Department of Computer Science | Institute of Systems Architecture | Chair of Computer Networks From Research T o T ransfer: User-Controllable Cloud Storage Josef Spillner mailto:josef.spillner@tu-dresden.de xmpp:josef.spillner@jabber.org 14 th TERENA TF-Storage Meeting February 10-11, 2014, Zurich, Switzerland
The T rouble with Cloud Storage proprietary client backup/sync vs. storage quality? trust? institutional utility billing pricing: progressive/ 3 months limit; degressive web access tariff choice file size limit only
T aming the T rouble: RAICs, RAOCs proxy cloud controller/ gateway integration storage targets: services, devices auto-selection scheduling auto-configuration service properties optimisation auto-repair service bundles extension backup sync sharing ...
The Cloud Storage Lab @ TU Dresden Short history: ● 2010 ideas and service modelling ● 2011 NubiSave storage controller started UCC paper on Cloud storage controllers ● 2012 Usable software, Debian packaging ● 2013 NubiVis ; π-Box personal Cloud VM FGCS article on optimal storage SUCRE summer school tutorial at KIT http://lab.nubisave.org/ ● 2014 NubiGate VM prototypical experiments widespread research software and pilot transfer production development installations deployments
Research
Storage Service Ecosystem Excursus: Service Science -> Bundles -> RAOC super-domain Infrastructure functionality non-functional domain 2 properties domain Cloud Storage domain 1 nfp nfp 1 backup nfp 2 sub-domain File Storage consumer client protocol description service interface SLAs messages service instance service impl service hosting engineer provider
Storage Service Descriptions Base Ontology (WSML) concept MeasurementUnit conversionFactor impliesType _double concept SpaceUnit subConceptOf MeasurementUnit instance GB memberOf SpaceUnit conversionFactor hasValue 1024.0 Instance Ontology webService GoogleStorage importsOntology { _"urn:ontology:conqo/CloudQoS.wsml#" } capability ServiceCapability postcondition definedBy ?serviceType memberOf cloud#CloudStorage . instance PricePerData memberOf { cloud#PricePerData, qos#ServiceSpec } qos#value hasValue 0.17 qos#unit hasValue qos#Euro
Storage Service Bundles $$ Service bundle a x replication s 1 (k+m) erasure coding secret sharing s 1 s 2 s 3 s 4 s 5 s 6 s 1 s 2 s 3 s 4 s 5 s 6
Storage Service Markets
Storage Flows sink transport source Storage Targets Data Flow Integration block File modifier USB stick modifier (compress) Splitter/ local (enc,dedup) Multiplexer directory block local modifier directory (stego) CIFS NFS Splitter/ proxy export Multiplexer Splitter/ Multiplexer web Splitter/ WebDAV application Multiplexer file Commercial modifier (Dropbox) (enc)
Reconfjgurable Storage Flows ... alias Software-Defined Storage Modules (i.e. FUSE modules): - Splitter (1:n) - Modifiers (1:1) * encryption * compression * ... - Transports * Directory * CIFS, NFS * WebDAV * SSHfs * S3fs * ... * CloudFusion - SDS propagation
Multi-User Storage Gateways FlexDDPL: Flexible Data Distribution Policy Language scopes → applied in contexts (e.g. user database) ~user !negation T:22:00-06:00 @group fragment mime:message/rfc822 rules → applied to targets (e.g. storage provider) store control adapt
Feature: Algorithms Selection Background ● secret sharing ● forward error correction / replication Striping (→ RAID0) ● capacity 1.0, safety 1.0, performance 2.0 Mirroring (→ RAID1) ● capacity 0.5, safety 2.0, performance 1.0 XOR parity (→ RAID5, RAID6) ● 4 disks: capacity 0.75, safety 1.33, performance 1.0 Erasure coding (→ RAIDn) ● n storage targets, k significant, m redundant: capacity k/m, safety 1+m/k ● (Rotated) Reed-Solomon: maximum-[Hamming-]distance separable (MDS) property ● Cauchy-Reed-Solomon, Vandermonde-Reed-Solomon: practically secure; CRS faster due to XOR operations compared to GF(2 w ) AONT: information-theoretically secure AONT-RS: blend of AONT with (C)RS; alternative: encrypted fragments
Feature: Nested Contexts Configuration Data context »normal« context »private«
Feature: Weights Assigning weights to the storage targets ● to fully utilise the capacity ● to exploit faster upload connections
Feature: Recursion Saving the database which contains all file and fragment metadata... ● to avoid single point of failure ● to allow for selective sharing
Feature: Streaming (1:x) Continuous operation during read or write of large file ● buffers on controller device can be kept small ● improved performance through parallel coding and transmission
Feature: Batching (x:1) Encoding speed of different batch sizes 15 nodes 100000 90000 80000 70000 M=0 Encoding [MBs] M=4 60000 M=8 50000 M=12 40000 30000 20000 10000 0 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 Filechunk size [KB]
Experiments Nubisave Read Performance - 1MB File - UseAllInParallel Read in KB/s 13000 14000 12000 12000 11000 10000 10000 9000 8000 8000 7000 6000 6000 4000 5000 0 10 20 30 40 50 60 70 80 90100 1 2 3 4 5 Redundancy 6 7 Storages 8 9 10 peaCS: Performance and Efficiency Analsis for Cloud Storage Experiment Week Feb 7-14 2014
Autonomous Storage Clients Scenario: A robot wants to store some data... Labcast available: http://www.youtube.com/ watch?v=NDlN0fG9Okk
Autonomous Storage Clients Scenario: Pervasive storage on all devices Credentials may be auto generated
Software
Comparison of Storage Controllers Prototype Origin Authors Approach Integration Public Availability Dependable T-Clouds, Bessani, Corr- data units, application no Univ. of Lisbon eia, Quaresma, byzantine fail, library Sky ('11) André, Sousa Rabin erasure ('11) HPI, Maxim Schnja- NFPs, propr. web interface no Potsdam kin, Christoph platform, Libe- Meinel ration erasure TrustedSafe eGovCD, Berlin/ (no publication) n fragments: Windows plugin no FhG FOKUS one on USB, ('10) others remote Cloud Chinese Nan Zang, Jiwu Two fragments: (not yet known) no Academy of Jing, Peng Liu one local, one Shredder ('11) Sciences remote SecCSIE ('11) T-Systems Ronny Seiger, Cachy Reed- CIFS proxy no MMS, Dresden Stephan Groß Solomon erasure RACS ('10) Cornell Abu-Libdeh, erasure coding: S3 proxy, can be yes University Princehouse, Zfec Reed- distributed w/ (www.cs.cornell Weatherspoon Solomon Zookeeper .edu) plus numerous commercial implementations: since 2012, community projects: Tahoe-LAFS externals, TCD, NetApp StorageGRID, Cleversafe, Trusted Safe etc. OpenStack/Intel - PyEC2Lib
OpenStack Intermezzo
NubiSave Cloud Storage Controller Installation of NubiSave ● Git repository git://nubisave.org/git/nubisave ● ● Debian packages http://nubisave.org/packages/ ● ● Integrated into π-Box or NubiGate VMs http://nubisave.org/downloads/ ●
NubiSave Storage Flow Editor
NubiSave Confjguration
NubiSave Confjguration
NubiSave T ools Controller operation Storages and database $ n u b i s a v e [ < i n s t a n c e > ] $ n u b i s a v e - s t a t u s Master script which starts both Lists all splitter instances and of the below combined. storage targets attached to them or as part of any storage flow. $ n u b i s a v e h e a d l e s s $ n u b i s a v e - m o u n t e r [ < m o d u l e > ] Starts a new instance of the splitter/dispersion file system. Mounts all splitters and/or storage targets. $ n u b i s a v e g u i $ n u b i s a v e - u n m o u n t e r [ < m o d u l e > ] Starts the storage flow editor with storage integration configuration. Unmounts; inverse of the above. $ n u b i s a v e s t o p $ n u b i s a v e - d a t a b a s e [ . . . ] Stops the splitter. Inspection of storage metadata.
Splitter-NG Framework Plugins: Jerasure, JSharing, RAID-1, more coming...
T ransport Module: CloudFusion
Additional Implementations
NubiVis: Distributed Data Visualisation
NubiVis Map View
NubiGate VM: Everything in one Box
NubiGate VM: OwnCloud Frontend Our instance „RN-Dropbox“ 141.76.41.232/owncloud Your instance...? 141.76.41.206 141.76.41.207
Future Vision
Future Work What's next in our lab ● Software improvements ● Configuration through the web interface (student thesis) ● Git-based service registry query ● Research: MapReduce-like processing over dispersed data ● Scaling large with more users and storage providers, HA setups etc. ● Federation between research institutions: reciprocative/collaborative storage resource sharing <-- role of NRENs
Recommend
More recommend
