In Place Analytics For File and Object Data
Download as PPTX, PDF0 likes669 views
The document discusses IBM Spectrum Scale's unified file and object access feature. It introduces Spectrum Scale and its support for file and object access. The unified file and object access feature allows data to be accessed as both files and objects without copying, through a single management plane. Use cases like in-place analytics for object data and common identity management across file and object access are enabled. A demo is presented where a file is uploaded as an object, analytics is run on it, and the result downloaded as an object, without data movement.
In Place Analytics For File and Object Data
- 1. #ibmedge© 2016 IBM Corporation Analytics for Object Storage Simplified - Spectrum Scale Object Storage with File Access for Hadoop Sandeep Patil, STSM, IBM Spectrum Scale Tomer Perry, Solution Architect, IBM Spectrum Scale Smita Raut , Object Development, IBM Spectrum Scale Acknowledgement : Bill Owen, Ashutosh Mate, Shou Feng, John Gu, Yong Zeng, Piyush Chaudhary, Wei Gong
- 2. #ibmedge Agenda • Introduction to Spectrum Scale • Introduction to Spectrum Scale Analytics • Introduction to Spectrum Scale Object Store • Unified File & Object Access (UFO) Feature Details • Use Cases Enabled By UFO • Deep Dive of In-Place Analytics Use Case • Demo • Q & A 1
- 3. #ibmedge 2
- 4. #ibmedge 3
- 5. #ibmedge 4
- 6. #ibmedge 5
- 7. #ibmedge 6
- 8. #ibmedge 7
- 9. #ibmedge Spectrum Scale Analytics Introduction 8
- 10. #ibmedge 9
- 11. #ibmedge 10
- 12. #ibmedge 11
- 13. #ibmedge 12
- 14. #ibmedge 13
- 15. #ibmedge 14
- 16. #ibmedge GPFS-FPO Advanced Storage for Map Reduce Data 15 Hadoop HDFS IBM GPFS Advantages HDFS NameNode is a single point of failure Large block-sizes – poor support for small files Non-POSIX file system – obscure commands Difficulty to ingest data – special tools required Single-purpose, Hadoop MapReduce only Not recommended for critical data No single point of failure, distributed metadata Variable block sizes – suited to multiple types of data and data access patterns POSIX file system – easy to use and manage Policy based data ingest Versatile, Multi-purpose Enterprise Class advanced storage features
- 17. #ibmedge Use Case: Big Data Analytics • Problem: Separate storage systems for ingest/distribution and analysis • Data movement overhead is a significant part of my time to insight. • Increased cost from data duplication & overhead • Inconsistent results • Solution: Native HDFS support • Decreased time to results • Run Map/Reduce directly • No waiting for data transfer between storage systems • Immediately share results 16 Spectrum Scale File/ ObjectFile/HDFS Global Ingest and Distribution Business Analytics Custom Applications Packaged Applications
- 18. #ibmedge Spectrum Scale Object Storage Introduction 17
- 19. #ibmedge IBM Spectrum Scale • Avoid vendor lock-in with true Software Defined Storage and Open Standards • Seamless performance & capacity scaling • Automate data management at scale • Enable global collaboration Data management at scale OpenStack and Spectrum Scale helps clients manage data at scale Business: I need virtually unlimited storage Operations: I need a flexible infrastructure that supports both object and file based storage Operations: I need to minimize the time it takes to perform common storage management tasks Collaboration: I need to share data between people, departments and sites with low latency. A single data plane that supports Cinder, Glance, Swift, Manila as well as NFS, et. al. A fully automated policy based data placement and migration tool An open & scalable cloud platform Sharing with a variety of WAN caching modes Results • Converge File and Object based storage under one roof • Employ enterprise features to protect data, e.g. Snapshots, Backup, and Disaster Recovery • Support native file, block and object sharing to data. Spectrum Scale NFS SMBPOSIX SSD Fast Disk Slow Disk Tape Swift HDFS Cinder Glance Manila Cognitive Services 18
- 20. #ibmedge Spectrum Scale Object Storage • Basic support added in 4.1.1 release & enhanced in 4.2 and 4.2.1 release • Based on Openstack Swift (Juno Release) • REST-based data access • Growing number of clients due to extremely simple protocol • Applications can easily save & access data from anywhere using HTTP • Simple set of atomic operations: – PUT (upload) – POST (update metadata) – GET (download) – DELETE • Amazon S3 Protocol support • High Availability with CES Integration • Simple and Automated Installation Process • Integrated authentication (Keystone) support • Native GPFS Command Line Interface to manage Object service (mmobj command) 19
- 21. #ibmedge Spectrum Scale Object Store – Additional Features • Unified file and object support with Hadoop connectors • Support for Encryption • Support for Compression • Only Object Store with Tape support for Backup • Object store with integrated transparent cloud tiering Support • Multi Region support • AD/LDAP support for authentication • ILM support for Object • Movement of Object across storage tiers based on access heat • Spectrum Scale Object with IBM DeepFlash becomes object store over all flash array for newer faster workloads. • Spectrum Scale Object with WAN caching support (AFM) 20
- 22. #ibmedge Spectrum Scale Object Vs Cleversafe 21
- 23. #ibmedge The right solution for the workload 22 Ideal Workloads • Big Data Analytics • High Performance Computing, e.g. Engineering Applications • Performance optimized Backup and Restore • Multi-Site file collaboration • Multi-tier File Synch and Share • Cold data archive with lowest cost data storage tier Differentiation • Designed for high performance • Unified Storage Infrastructure: Native File, Object & Hadoop • Robust Tiering with policy based data placement and data movement • Multi site collaboration with advanced routing and caching • Enterprise Features, e.g. Encryption, Compression, QoS, & Disaster Recovery Ideal Workloads • Active Archive (warm data, mostly static) • Cost optimized Cloud backup target • Web app content • Remote office storage consolidation • Storage as a service Differentiation • Designed for easy deployment and management at scale • Always-on architecture • Geo-dispersed erasure coding for site fault tolerance and DR • Simple keyless native encryption and multi- tenant security • Reduced cost and complexity Spectrum Scale IBM Cloud Object Store (Cleversafe)
- 24. #ibmedge IBM Spectrum Scale: Unified File and Object Access Feature Overview 23
- 25. #ibmedge Unified File & Object (UFO) Support • Challenge • The world is not converged/file/object/HDFS today! • and never will be completely… • Unified Scale-out Content Repository • File or object in. Object or file out. • Integrated big data analytics support • Native protocol support • High-performance that scales • Single Management Plane 24 Spectrum Scale NFS SMBPOSIX SSD Fast Disk Slow Disk Tape Swift/S3HDFS Spectrum Scale: Redefining Unified Storage
- 26. #ibmedge Spectrum Scale Unified File & Object • Access same content both as a File & as an Object without making a copy or needing File or Object Gateways! • File-In-Object-Out and Object-In-File-Out Support • Support for File Access Protocols (NFS/SMB/POSIX) and Object Access Protocols (Swift/S3) • Objects ingested into designated Unified Container available as Files and Files ingested into it available as Objects. • Support for File & Object ACLs with Unified Mode ID Mapping 25
- 27. #ibmedge Unified File and Object Access – What is it ? 26
- 28. #ibmedge What is Unified File and Object Access ? • Accessing object using file interfaces (SMB/NFS/POSIX) and accessing file using object interfaces (REST) helps legacy applications designed for file to seamlessly start integrating into the object world. • It allows object data to be accessed using applications designed to process files. It allows file data to be published as objects. • Multi protocol access for file and object in the same namespace (with common User ID management capability) allows supporting and hosting data oceans of different types of data with multiple access options. • Optimizes various use cases and solution architectures resulting in better efficiency as well as cost savings. 27 <Clustered file system> Swift (With Swift on File) NFS/SMB/POSIXObject(http) 2 1 <Container> File Exports created on container level OR POSIX access from container level Objects accessed as FilesData ingested as Objects 3 Data ingested as Files4 Files accessed as Objects
- 29. #ibmedge Flexible Identity Management Modes • Support’s Two Identity Management Modes • Administrators can choose based on their need and use-case using CLI --------------> 28 #mmobj config change --ccrfile object-server-sof.conf -- section DEFAULT --property id_mgmt --value unified_mode | local_mode Local_Mode Unified_Mode Identity Management Modes Object created by Object interface will be owned by internal “swift” user Application processing the object data from file interface will need the required file ACL to access the data. Object authentication setup is independent of File Authentication setup Object created from Object interface should be owned by the user doing the Object PUT (i.e FILE will be owned by UID/GID of the user) Users from Object and File are expected to be common auth and coming from same directory service (only AD+RFC 2307 or LDAP) Owner of the object will own and have access to the data from file interface. Suitable for unified file and object access for end users. Leverage common ILM policies for file and object data based on data ownership Suitable when auth schemes for file and object are different and unified access is for applications
- 30. #ibmedge Use Cases Enabled by Unified File Object 29
- 31. #ibmedge Use case 1 – Enabling “In-Place” analytics for Object data repository with analytic results available as objects 30 Clustered file system <SOF_Fileset>/<Device> Object (http) Data ingested as Objects Spark or Hadoop MapReduce In-Place Analytics Source:https://aws.amazon.com/elasticmapreduce/ Traditional object store – Data to be copied from object store to dedicated cluster , do the analysis and copy the result back to object store for publishing Object store with Unified File and Object Access – Object Data available as File on the same fileset. Analytics systems like Hadoop MapReduce or Spark allow the data to be directly leveraged for analytics. No data movement i.e. In-Place immediate data analytics. Analytics With Unified File and Object AccessAnalytics on Traditional Object Store Explicit Data movement Results Published as Objects with no data movement Results returned in place
- 32. #ibmedge Use case 2 : Process Object Data with File-Oriented Applications and Publish Outcomes as Objects 31 Swift on file Container1 Virtual Machine Instances Virtual Machine Instances Container2 Subsidiary 1 Subsidiary 2 NFS Export on Container 1 NFS Export on Container 2 Virtual Machine Instances Virtual Machine Instances VM Farm for Subsidiary 1 for video processing VM Farm for Subsidiary 2 for video processing …. …. Ingest Media Objects Media House OpenStack Cloud Platform (Tenant = Media House Subsidiaries) Manila Shares (NFS) exported only for Subsidiary1 Publishing Channels Final Video (as objects) available for streaming Final processed videos available as Objects in container which is used for external publishing Raw media content sent for media processing which happens over files (Object to File access) NFS Export on Container 1’ Container 1’ Manila Shares (NFS) exported only for Subsidiary2 Files converted into objects for publishing (File to Object access)
- 33. #ibmedge Use case 3 : Users read/write data via File and Object with Common User Authentication and Identity 32 Clustered file system Data N F S S M B O b je c t Data N F S S M B O b je c t User: John User: Riya Access Common Data using the same User Credentials across all protocols Corporate User Directory (Active Directory/LDAP) Riya’s data Read/Written from Object should be owned by Riya when accessed from File (SMB/NFS/POSIX) User: Riya UID: 1001 GID: 2000 Domain: XYZ
- 34. #ibmedge Deep Dive on In-Place Analytics Use Case 33
- 35. #ibmedge Analytics use case 34 What is In-place Analytics ? What is unified storage ? Setup Details Prerequisites for demo Demo content
- 36. #ibmedge What is In-place Analytics ? 35
- 37. #ibmedge Setup Details 36 /dev/dm-3 viknode1 Roles – Admin, quorum, NSD viknode2 Roles – Quorum, NSD, CES Node viknode3 Roles – Quorum, CES Node Spectrum Scale Cluster IBM BigInsight with Spectrum Scale Demo Setup /dev/dm-2 Disks Ambari Server IBM BigInsightsYarn Spark HiveOozie Slider Knox
- 38. #ibmedge Prerequisites For Demo 37 Setup a Spectrum Scale Cluster with NFS, SMB and Object Protocols Enabled Setup same authentication for File and Object Enable unified access mode Enable file access capabilities Create a swift storage policy with File access enabled Install BigInsights Start Ambari server Configure Cyberduck Client to access object store
- 39. #ibmedge Demo Content 38 Upload a file using Cyberduck Run Analytics on file Download result using Cyberduck
- 40. #ibmedge Demo
- 41. #ibmedge Spectrum Scale User Group • The Spectrum Scale User Group is free to join and open to all using, interested in using or integrating Spectrum Scale. • Join the User Group activities to meet your peers and get access to experts from partners and IBM. • Next meetings: - APAC: October 14, Melbourne - Global at SC16 : November 13 1pm to 5pm, Salt Lake City • Web page: http://www.spectrumscale.org/ • Presentations: http://www.spectrumscale.org/presentations/ • Mailing list: http://www.spectrumscale.org/join/ • Contact: http://www.spectrumscale.org/committee/ • Meet Bob Oesterlin (US Co-Principal) at Edge2016: [email protected]
- 42. #ibmedge Session : Futures of IBM Spectrum Scale NDA & Customers ONLY • Who: IBM Spectrum Scale Offering Management • Carl Zetie, Ron Riffe • When: Tuesday, September 20, 2016 • 1pm to 2pm • Where: MGM Grand, Signature Tower 3 • Meeting Room D • Contact (if any questions) • [email protected], [email protected] 41
- 43. #ibmedge Session : How to apply Flash benefits to big data analytics and unstructured data NDA & Customers ONLY • Who: IBM Elastic Storage Server Offering Management • Alex Chen • When: Thursday, September 22, 2016 • 1:15pm to 2:15pm • Where: Grand Garden Arena, Lower Level, MGM, Studio 10 • Contact(if any questions) • • [email protected], [email protected] 42
- 44. #ibmedge Trial VM • Download the IBM Spectrum Scale Trial VM from : http://www- 03.ibm.com/systems/storage/spectrum/scale/trial.html 43
- 45. #ibmedge References Write a File, read as an Object: Openstack Summit, Austin, TX Apr 2016 https://www.youtube.com/watch?v=6ovLb6aktbM&feature=youtu.be&t=2 Amalgamating Manila and Swift for Unified Data Sharing: Openstack Summit, Austin, TX Apr 2016 https://www.youtube.com/watch?v=3MMrMUaA_Mg Hadoop HDFS Vs Spectrum Scale: https://www.youtube.com/watch?v=kOeEbdO8F4A From Archive to Insight: Debunking Myths of Analytics on Object Stores – Dean Hildebrand, Bill Owen, Simon Lorenz, Luis Pabon, Rui Zhang. Vancouver Summit, Spring 2015. https://www.youtube.com/watch?v=brhEUptD3JQ Deploying Swift on a File System – Bill Owen, Thiago Da Silva. BrownBag at OpenStack Paris, Fall 2014 https://www.youtube.com/watch?v=vPn2uZF4yWo Breaking the Mold with OpenStack Swift and GlusterFS – Jon Dickinson, Luis Pabo. Atlanta Summit, Spring 2014 https://www.youtube.com/watch?v=pSWdzjA8WuA SNIA SDC 2015 http://www.snia.org/sites/default/files/SDC15_presentations/security/DeanHildebrand_Sasi__OpenStack%20SwiftOnFile.pdf Spectrum Scale Infocenter http://www.ibm.com/support/knowledgecenter/#!/STXKQY_4.2.0/com.ibm.spectrum.scale.v4r2.adm.doc/bl1adm_manageunifieda ccess.htm 44
- 46. #ibmedge OpenStack Summit 2016: IBM Spectrum Scale in an OpenStack Environment Redpaper Published. 45 http://www.redbooks.ibm.com/abstracts/redp5331.html
- 47. © 2016 IBM Corporation #ibmedge Thank You
- 48. #ibmedge IBM Spectrum Scale - Unified File and Object Access Feature Overview • Multi protocol access for file and object in the same namespace • Access object as file from POSIX, NFS and SMB • Access file as object – Provision to convert files to object automatically via background service called ‘objectizer’ – Provision to explicitly and immediately convert files to objects using CLI • Feature is specifically made available as an “object storage policy” • Allows to coexists with traditional object and other policies • Create multiple unified file and object access policies • Since policies are applicable per container , it gives end user the flexibility to create certain containers with Unified File and Object Access policy and certain without it. Flexible Identity Management Mode Support • Local Mode: Suitable when auth schemes for file and object are different and unified access is for applications • Object created by Object interface will be owned by internal “swift” user • Unified Mode: Suitable for unified file and object access by end users. Leverage common ILM policies for file and object data based on data ownership. • Object created from Object interface should be owned by the user doing the Object PUT (i.e. FILE will be owned by UID/GID of the user) • Ability to run in-place analytics of object data using Spectrum Scale Hadoop connectors via POSIX interface. 47
- 49. #ibmedge Filesystem Layout (Traditional Vs Unified File and Object Access) • One of the key advantages of unified file and object access is the placement and naming of objects when stored on the file system. In unified file and object access stores objects following the same path hierarchy as the object's URL. • In contrast, the default object implementation stores the object following the mapping given by the ring, and its final file path cannot be determined by the user easily. 48 ibm/gpfs0/ Object ingest object_fileset/ o/z1device108/objects/7551/125 75fc66179f12dc513580a239e92c3125 a.jpg a.jpg Object ingest ibm/gpfs0/ <Sof_policy_fileset>/<device>/ AUTH_acctID/cont/ a.jpg Traditional SWIFT Unified File and Object Access Ingest object URL: https://swift.example.com/v1/acct/cont/a.jpg
- 50. #ibmedge Easy Access Of Objects as Files via supported File Interfaces (NFS/SMB/POSIX) • Objects ingested are available immediately for File access via the 3 supported file protocols. • ID management modes (explained later) gives flexibility of assigning/retaining of owners, generally required by file protocols. • Object authorization semantics are used during object access and file authorization semantics are used during file access of the same data – thus ensuring compatibility of object and file applications 49 <Spectrum Scale Filesystem> <SOF_Fileset>/<Device> NFS/SMB/POSIXObject (http) 2 1 <AUTH_account_ID> <Container> File Exports created on container level OR POSIX access from container level Objects accessed as Files Data ingested as Objects
- 51. #ibmedge Objectization – Making Files as Objects (Accessing File via Object interface) • Spectrum Scale 4.2 features with a system service called ibmobjectizer responsible for objectization. • Objectization is a process that converts files ingested from the file interface on unified file and object access enabled container path to be available from the object interface. • When new files are added from the file interface, they need to be visible to the Swift database to show correct container listing and container or account statistics. 50 Spectrum Scale Filesystem Unified File and Object Fileset NFS/SMB/POSIXObject (http) ibmobjectizer objectization 1 2 3 Data ingested as Files Files accessed as Objects
- 52. #ibmedge Unified File and Object Access – Policy Integration for Flexibility • This feature is specifically made available as an “object storage policy” as it gives the following advantages: • Flexibility for administrator to manage unified file and object access separately • Allows to coexists with traditional object and other policies • Create multiple unified file and object access policies which can vary based on underlying storage • Since policies are applicable per container , it gives end user the flexibility to create certain containers with Unified File and Object Access policy and certain without it. • Example: mmobj policy create SwiftOnFileFS --enable-file-access 51
- 53. #ibmedge Notices and Disclaimers 52 Copyright © 2016 by International Business Machines Corporation (IBM). No part of this document may be reproduced or transmitted in any form without written permission from IBM. U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM. Information in these presentations (including information relating to products that have not yet been announced by IBM) has been reviewed for accuracy as of the date of initial publication and could include unintentional technical or typographical errors. IBM shall have no responsibility to update this information. THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY, EITHER EXPRESS OR IMPLIED. IN NO EVENT SHALL IBM BE LIABLE FOR ANY DAMAGE ARISING FROM THE USE OF THIS INFORMATION, INCLUDING BUT NOT LIMITED TO, LOSS OF DATA, BUSINESS INTERRUPTION, LOSS OF PROFIT OR LOSS OF OPPORTUNITY. IBM products and services are warranted according to the terms and conditions of the agreements under which they are provided. IBM products are manufactured from new parts or new and used parts. In some cases, a product may not be new and may have been previously installed. Regardless, our warranty terms apply.” Any statements regarding IBM's future direction, intent or product plans are subject to change or withdrawal without notice. Performance data contained herein was generally obtained in a controlled, isolated environments. Customer examples are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual performance, cost, savings or other results in other operating environments may vary. References in this document to IBM products, programs, or services does not imply that IBM intends to make such products, programs or services available in all countries in which IBM operates or does business. Workshops, sessions and associated materials may have been prepared by independent session speakers, and do not necessarily reflect the views of IBM. All materials and discussions are provided for informational purposes only, and are neither intended to, nor shall constitute legal or other guidance or advice to any individual participant or their specific situation. It is the customer’s responsibility to insure its own compliance with legal requirements and to obtain advice of competent legal counsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customer’s business and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the customer is in compliance with any law
- 54. #ibmedge Notices and Disclaimers Con’t. 53 Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products in connection with this publication and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM does not warrant the quality of any third-party products, or the ability of any such third-party products to interoperate with IBM’s products. IBM EXPRESSLY DISCLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. The provision of the information contained h erein is not intended to, and does not, grant any right or license under any IBM patents, copyrights, trademarks or other intellectual property right. IBM, the IBM logo, ibm.com, Aspera®, Bluemix, Blueworks Live, CICS, Clearcase, Cognos®, DOORS®, Emptoris®, Enterprise Document Management System™, FASP®, FileNet®, Global Business Services ®, Global Technology Services ®, IBM ExperienceOne™, IBM SmartCloud®, IBM Social Business®, Information on Demand, ILOG, Maximo®, MQIntegrator®, MQSeries®, Netcool®, OMEGAMON, OpenPower, PureAnalytics™, PureApplication®, pureCluster™, PureCoverage®, PureData®, PureExperience®, PureFlex®, pureQuery®, pureScale®, PureSystems®, QRadar®, Rational®, Rhapsody®, Smarter Commerce®, SoDA, SPSS, Sterling Commerce®, StoredIQ, Tealeaf®, Tivoli®, Trusteer®, Unica®, urban{code}®, Watson, WebSphere®, Worklight®, X-Force® and System z® Z/OS, are trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at: www.ibm.com/legal/copytrade.shtml.