DOCSLIB.ORG

Johann Schleier-Smith and Joseph M. Hellerstein Presented at HPTS November 5, 2019 Outline

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Johann Schleier-Smith and Joseph M. Hellerstein Presented at HPTS November 5, 2019 Outline A File System for Serverless Computing Johann Schleier-Smith and Joseph M. Hellerstein Presented at HPTS November 5, 2019 Outline Serverless computing background Limitations The Cloud Function File System (CFFS) Evaluation and learnings 2 ©2019 RISELab Serverless computing in 2014 AWS announced Lambda Function as a Service (FaaS) in 2014, others clouds followed quickly Write code, upload it to the cloud, run at any scale Successful for web APIs and event processing, limited in stateful applications Thumbnail 3 ©2019 RISELab Serverless computing in 2014 AWS announced Lambda Function as a Service (FaaS) in 2014, others clouds followed quickly Write code, upload it to the cloud, run at any scale Successful for web APIs and event processing, limited in stateful applications Thumbnail 4 ©2019 RISELab Defining characteristics of serverless abstractions Hiding the servers and the complexity of programming them Consumption-based costs and no charge for idle resources Excellent autoscaling so resources match demand closely 5 ©2019 RISELab Understanding serverless computing’s impact Resources People On-premise Serverful cloud Serverless cloud Job Serverful Cloud Serverless Cloud Infrastructure Outsourced job Outsourced job Administration System Simplified job Outsourced job Administration Resources Software Allocated & Charged Allocated Little change Simplified job Development Time 6 ©2019 RISELab Serverless as next phase of cloud computing Serverless abstractions offer Simplified programming Outsourced operations Improved resource utilization 7 ©2019 RISELab Serverless is much more than FaaS Function as a Service Big Data Processing AWS Lambda Object Storage Google Cloud Dataflow Google Cloud Functions AWS S3 AWS Glue Google Cloud Run Azure Blobs AWS Athena Azure Functions Google Cloud Storage AWS Redshift Queue Service Mobile back end Google Firebase AWS SQS Key-Value Store AWS AppSync Google Cloud Pub/Sub AWS DynamoDB Azure CosmosDB Google Cloud Datastore AWS Serverless Aurora Google App Engine 8 ©2019 RISELab Fertile ground for research 120 Actual Extrapolated 100 80 60 40 Number of Publications 20 0 2012 2013 2014 2015 2016 2017 2018 2019 Year Source: dblp computer science bibliography. Outline Serverless computing background Limitations The Cloud Function File System (CFFS) Evaluation and learnings 10 ©2019 RISELab Limitations of FaaS No inbound network Limited runtime connections No specialized Ephemeral state hardware, e.g., GPU Plenty of complementary stateful serverless services Others Object Storage Key-Value Store • AWS Aurora Serverless • AWS S3 • AWS DynamoDB • Google Firebase • Azure Blobs • Azure CosmosDB • Google Cloud Storage • Google Cloud Datastore • Anna KVS (Berkeley) Combine with FaaS to build applications Compute: λ λ isolated & ephemeral state Storage: Object Key-Value Etc. shared & durable state Storage Store Allows independent scaling λ λ λ λ λ Object Key-Value Etc. Etc. Etc. Storage Store How happy are we? How happy are we? Two main problems Latency API Can I please have something like local disk, but for the cloud? File systems let us run so much software Data analysis with Pandas Machine learning with TensorFlow Software builds with Make Search indexes with Sphinx Image rendering with Radiance Databases with SQLite Web serving with Nginx Email with Postfix 20 ©2019 RISELab Objections It won’t scale Need a simpler data model (key-value store, immutable objects) Need a weaker consistency model (eventual consistency) Performance and reliability will suffer otherwise You don’t need it You should be rewriting your software for the cloud anyhow Why not just use use a key-value store? 21 ©2019 RISELab How does this make me feel? 22 ©2019 RISELab Outline Serverless computing background Limitations The Cloud Function File System (CFFS) Evaluation and learnings 23 ©2019 RISELab Introducing the Cloud Function File System (CFFS) POSIX semantics, including strong consistency Local caches for local disk performance Works under autoscaling, extreme elasticity, and FaaS limitations Implemented as a transaction system 24 ©2019 RISELab What is special about the FaaS environment? Function invocations have well defined beginning and end At-least-once execution—expects idempotent code Constrained execution model Clients frozen in between invocations No inbound network connections No root access 25 ©2019 RISELab CFFS Architecture CFFSλ CFFSλ CFFSλ CFFSλ CFFSλ CFFSλ CFFS Back end transactional system 26 ©2019 RISELab FaaS instance caching “Function as a Service” suggests statelessness, but most implementations reuse instances and preserve their state Setup of sandboxed environment takes time Loads selected runtime (e.g., JavaScript, Python, C#, etc.) Configures network endpoint, IAM privileges Loads user code Runs user initialization Caching is key to amortizing instance setup 27 ©2019 RISELab Cloud Function File System (CFFS) Application — POSIX API — Our Txn Buffer emphasis Cache Cache Logs Txn Commit 28 Back end transactional©2019 RISELab system Core API implemented in CFFS open New descriptor (handle) for file or mkdir Create directory directory rename Rename file / directory close Close descriptor unlink Delete file / directory write / pwrite Write / positioned write chmod Set access permissions read / pread Read / positioned read chown Set ownership stat Get size, ownership, access, utimes Update modified / accessed permissions, last modified timestamps seek Set descriptor position clock_gettime Get current time dup / dup2 Copy descriptor chdir Set working directory truncate Set file size getcwd Get working directory flock Byte range lock and unlock begin Start transaction commit / abort End transaction 29 ©2019 RISELab POSIX guarantees - language from the spec • Writes can be serialized with respect to other reads and “ writes. ” • If a read() of file data can be proven (by any means) to occur after a write() of the data, it must reflect that write()… • A similar requirement applies to multiple write operations to the same file position… • This requirement is particularly significant for networked file systems, where some caching schemes violate these semantics. https://pubs.opengroup.org/onlinepubs/9699919799/ POSIX guarantees in database terms Atomic operations Each operation (at the API level) is observed entirely or not at all Some violations in practice Consistency model Spec references time, technically requires strict consistency (shared global clock) Actually implemented as sequential consistency (global order exists, consistent with order at each processor) We use serializability to provide isolation and atomicity at function granularity Open question: what guarantees do applications actually rely on? 31 ©2019 RISELab Implementation highlights Choice of transaction mechanism not fundamental Implemented timestamp-order serilizable Optimistic protocols can be a good fit—FaaS side effects must be idempotent State-of-the-art protocols promise lower abort rates, more effective local caches (e.g., Yu, et al., VLDB 2018) Cache updates through on-demand filtererd log shipping Check for updates when function starts execution Eviction messages help back-end track client cache content 32 ©2019 RISELab CFFS in context: Transactional file systems QuickSilver distributed system – IBM, 1991 Very close in spirit No FaaS No caching Inversion File System – Berkeley, 1993 Built on top of PostgreSQL Access through custom library Transactional NTFS (TxF) – Microsoft, 2006 Shipping in Windows Deprecated on account of complexity There are many non-transactional shared & distributed file systems 33 ©2019 RISELab CFFS in context: Shared file systems Must choose between consistency and latency Eventual consistency Delegation/lock-based caching No caching HPC Big data Client-server Lustre HDFS NFS GPFS (IBM) GFS (Google) SMB GlusterFS (RedHat / IBM) Ceph (IBM) GFS (Google File System) Mainframe MooseFS MapR-FS zFS LizardFS Alluxio BeeGFS 34 ©2019 RISELab Outline Serverless computing background Limitations The Cloud Function File System (CFFS) Evaluation and learnings 35 ©2019 RISELab Sample workload call frequencies 120,000 make OpenSSH 100,000 TPC-C on SQLite 80,000 60,000 40,000 20,000 - stat dup read seek write chdir close open unlink chmod getdent rename 36 ©2019 RISELab Caching benefits (TPC-C / SQLite) 9,000 8,105 8,000 7,000 6,000 5,000 tpmC 4,000 3,000 2,000 1,000 817 142 - Local NFS CFFS 37 ©2019 RISELab Scaling in AWS Lambda 180,000 Uncached Cached 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 - 10 20 30 40 50 60 70 80 90 100 4k random reads 38 ©2019 RISELab Returning to objections It won’t scale—contention risks File length: must be checked on every read Stat command: mainly used use it to check the file length or permissions, but also returns modification time and access time Challenges here, optimistic they will be overcome You don’t need it I think it will be pretty useful 39 ©2019 RISELab How happy are we? How happy are we? CFFS Summary Transactions are a natural fit for FaaS BEGIN and END from function context At-least-once execution goes well with optimistic transactions On-demand filtered log shipping allows asynchronous cache updates Overcomes limitations of FaaS & traditional shared file systems Allows caching for lower latency, preserving consistency Highly scalable, especially with snapshot reads POSIX API enables vast range of tools and libraries 42 ©2019 RISELab.
Load more

Recommended publications
  • FOSDEM 2017 Schedule
    FOSDEM 2017 - Saturday 2017-02-04 (1/9) Janson K.1.105 (La H.2215 (Ferrer) H.1301 (Cornil) H.1302 (Depage) H.1308 (Rolin) H.1309 (Van Rijn) H.2111 H.2213 H.2214 H.3227 H.3228 Fontaine)… 09:30 Welcome to FOSDEM 2017 09:45 10:00 Kubernetes on the road to GIFEE 10:15 10:30 Welcome to the Legal Python Winding Itself MySQL & Friends Opening Intro to Graph … Around Datacubes Devroom databases Free/open source Portability of containers software and drones Optimizing MySQL across diverse HPC 10:45 without SQL or touching resources with my.cnf Singularity Welcome! 11:00 Software Heritage The Veripeditus AR Let's talk about The State of OpenJDK MSS - Software for The birth of HPC Cuba Game Framework hardware: The POWER Make your Corporate planning research Applying profilers to of open. CLA easy to use, aircraft missions MySQL Using graph databases please! 11:15 in popular open source CMSs 11:30 Jockeying the Jigsaw The power of duck Instrumenting plugins Optimized and Mixed License FOSS typing and linear for Performance reproducible HPC Projects algrebra Schema Software deployment 11:45 Incremental Graph Queries with 12:00 CloudABI LoRaWAN for exploring Open J9 - The Next Free It's time for datetime Reproducible HPC openCypher the Internet of Things Java VM sysbench 1.0: teaching Software Installation on an old dog new tricks Cray Systems with EasyBuild 12:15 Making License 12:30 Compliance Easy: Step Diagnosing Issues in Webpush notifications Putting Your Jobs Under Twitter Streaming by Open Source Step. Java Apps using for Kinto Introducing gh-ost the Microscope using Graph with Gephi Thermostat and OGRT Byteman.
    [Show full text]
  • ZFS Basics Various ZFS RAID Lebels = & . = a Little About Zetabyte File System ( ZFS / Openzfs)
    = BDNOG11 | Cox's Bazar | Day # 3 | ZFS Basics & Various ZFS RAID Lebels. = = ZFS Basics & Various ZFS RAID Lebels . = A Little About Zetabyte File System ( ZFS / OpenZFS) =1= = BDNOG11 | Cox's Bazar | Day # 3 | ZFS Basics & Various ZFS RAID Lebels. = Disclaimer: The scope of this topic here is not to discuss in detail about the architecture of the ZFS (OpenZFS) rather Features, Use Cases and Operational Method. =2= = BDNOG11 | Cox's Bazar | Day # 3 | ZFS Basics & Various ZFS RAID Lebels. = What is ZFS? =0= ZFS is a combined file system and logical volume manager designed by Sun Microsystems and now owned by Oracle Corporation. It was designed and implemented by a team at Sun Microsystems led by Jeff Bonwick and Matthew Ahrens. Matt is the founding member of OpenZFS. =0= Its development started in 2001 and it was officially announced in 2004. In 2005 it was integrated into the main trunk of Solaris and released as part of OpenSolaris. =0= It was described by one analyst as "the only proven Open Source data-validating enterprise file system". This file =3= = BDNOG11 | Cox's Bazar | Day # 3 | ZFS Basics & Various ZFS RAID Lebels. = system also termed as the “world’s safest file system” by some analyst. =0= ZFS is scalable, and includes extensive protection against data corruption, support for high storage capacities, efficient data compression and deduplication. =0= ZFS is available for Solaris (and its variants), BSD (and its variants) & Linux (Canonical’s Ubuntu Integrated as native kernel module from version 16.04x) =0= OpenZFS was announced in September 2013 as the truly open source successor to the ZFS project.
    [Show full text]
  • Executive Summary
    LizardFS DISTRIBUTED, PARALLEL, FAULT–TOLERANT, HIGHLY AVAILABLE, GEO–REDUNDANT FILE SYSTEM BUILD YOUR STORAGE EVEN FROM COMMODITY HARDWARE, USE YOUR CURRENT INFRASTRUCTURE YOUR BENEFITS USE CASES Scale up to exabytes FINANCE Big Data Elastic Search Reduce your storage cost SAS Analytics Reduce time spent on storage DATA CENTER maintenance Backup Big Data Archive Cloud Drive Virtualization Increase your usable capacity MEDIA Eliminate storage–related downtime Animation Render Farm Storage Special Effects (SFX) Visual Effects (VFX) No vendor lock-in Postproduction CCTV Recordings TELECOMS Billing Big Data SAS Analytics CCTV Recordings SCIENCE Satellite Images Seismologic Data Microscope Images Genome Sequencing DATA SAFETY REDUCING RTO Thanks to High Availability and Erasure Coding Thanks to LizardFS, the time needed to mechanisms, server failures will not affect your recover from a disaster is reduced business continuity. Your data will always stay massively. The stability of your business is safe. what really matters. REDUCED MAINTENANCE TIME FULL TRANSPARENCY Disc failures are handled automatically with no LizardFS is an open-source solution. You can need of human intervention. Thanks to that you trust hundreds of our Community don’t have to hire an army of sysadmins. members, not just us. USE YOUR CURRENT INFRASTRUCTURE GEO REPLICATION LizardFS is 100% hardware agnostic. You can run it It allows you to create geo stretched clusters on any hardware you want. stretching virtually any distance. PLUG AND SCALE INCREASED PERFORMANCE Scale vertically as well as horizontally by simply LizardFS allows for parallel writes and reads, adding or removing a single drive or node. thus enhancing performance. COST EFFICIENCY EASY TO INSTALL AND MAINTAIN With us your existing storage keeps growing inste- You don’t have to graduate from the School of ad of being replaced every few years.
    [Show full text]
  • Freebsd Enterprise Storage Polish BSD User Group Welcome 2020/02/11 Freebsd Enterprise Storage
    FreeBSD Enterprise Storage Polish BSD User Group Welcome 2020/02/11 FreeBSD Enterprise Storage Sławomir Wojciech Wojtczak [email protected] vermaden.wordpress.com twitter.com/vermaden bsd.network/@vermaden https://is.gd/bsdstg FreeBSD Enterprise Storage Polish BSD User Group What is !nterprise" 2020/02/11 What is Enterprise Storage? The wikipedia.org/wiki/enterprise_storage page tells nothing about enterprise. Actually just redirects to wikipedia.org/wiki/data_storage page. The other wikipedia.org/wiki/computer_data_storage page also does the same. The wikipedia.org/wiki/enterprise is just meta page with lin s. FreeBSD Enterprise Storage Polish BSD User Group What is !nterprise" 2020/02/11 Common Charasteristics o Enterprise Storage ● Category that includes ser$ices/products designed &or !arge organizations. ● Can handle !arge "o!umes o data and !arge num%ers o sim#!tano#s users. ● 'n$olves centra!ized storage repositories such as SA( or NAS de$ices. ● )equires more time and experience%expertise to set up and operate. ● Generally costs more than consumer or small business storage de$ices. ● Generally o&&ers higher re!ia%i!it'%a"aila%i!it'%sca!a%i!it'. FreeBSD Enterprise Storage Polish BSD User Group What is !nterprise" 2020/02/11 EnterpriCe or EnterpriSe? DuckDuckGo does not pro$ide search results count +, Goog!e search &or enterprice word gi$es ~ 1 )00 000 results. Goog!e search &or enterprise word gi$es ~ 1 000 000 000 results ,1000 times more). ● /ost dictionaries &or enterprice word sends you to enterprise term. ● Given the *+,CE o& many enterprise solutions it could be enterPRICE 0 ● 0 or enterpri$e as well +.
    [Show full text]
  • Survey of File Systems Yale, August 13Th – 17Th 2018
    Linux Clusters Institute: Survey of File Systems Yale, August 13th – 17th 2018 J.D. Maloney | Storage Engineer National Center for Supercomputing Applications (NCSA) [email protected] Baseline • Look over slides from Irfan Elahi’s talks at the beginner workshop in May 2018 • http://linuxclustersinstitute.org/workshops/intro18/program.html • Segments of some following slides were adapted from her work • Have an grasp on: • Definition of Parallel file system • Types of file systems (home, sCratCh, projeCts, arChive) August 13th-17th 2018 3 Understanding Your Workloads August 13th-17th 2018 4 Finding Features You Need • What features/traits do you need your file system to have? • Some Examples: • Encryption at Rest • Tiered Storage/HSM • Simultaneous Multi-FaBric Support • High sequential throughput • High metadata performance • Open Source (aka no licenses) • Object Storage • Less FTE intensive to operate • Local node cache support • Very high client count • Rank the order of importance, nothing has all the upsides with no downsides August 13th-17th 2018 5 Look at Current Metrics • Similar to when exploring hardware, look at your current file system usage • Do you see a need for high metadata performance, sequential I/O performance? • Is there a need for growth to support a high number of clients (many 10,000s) over the lifetime of the system? • Does your data center require the file system to interface with multiple network fabrics and/or clusters? • Look at your growth model to see how you’re going to need to add capacity or performance
    [Show full text]
  • Analiza Właściwości Wybranych Rozwiązań Rozproszonych Systemów Plików
    PRZEGLĄD TELEINFORMATYCZNY NR 3, 2017 Analiza właściwości wybranych rozwiązań rozproszonych systemów plików Karolina KULA, Marcin MARKOWSKI Politechnika Wrocławska, Wybrzeże Wyspiańskiego 27, 50-370 Wrocław [email protected], [email protected] STRESZCZENIE: W artykule przedstawiono analizę porównawczą trzech najpopularniejszych rozproszonych systemów plików: Ceph, LizardFS, GlusterFS. Głównym kryterium oceny badanych systemów była ich wydajność, rozumiana jako liczba operacji zapisu i odczytu w jednostce czasu oraz maksymalny transfer danych. Analizę oparto na wynikach licznych eksperymentów przeprowadzonych w opracowanym w tym celu stanowisku badawczym. Część badawczą poprzedzono przeglądem technologii oraz najpopularniejszych istniejących systemów tego typu. W pracy przedstawiono wyniki eksperymentów, ich analizę oraz praktyczne wnioski. SŁOWA KLUCZOWE: rozproszone systemy plików, wirtualizacja, usługi chmurowe, Ceph, LizardFS, GlusterFS 1. Wprowadzenie Zakres świadczenia usług informatycznych przez firmy IT stale się poszerza. Rośnie również zapotrzebowanie na zasoby informatyczne. Nie wszystkie technologie stosowane w firmach usługowych dają się skalować. Możliwe jest także, że własne wdrażanie udoskonaleń, zabezpieczeń i redundancji jest zbyt drogie lub ryzykowne. Dlatego wraz ze wzrostem zapotrzebowania na usługi, rozwija się technologia chmur obliczeniowych (ang. Cloud Computing), umożliwiająca zdalne świadczenie wszelkiego rodzaju usług dla różnych typów użytkowników (korporacji, małych firm, prywatnych użytkowników).
    [Show full text]
  • Leveraging Information-Driven Storage in an NFV Marketplace in the 5G Ecosystem
    University of Piraeus Department of Digital Systems Postgraduate Programme in Big Data & Analytics Leveraging Information-driven Storage in an NFV Marketplace in the 5G Ecosystem MSc Thesis Stavrianos Panagiotis Registration Number: ME1732 Supervisor Dimosthenis Kyriazis Assistant Professor ii Leveraging Information-driven Storage in an NFV Marketplace in the 5G Ecosystem Panagiotis Stavrianos Copyright © Panagiotis Stavrianos All rights reserved It is forbidden to copy, store and distribute this work, in whole or in part, for commercial purposes. Re- production, storage and distribution may be permitted for non-profit, educational or research purposes, only by providing the source of origin and the message is retained. Questions about the use of work for profit should be addressed to the writers. The views and conclusions contained in this document reflect the authors and should not be interpreted as representing the official positions of the University of Piraeus. University of Piraeus Department of Digital Systems Postgraduate Programme in Big Data & Analytics Leveraging Information-driven Storage in an NFV Marketplace in the 5G Ecosystem MSc Thesis Supervisor Dimosthenis Kyriazis Assistant Professor Dimosthenis Kyriazis Ilias Maglogiannis Andriana Prentza Assistant Professor Associate Professor Associate Professor Consistency is the hallmark of the unimaginative -Oscar Wilde- Abstract In the previous years, preliminary interest and discussions about a possible 5G standard have evolved into a full-fledged conversation that has captured the attention and imagination of researchers and engineers around the world. 5G networks expect to provide significant advances in network management compared to traditional mobile infrastructures by leveraging intelligence capabilities such as data analysis, prediction, pattern recog- nition and artificial intelligence.
    [Show full text]
  • CEPH Storage [Open Source Osijek]
    Open Source Osijek Licenca : GPL Autor: Hrvoje Horvat CEPH sustav 1 Sadržaj CEPH storage .................................................................................................................................. 4 NAS i SAN ................................................................................................................................. 4 Što je uopće NAS sustav ? ...................................................................................................... 4 A što je SAN sustav ................................................................................................................ 8 Zbog čega uopće koristiti NAS ili SAN sustave ......................................................................... 9 U čemu je problem sa standardnim NAS ili SAN sustavima ............................................... 11 Slijedeći korak: Klasterski i/ili redundantni NAS sustavi ........................................................ 11 Redundantni ili Klasterizirani NAS Sustavi ......................................................................... 12 Redundantni ili Klasterizirani SAN Sustavi ......................................................................... 14 ZFS - negdje između ............................................................................................................. 14 Proces učenja ........................................................................................................................ 17 Što je slijedeće .............................................................................................................................
    [Show full text]
  • (FSS) in Oracle Cloud Infrastructure Bradley C
    Everyone Loves File: File Storage Service (FSS) in Oracle Cloud Infrastructure Bradley C. Kuszmaul, Matteo Frigo, Justin Mazzola Paluska, and Alexander (Sasha) Sandler, Oracle Corporation https://www.usenix.org/conference/atc19/presentation/frigo This paper is included in the Proceedings of the 2019 USENIX Annual Technical Conference. July 10–12, 2019 • Renton, WA, USA ISBN 978-1-939133-03-8 Open access to the Proceedings of the 2019 USENIX Annual Technical Conference is sponsored by USENIX. Everyone Loves File: File Storage Service (FSS) in Oracle Cloud Infrastructure Bradley C. Kuszmaul Matteo Frigo Justin Mazzola Paluska Alexander (Sasha) Sandler Oracle Corporation Abstract 100 MB=s of bandwidth and 3000 operations per sec- ond for every terabyte stored. Customers can mount a File Storage Service (FSS) is an elastic filesystem pro- filesystem on an arbitrary number of NFS clients. The vided as a managed NFS service in Oracle Cloud In- size of a file or filesystem is essentially unbounded, lim- frastructure. Using a pipelined Paxos implementation, ited only by the practical concerns that the NFS pro- we implemented a scalable block store that provides lin- tocol cannot cope with files bigger than 16 EiB and earizable multipage limited-size transactions. On top of that we would need to deploy close to a million hosts the block store, we built a scalable B-tree that provides to store multiple exabytes. FSS provides the ability linearizable multikey limited-size transactions. By us- to take a snapshot of a filesystem using copy-on-write ing self-validating B-tree nodes and performing all B- techniques.
    [Show full text]
  • Lizardfs Documentation Release Latest
    LizardFS Documentation Release latest Michal Bielicki Jun 15, 2018 Contents 1 ToDo-List 3 2 Introduction to LizardFS 5 2.1 Architecture...............................................5 2.2 Replication-Modes............................................6 2.3 Possible application of LizardFS.....................................6 2.4 Hardware recommendation.......................................7 2.5 Additional Features...........................................7 3 LizardFS QuickStart on Debian9 3.1 Master server installation........................................9 3.2 Shadow master installation........................................ 10 3.3 Metalogger installation.......................................... 11 3.4 Chunk server installation......................................... 11 3.5 Cgi server installation.......................................... 12 3.6 Command line administration tools................................... 13 4 LizardFS Administration Guide 15 4.1 Installing LizardFS............................................ 15 4.2 Basic Configuration........................................... 18 4.3 Configuring Replication Modes..................................... 26 4.4 Advanced configuration......................................... 29 4.5 Configuring Optional Features...................................... 34 4.6 Connecting Clients to your LizardFS installation............................ 36 4.7 Securing your LizardFS installation................................... 40 4.8 Optimizing your LizardFS setup....................................
    [Show full text]
  • Ütläärschulr Dekanat - Prüfungsausschuss LJ LUBECK University of Applied §Ciences
    Fachbereich Elektrotechnik und lnformatik ÜtläärscHulr Dekanat - Prüfungsausschuss LJ LUBECK University of Applied §ciences Thema der Masterarbeit für Herrn Thomas F in n e r n Analyse und lntegration von Storage-Clustern in elastische Container Plattformen ofessor Dr. Kratzke gefordert durch: {r Ausgabedatum: 02. Januar 2017 rf Abgabedatum: 03. Juli 2017 E -q**V,.,1 s& -4, #w r+ (Professor Dr. Andreas Hanemann) Vorsitzender des Prüfu ngsaussch usses L# ffi Seite 2 Thema der Masterarbeit für Herrn Thomas Finnern Aufgabenstellung: Elastische Container Plattformen können auf Maschinen von verschiedenen Cloud Service Providern (2.8. AWS, GCE, Azure) ausgebracht werden und ermöglichen eine Migration von Workloads zwischen Cloud-Anbietern. Eine Realisierung dieses Ansat- zes wurde im Forschungsprojekt Cloud TRANSIT bereits teilweise durch die Entwick- lung eines Cluster-Deployers für Kubernetes auf Amazon AWS und OpenStack umge- setzt. Aktuell können zustandslose Applikationen providerunabhängig bereitgestellt werden. Mit einem ergänzenden Storage-Cluster sollen nun auch zustandsbehaftete Applikationen u nterstützt werden. In dieser Masterarbeit sollen dafür existierende Open Source Storage-Cluster Lösun- gen analysiert, integriert und hinsichtlich Migrierbarkeit evaluiert werden. Hierzu sind im Detailfolgende TeilaufEaben zu bearbeiten und zu dokumentieren (Masterarbeit): . Automatisches Bereitstellen von Storage-Cluster Lösungen (lntegration) . Evaluierung der Migrierbarkeit von Storage-Clustern Kratzke Fachbereich Elektrotechnik und Informatik Dekanat - Prüfungsausschuss University of Applied Sciences Erklärung zur Abschlussarbeit Ich versichere, dass ich die Arbeit selbständig, ohne fremde Hilfe verfasst habe. Bei der Abfassung der Arbeit sind nur die angegebenen Quellen benutzt worden. Wörtlich oder dem Sinne nach entnommene Stellen sind als solche gekennzeichnet. Ich bin damit einverstanden, dass meine Arbeit veröffentlicht wird, insbesondere dass die Arbeit Dritten zur Einsichtnahme vorgelegt oder Kopien der Arbeit zur Weitergabe an Dritte angefertigt werden.
    [Show full text]
  • Freebsd Enterprise Storage Polish BSD User Group Welcome 2020/02/11 Freebsd Enterprise Storage
    FreeBSD Enterprise Storage Polish BSD User Group Welcome 2020/02/11 FreeBSD Enterprise Storage Sławomir Wojciech Wojtczak [email protected] vermaden.wordpress.com twitter.com/vermaden bsd.network/@vermaden https://is.gd/bsdstg FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 What is Enterprise Storage? FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 What is Enterprise Storage? The wikipedia.org/wiki/enterprise_storage page tells nothing about enterprise. FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 What is Enterprise Storage? The wikipedia.org/wiki/enterprise_storage page tells nothing about enterprise. Actually just re irects to wikipedia.org/wiki/data_storage page. FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 What is Enterprise Storage? The wikipedia.org/wiki/enterprise_storage page tells nothing about enterprise. Actually just re irects to wikipedia.org/wiki/data_storage page. The other wikipedia.org/wiki/computer_data_storage page also oes the same. FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 What is Enterprise Storage? The wikipedia.org/wiki/enterprise_storage page tells nothing about enterprise. Actually just re irects to wikipedia.org/wiki/data_storage page. The other wikipedia.org/wiki/computer_data_storage page also oes the same. The wikipedia.org/wiki/enterprise is just meta page !ith lin"s. FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 Common Charasteristics o Enterprise Storage FreeBSD Enterprise Storage Polish BSD User Group What is Enterprise? 2020/02/11 Common Charasteristics o Enterprise Storage ● Category that inclu es ser$ices/pro ucts esigne &or !arge organizations.
    [Show full text]