1 / 26

19. April 2013

19. April 2013. Toine Beckers. tbeckers@ddn.com. Big Data Evolution. Top500 – Storage, Nov 2012. Accelerating Accelerators. DDN is the leading provider of affordable, high-availability storage for the next generation of particle physics research.

ivan
Download Presentation

19. April 2013

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. 19. April 2013 Toine Beckers tbeckers@ddn.com Big Data Evolution

  2. Top500 – Storage, Nov 2012

  3. Accelerating Accelerators DDN is the leading provider of affordable, high-availability storage for the next generation of particle physics research. DDN Supplied over 40PB of Storage to the LHC Community in the last 4 years

  4. LHC Customer Base • Tier 0 • CERN-LHCb (1*S2A9900 SFA10K, 100TB) • Tier 1 • SARA/NIKHEF (13*S2A9900, 6 PB) • KIT (10*S2A9900, 1*SFA10K, 20PB) • IN2P3 (7*DCS9550, 1.5PB) • PIC (2*S2A9900, 2.4PB) • INFN-CNAF (5*S2A9900, 1*SFA10K, 10PB) • Tier 2 • DESY (2*S2A9900, 2*SFA10K, 3PB) • NBI (1*S2A6620, 60*2TB) • INFN-PISA (2*S2A9900, 1*SFA12K, 1PB) • INFN-PADOVA (1*S2A9900, 240TB) • IFCA (1*S2A9900, 1.2PB) • TRIUMF (2*DCS9900, 0.6PB) • SFU (1*S2A9900, 1PB) • UNIV. ALBERTA (1*S2A9550, 100TB) • UNIV. VICTORIA (1*S2A9900, 500TB) • SCINET (2*S2A9900, 1PB) • McGill UNIV. (2*SFA10K, 1PB)

  5. DDN Portfolio Block Embedded systems Parallel File Storage Cloud Tiering Analytics Filesystems, customer applications EXAScaler™ 10Ks of Clients 1TB/s+, HSM NFS, CIFS GRIDScaler™ 1Ks of Clients 1TB/s+, HSM NFS, CIFS Cloud Storage Storage Fusion Architecture Storage Appliances 7700 12K 10GB/s, 600K IOPS 60 Drives in 4U; 396 Drives in 20U Embedded Computing (tba) 40GB/s/1.7M IOPS 1,680 Drives: 2 Racks Embedded Computing DirectMon Enterprise Platform Management WOS® 2.5 256 Billion Objects GeoReplicated Cloud Foundation Mobile Cloud Access Storage Fusion Xcelerator (SFX) Flash Acceleration SFX Context Commit SFX Instant Commit SFX Read SFX Write Flexible Media Configuration SAS SATA SSD

  6. SS8460 – Highest Density Enclosure 84 Drives – SSD, SAS, SATA - in 4 rack units Up to 336 TB (84 x 4)

  7. SFA12K-40 (Block Appliance) 40 GB/s Read & Write Speed Highly Parallelized SFA Storage Processing Engine Active/Active Storage Design 35-40GB/s Read & Write Speed Up to 6.7PB of Disk 2.4+ Million Burst IOPS 700K+ Random Spinning Disk IOPS 1.7M Sustained Random SSD IOPS 64GB+ Mirrored Cache (Protected) RAID 1/5/6 Intelligent Block Striping DirectProtect™ GUI, SNMP, CLI, API 16 x FDR IB Host-Ports 8RU Height 16 x FDR InfiniBand Host Ports SFA Interface Virtualization SFA Interface Virtualization 32-64GB High-Speed Cache 32-64GB High-Speed Cache 240Gb/s Cache Link Internal SAS Switching Internal SAS Switching 960Gb/s Internal SAS Storage Management Network 1 2 3 4 5 6 7 8 P RAID 5,6 Q RAID 6 RAID 5,6 1 2 3 4 P RAID 5,6 Q RAID 6 SFA RAID 5,6 1 1m SFA RAID 1

  8. Big Data Solution Shapes • Data separate from compute • Data inside compute • Compute and Data all over the place • Compute inside the Data • Data Separate from Compute DDN Confidential

  9. Traditional Parallel Filesystems • Data separate from compute ‘Scaler File System Family In-Storage Processing™ Engine & DMA Driver Storage Fusion Architecture™ [Core Storage S/W Engine] Low-Latency Connect: FC, IB, Memory DirectMon™: Infrastructure Management Interrupt-Free Storage Processing ReACT™ Adaptive Cache Technology DirectProtect™ Data Integrity Management Storage Fusion Xcelerator (SFX) Quality of Service Engine Storage Fusion Fabric™ DDN Confidential

  10. High Single Client Performance • Data Separate from Compute Best Runtime Ever for Drug Discovery, Warranty, Risk Analytics 1-Trillion Row Big Data Queries in less than 20s. Up to 570% faster FSI back-testing and risk management 10 DDN Confidential

  11. Move Compute to the Data • Data inside compute End:End RDMA for Hadoop Shuffle Innovative I/O Node Data Pipelining Full Data Protection Offload 8+2 Data Protection w/ Real-Time I/O Hybrid HDD & SSD Configuration 300%+ Density; Flexible Scaling DDN Confidential

  12. Embedded Systems • Compute inside the Data Back-End Storage Enclosures Filesystem Clients Dedicated I/O Bridge Application Memory Dedicated I/O Bridge Cache Memory Multi-core CPU Application Processor (AP) Multi-core CPU RAID Processor (RP) Dedicated PCI-e I/O High Speed Bus Memory Pointers (Virtual Disks) File Server File Server File Server Multi-Threaded Real-Time RAID Engine, Hypervisor Virtual Disk Block Driver …… DDN Confidential

  13. Compute and Data all over the place DDN Confidential

  14. Simplified Data Lifecycle Ingest Process Distribute Store

  15. WOS: WEB Object Store • Not POSIX-based • Not RAID-based • No Spare Drives • No inode references, no FAT, no extent lists • No more running fsck • No more volume management • Not based on single-site/box architecture • 3 commands only: PUT, GET, DELETE

  16. Confidential - To Be Shared Under NDA Only

  17. WOS Puts & Gets Application returns file to user. A file is uploaded to the application or web server. A user needs to retrieve a file. The WOS client automatically determines what nodes have the requested object, retrieves the object from the lowest latency source, and rapidly returns it to the application. The WOS client returns a unique Object ID which the application stores in lieu of a file path. The application registers this OID with the content database. Application makes a call to the WOS client to read (GET) the object. The unique Object ID is passed to the WOS client. Application makes a call to the WOS client to store (PUT) a new object The WOS client stores the object on a node. Subsequent objects are automatically load balanced across the cloud. App/Web Servers OID = 5718a36143521602 OID = 5718a36143521602 Database The system then replicates the data according to the WOS policy, in this case the file is replicated to Zone 2. LAN/WAN Zone 1 Zone 2

  18. DDN | WOS® Data Protection Replicate and/or Erasure Coding Global, Peer:Peer Distribute data across 100s of sites in one namespace WOS API • C++, Python, Java, PHP, • HTTP, REST interfaces • PUT, GET, DELETE • S3 & WebDAV APIs • IOS Smartphones and Tablets • Multi-tenancy, Reporting and Billing Object ID Management NFS iRODS Connectors CIFS Object Placement Self-Healing Intelligent Data Management system recovers from failures rapidly and autonomously WOS Core [Peer:Peer Object Storage] API-based Integrate applications and devices more robustly Latency-Aware Access Manager Self-Healing Object Storage Clustering WOS Cluster Management WOS Policy Engine Policy driven Manage truly via policy, rather than micromanaging multiple layers of traditional filesystems Replication Engine ObjectAssure™ Erasure Coding De-clustered Data Management Small files, large files, streaming files Low seek times to get data WOS caching servers for massive streaming data

  19. DDN |WOS™ Deployment & Provisioning DDN | WOS building blocks are easy to deploy & provision – in 10 minutes or less • Provide power & network for the WOS Node • Assign IP address to WOS Node& specify cluster name (“Acme WOS 1”) • Go to WOS Admin UI. WOS Node appears in “Pending Nodes” List for that cluster San Francisco New York London Tokyo Simply drag new nodes to any zone to extend storage • Drag & Drop the node into the desired zone • Assign replication policy (if needed) NoFS It’s that simple to add 90TB (30x3) to your WOS cluster!

  20. WOS Screenshots Confidential - To Be Shared Under NDA Only

  21. Big Data ConnectediRODS Integration Site 1 • Now: iRODS/WOS compound resource • E-iRODS will bring a composable resources • Now iRODS uses WOS as a global storage system • iRODS Talks to all WOS IP addresses via REST interface • WOS handles the replica management • Interconnect Parallel Filesystem, Analytics and Cloud storage with Policy-based management .. iRODS Server iRODS Server ICAT DB .. iRODS Server iRODS Server SFA 10K ICAT DB JBOD Site 2 .. iRODS Server iRODS Server ICAT DB WOS Cluster Site 3

  22. Enable High Speed Edge Computing with GRIDScaler-WOS Bridge • Offload – Offload/archive GRIDScaler files to WOS to free up space & improve performance • Distribute & Federate – Replicate & federate files across other sites for collaboration & disaster protection • Collaborate – NFS users at remote sites can review & update files at local LAN speeds & share with GRIDScaler users GRIDScaler GRIDScaler

  23. Local Ingest Remote DistributionArchive & Distribution Use Case Ingest to GS for analysis, GS w/connector distributes to WOS for viewing/processing, GS to WOS DB Sync federates GS & WOS WS GS Ingest Process Flow, GS to WOS connector • Raw data “A” ingested into GS via WS • Analysis App processes raw sequencer data, writes resultant file “B” to WOS via GS-WOS Bridge • WOS Bridge DB synchronizes with WOS Access Sites • WOS Access DB syncs across sites which federates the NFS view across all sites Users utilizing review and/or processing applications @ remote sites can access the resultant files. 1 2 WOS Zone 2 WOS Zone 1 Viewing App 2 3 NFS or CIFS B 1 WOS Access NFS 4 3 Analysis WOS REST or WOSLIB WOS REST or WOSLIB 2 4 Native GPFS or nfs Samba or native CIFS 3 4 WOS Access GPFS GPFS / WOS Access DB Sync GPFS –WOS DB Sync Compute Cluster A B WOS Policy driven replication B Compute Cluster

  24. Big Data Connected • Use GPFS HSM features to create candidate file lists • Fast data movements into object store via 10Gbe/IB • Maintain metadata and file stubs in GPFS • Data in WOS moves according to policy Confidential - To Be Shared Under NDA Only

  25. WOS: Performance Comparison Google! World’s Fastest POSIX FS (new: 2H11) World’s Fastest POSIX FS EMC: http://reg.cx/1P1E Lustre; http://wiki.lustre.org/images/1/16/LUG08_Cray_HPCS.pdf GPFS: http://www.spscicomp.org/ScicomP13/Presentations/IBM/GPFSGunda.pdf Megastore: http://highscalability.com/blog/2011/1/11/google-megastore-3-billion-writes-and-20-billion-read-transa.html

  26. Summary • Scaling Up and Out and Deep across the Big Data Landscape requires a range of building block shapes • Managing the complexity of global distribution requires an alternative model for data management • Connecting Big Data solutions is the next challenge • More details: www.ddn.com http://cern.ch/Computing.Seminars/2013/0131

More Related