1 Conquering Big Data with Apache Spark Ion Stoica November 1 st, 2015 UC BERKELEY2 The Berkeley AMPLab January faculty > 50 students 3 software en...
1 Apache Flink Big Data Stream Processing Tilmann Rabl Berlin Big Data Center bbdc.berlin XLDB Berlin Big Data Center All Rights Reserved DIMA 20172 ...
1 Big Data Analysis with Apache Spark UC#BERKELEY2 This Lecture Course Objectives and Prerequisites Brief History of Data Analysis Correlation, Causat...
1 Big Data Analytics with Apache Spark (Part 2) Nastaran Fatemi2 What we ve seen so far Spark s Programming Model We saw that, at a glance, Spark look...
1 Big Data Analytics using Apache Hadoop and Spark with Scala Training Highlights : 80% of the training is with Practical Demo (On Custom Cloudera and...
1 Big Data Analysis with Apache Spark UC#BERKELEY2 This Lecture: Relation between Variables An association A trend» Positive association or Nega...
1 Data Processing with Apache Beam (incubating) and Google Cloud Dataflow Jelena Pjesivac-Grbovic Staff software engineer Cloud Big Data In collaborat...
1 BIG DATA ANALYTICS USING HADOOP TOOLS APACHE HIVE VS APACHE PIG Prof R.Angelin Preethi #1 and Prof J.Elavarasi *2 # Department of Computer Science, ...
1 Cassandra multi-datacenter operations essentials Apache: Big Data Vancouver, CA Julien Anguenot2 agenda key notions configuration and tuning tools a...
1 DOI: / This open source computing framework unifies streaming, batch, and interactive big data workloads to unlock new applications. BY MATEI ZAHARI...
Standards-Compliant Cloud Orchestration with
Apache: Big Data North America, May 2017, Miami Presented by Tal Liron, GigaSpaces
Farm to Table
MyCorpRoom1 Cloud Service Archive: ● ●
TOSCA YAML file Artifacts: ● Orchestration scripts ● VM images
Service Template LoadBalancer Template FrontendVM Template VideoStreamer Template
AriaTosca AuthenticationServer Template
Service Instance LoadBalancer1
Internal Scale: Big vs. Small
External Scale: Number of Services in Enterprise MyCorpRoom1
Meta-External Scale: Number of Services for ISPs Customer123-Room1
Meta-External Scale: Number of Services for NFV vCPE-12345678
Who? (As of May 2017) ASG, Bank of America, Brocade, CA, CENX, Cisco, Cloudsoft, CSI, FastConnect, Fujitsu, GigaSpaces, Google, HP, Huawei, IBM, iMinds, Intel, JPMorgan Chase, Mitre, NetApp, NetCracker, NIST, Nokia, Politecnico di Milano, Primeton, Quali, Red Hat, SAP, SINTEF, UNIPI, VMware, Vnomic, WIPRO, Yanna, Zenoss, ZTE ● ● ● ● ● ●
Cloud providers ISPs and mobile operators Enterprises (banks, government) Compute infrastructure Network infrastructure Agile innovators
When? ● ● ● ● ● ● ● ● ● ● ● ●
1993: “SGML Open” in Massachusetts 1998: Renamed to “OASIS”: change of emphasis to XML 1999: ebXML (with UN) 2005: ODF (clash with Microsoft’s Open Document) 2011: Call for participation for TOSCA 2012: GigaSpaces releases Cloudify 1.0 2012: First draft of TOSCA 1.0 (in XML) 2015: First draft of TOSCA Simple Profile in YAML 1.0 2016: TOSCA 1.0 released, first draft of TOSCA Simple Profile in YAML 1.1 2016: AriaTosca joins Apache Software Foundation as incubator project, 2016: AriaTosca joins OPEN-O (Linux Foundation) 2017: AriaTosca joins ONAP = OPEN-O + OpenECOMP (Linux Foundation)
Operas, Orchestras, and … Napoleon? TOSCA = Topology and Orchestration Specification for Cloud Applications Tosca was an Italian opera by Giacomo Puccini that premiered in 1900. Based on Victorien Sardou’s French play about the Kingdom of Naples’ defense of Rome from Napoleon in 1800. It contains depictions of torture, murder and suicide, as well as some of Puccini’s best-known lyrical arias. (Wikipedia) ARIA = Agile Reference Implementation of Automation Aria is an expressive melody, usually, but not always, performed by a singer. A self-contained piece for one voice, normally part of a larger work. (Wikipedia)
The TOSCA Balancing Act ●
Strict ○ ○ ○
Object-oriented, strictly typed, polymorphic Rich set of base types (Simple Profile + Simple Profile for NFV) Normative lifecycle (install, uninstall, start, stop)
Agnostic ○ ○ ○ ○ ○
Not specific to any cloud provider (multi-VIM is hard) Not specific to any machine technology (tosca.nodes.Compute = VM, container, cage, or…?) Base types designed to be lowest-common denominators (politics) Base types are optional Support for generic workflows in TOSCA 1.1
AriaTosca ● ● ● ● ●
Fanatic compliance with TOSCA spec (1.0, 1.1) Python 2.6 Rich CLI Rich API (Python and REST) SQL modeling ○
Converts to JSON or YAML
Batteries included: execute scripts locally and remotely, support dry runs Extensible plugin system ○ ○ ○
VIMs: CloudStack, OpenStack, VMware, AWS, Azure Orchestration: Puppet, Chef, Juju, Ansible, BPML/WSO2, Jenkins Custom: plugin can be included as artifact in .csar file
An AriaTosca Future ●
Rich plugin ecosystem ○
Community repository of .csar files ○
● ● ●
Every Apache Software Foundation project should publish one or more .csar files
Integrated into every cloud admin (native speaking of TOSCA) Cloud-based integration testing TOSCA IDE ○ ○ ○
Integration with open and proprietary orchestration systems
Autocompletion Inline validation Graphical modeling studio
Contribute to TOSCA evolution in OASIS
Thank You Today’s presenter: Tal Liron, GigaSpaces [email protected] For more information: http://ariatosca.org/ https://github.com/apache/incubator-ariatosca