mongo, and rabbitmq * Blazing fast UI * Multi-cloud * More at http://scalr.net/features/ ROADMAP * http://wiki.scalr.net/Roadmap
- Software Defined Networking (SDN) is an emerging network architecture where network control is decoupled from forwarding and is directly programmable. This migration of control, formerly tightly bound in individual network devices, into accessible computing devices enables the underlying infrastructure to be abstracted for applications and network services, which can treat the network as a logical or virtual entity. This figure depicts a logical view of the SDN architecture. Network intelligence is (logically) centralized in software-based SDN controllers, which maintain a global view of the network. As a result, the network appears to the applications and policy engines as a single, logical switch. With SDN, enterprises and carriers gain vendor-independent control over the entire network from a single logical point, which greatly simplifies the network design and operation. SDN also greatly simplifies the network devices themselves, since they no longer need to understand and process thousands of protocol standards but merely accept instructions from the SDN controllers.
- http://www.opendaylight.org/publications/sdn-nfv-and-open-source-operators-view
- Open Flow OpenFlow is an open standard that enables researchers to run experimental protocols in the campus networks we use every day. OpenFlow is added as a feature to commercial Ethernet switches, routers and wireless access points – and provides a standardized hook to allow researchers to run experiments, without requiring vendors to expose the internal workings of their network devices. OpenFlow is currently being implemented by major vendors, with OpenFlow-enabled switches now commercially available. In a classical router or switch, the fast packet forwarding (data path) and the high level routing decisions (control path) occur on the same device. An OpenFlow Switch separates these two functions. The data path portion still resides on the switch, while high-level routing decisions are moved to a separate controller, typically a standard server. The OpenFlow Switch and Controller communicate via the OpenFlow protocol, which defines messages, such as packet-received, send-packet-out, modify-forwarding-table, and get-stats. The data path of an OpenFlow Switch presents a clean flow table abstraction; each flow table entry contains a set of packet fields to match, and an action (such as send-out-port, modify-field, or drop). When an OpenFlow Switch receives a packet it has never seen before, for which it has no matching flow entries, it sends this packet to the controller. The controller then makes a decision on how to handle this packet. It can drop the packet, or it can add a flow entry directing the switch on how to forward similar packets in the future. OpenFlow is the first standard communications interface defined betweenthe control and forwarding layers of an SDN architecture. OpenFlow allows direct access to and manipulation of the forwarding plane of network devices such as switches and routers, both physical and virtual (hypervisor-based). It is the absence of an open interface to the forwarding plane that has led to the characterization of today’s networking devices as monolithic, closed, and mainframe-like. No other standard protocol does what OpenFlow does, and a protocol like OpenFlow is needed to move network control out of the networking switches to logically centralized control software
- Floodlight - http://www.projectfloodlight.org/floodlight/ - OpenFlow – works with physical- and virtual- switches that speak the OpenFlow protocol - Apache-licensed – lets you use Floodlight for almost any purpose Open community Floodlight is developed by an open community of developers. We welcome code contributions from active participants and we’ll openly share information on project status, roadmap, bugs, etc. Easy to Use- Floodlight is drop dead simple to build and run. Read through the Documentation (link) Tested and Supported – Floodlight is the core of a commercial controller product from Big Switch Networks (link) and is actively tested and improved by a community of professional developers. Indigo - http://www.projectfloodlight.org/indigo/ Indigo is an open source project aimed at enabling support for OpenFlow on physical and hypervisor switches. Big Switch has helped numerous companies OpenFlow enable their equipment, and we provide firmware for a number of popular switches. Indigo is the basis of Switch Light by Big Switch Networks. - See more at: http://www.projectfloodlight.org/indigo/#sthash.K7LiHcqc.dpuf Lincx - https://github.com/FlowForwarding/lincx LINCX is a pure OpenFlow software switch written in Erlang. It runs within a separate domain under Xen hypervisor using LING (erlangonxen.org). LINCX is a new faster version of LINC-Switch. Open Daylight – http://www.opendaylight.com The adoption of new technologies and pursuit of programmable networks has the potential to significantly improve levels of functionality, flexibility and adaptability of mainstream datacenter architectures. To leverage this abstraction to its fullest requires the network to adapt and evolve to a Software-Defined architecture. One of the architectural elements required to achieve this goal is a Software-Defined-Networking (SDN) platform that enables network control and programmability. Open vSwitch Open vSwitch is a production quality, multilayer virtual switch licensed under the open source Apache 2.0 license. It is designed to enable massive network automation through programmatic extension, while still supporting standard management interfaces and protocols (e.g. NetFlow, sFlow, SPAN, RSPAN, CLI, LACP, 802.1ag). In addition, it is designed to support distribution across multiple physical servers similar to VMware's vNetwork distributed vswitch or Cisco's Nexus 1000V. See the full feature list here
- Why Open vSwitch - http://git.openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=blob_plain;f=WHY-OVS;hb=HEAD Hypervisors need the ability to bridge traffic between VMs and with theoutside world. On Linux-based hypervisors, this used to mean using thebuilt-in L2 switch (the Linux bridge), which is fast and reliable. So, it is reasonable to ask why Open vSwitch is used. The answer is that Open vSwitch is targeted at multi-server virtualization deployments, a landscape for which the previous stack is not well suited. These environments are often characterized by highly dynamic end-points, the maintenance of logical abstractions, and (sometimes) integration with or offloading to special purpose switching hardware.
- NetFlix AWS Toolbag – http://netflix.github.com Over 25 projects developed by NetFlix to manager their cloud deployments. Asgard Asgard is a web-based tool for managing cloud-based applications and infrastructure. Astyanaz Astyanax is a high level Java client for Apache Cassandra. Apache Cassandra is a highly available column oriented database. Edda Edda is a Service to track changes in your cloud deployments. Eureka Eureka is a REST (Representational State Transfer) based service that is primarily used in the AWS cloud for locating services for the purpose of load balancing and failover of middle-tier servers. At Netflix, Eureka is used for the following purposes apart from playing a critical part in mid-tier load balancing. For aiding Netflix Asgard - an open source service which makes cloud deployments easier, in Fast rollback of versions in case of problems avoiding the re-launch of 100's of instances which could take a long time. In rolling pushes, for avoiding propagation of a new version to all instances in case of problems. For our cassandra deployments to take instances out of traffic for maintenance. For our memcached caching services to identify the list of nodes in the ring. Priam Priam is a process/tool that runs alongside Apache Cassandra to automate the following: - Backup and recovery (Complete and incremental) - Token management - Seed discovery - Configuration Support AWS environment Simian Army The Simian Army is a suite of tools for keeping your cloud operating in top form. Chaos Monkey, the first member, is a resiliency tool that helps ensure that your applications can tolerate random instance failures
- Salt - https://github.com/saltstack/salt
- Cacti Cacti is a complete network graphing solution designed to harness the power of RRDTool's data storage and graphing functionality. Cacti provides a fast poller, advanced graph templating, multiple data acquisition methods, and user management features out of the box. All of this is wrapped in an intuitive, easy to use interface that makes sense for LAN-sized installations up to complex networks with hundreds of devices. RRDTool RRDtool is the OpenSource industry standard, high performance data logging and graphing system for time series data. RRDtool can be easily integrated in shell scripts, perl, python, ruby, lua or tcl applications. Graphite Graphite is a highly scalable real-time graphing system. As a user, you write an application that collects numeric time-series data that you are interested in graphing, and send it to Graphite's processing backend, carbon, which stores the data in Graphite's specialized database. The data can then be visualized through graphite's web interfaces.
- These tools are all appropriate for Linux guest operating systems, Windows operating system provisioning is not well addressed in OSS. Axembler Provisonr Provisionr solves the problem of cloud portability by hiding completely the APIs and only focusing on building a cluster that matches the same set of assumptions on all clouds, assumptions like: a specific OS, pre-installed packages and binaries, sane dns settings, ssh & vpn access etc. - think a solid foundation for configuration. As a secondary goal Provisionr will also provide primitives for building automatic or semi-automatic workflows for configuring and monitoring services, workflows that assume that all the machines share a common set of characteristics as described above. Cobbler Cobbler is a Linux installation server that allows for rapid setup of network installation environments. It glues together and automates many associated Linux tasks so you do not have to hop between lots of various commands and applications when rolling out new systems, and, in some cases, changing existing ones. With a simple series of commands, network installs can be configured for PXE, reinstallations, media-based net-installs, and virtualized installs (supporting Xen, qemu, KVM, and some variants of VMware). Cobbler uses a helper program called 'koan' (which interacts with Cobbler) for reinstallation and virtualization support. Crowbar Bare metal provisioning for CloudStack developed by Dell using Opscode Chef. Juju Metal as a Service (MAAS) MAAS offers a nice UI to provision your Ubuntu servers. Each physical server (“node”) will be commissioned automatically on first boot. During the commissioning process administrators are able to configure hardware settings manually before an automated smoke test and burn-in test are done. Once commissioned, a node can be deployed on demand by name, or allocated to a queue for dynamic allocation to services being deployed on this MAAS. Salt Cloud Salt Cloud is a tool for provisioning salted minions across various cloud providers. Currently supported providers are: - Amazon EC2 - GoGrid - HP Cloud (using OpenStack) - Joyent - Linode - OpenStack - Rackspace (using OpenStack) The salt-cloud command can be used to query configured providers, create VMs on them, deploy salt-minion on those VMs and destroy them when no longer needed. Salt Cloud requires Salt to be installed, but does not require any Salt daemons to be running. However, if used in a salted environment, it is best to run Salt Cloud on the salt-master, so that it can properly lay down salt keys when it deploys machines, and then properly remove them later. If Salt Cloud is run in this manner, minions will automatically be approved by the master; no need to manually authenticate them later. Deprecated Spacewalk Spacewalk manages software content updates for Red Hat derived distributions such as Fedora, CentOS, and Scientific Linux, within your firewall. You can stage software content through different environments, managing the deployment of updates to systems and allowing you to view at which update level any given system is at across your deployment. A clean central web interface allows viewing of systems and their software update status, and initiating update actions.
- Big data the term for a collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools or traditional data processing applications. The challenges include capture, curation, storage, search, sharing, transfer, analysis[4] and visualization. The trend to larger data sets is due to the additional information derivable from analysis of a single large set of related data, as compared to separate smaller sets with the same total amount of data, allowing correlations to be found to "spot business trends, determine quality of research, prevent diseases, link legal citations, combat crime, and determine real-time roadway traffic conditions.
- NoSQL In computing, NoSQL (commonly interpreted as "not only SQL"[1]) is a broad class of database management systems identified by non-adherence to the widely used relational database management system model. NoSQL databases are not built primarily on tables, and generally do not use SQL for data manipulation. NoSQL database systems are often highly optimized for retrieval and appending operations and often offer little functionality beyond record storage (e.g. key–value stores). The reduced run-time flexibility compared to full SQL systems is compensated by marked gains in scalability and performance for certain data models. Apache Cassandra The Apache Cassandra database is the right choice when you need scalability and high availability without compromising performance. Linear scalability and proven fault-tolerance on commodity hardware or cloud infrastructure make it the perfect platform for mission-critical data. Cassandra's support for replicating across multiple datacenters is best-in-class, providing lower latency for your users and the peace of mind of knowing that you can survive regional outages. Cassandra's ColumnFamily data model offers the convenience of column indexes with the performance of log-structured updates, strong support for materialized views, and powerful built-in caching. Cassandra is in use at Netflix, Twitter, Urban Airship, Constant Contact, Reddit, Cisco, OpenX, Digg, CloudKick, Ooyala, and more companies that have large, active data sets. The largest known Cassandra cluster has over 300 TB of data in over 400 machines. Hypertable Hypertable is based on a design developed by Googl(e.g. BigTable clone) to meet their scalability requirements and solves the scale problem better than any of the other NoSQL solutions out there. Mongo DB MongoDB (from "humongous") is a cross-platform document-oriented database system. Redis Redis is an open source, BSD licensed, advanced key-value store. It is often referred to as a data structure server since keys can contain strings, hashes, lists, sets and sorted sets. Riak Riak is known for its ability to distribute data across nodes using consistent hashing in a simple key/value scheme in namespaces called buckets.
- MapReduce is a programming model for processing large data sets with a parallel, distributed algorithm on a cluster. A MapReduce program is composed of a Map() procedure that performs filtering and sorting (such as sorting students by first name into queues, one queue for each name) and a Reduce() procedure that performs a summary operation (such as counting the number of students in each queue, yielding name frequencies). The "MapReduce System" (also called "infrastructure" or "framework") orchestrates by marshalling the distributed servers, running the various tasks in parallel, managing all communications and data transfers between the various parts of the system, and providing for redundancy and fault tolerance.