Resource Representations in GENI: A path forward Ilia Baldine, Yufeng Xin Renaissance Computing Institute,

Slides:



Advertisements
Similar presentations
Network II.5 simulator ..
Advertisements

Network Systems Sales LLC
Proposal by CA Technologies, IBM, SAP, Vnomic
ExoGENI Rack Architecture Ilia Baldine Jeff Chase Chris Heermann Brad Viviano
Omniran TG 1 Cooperation for OmniRAN P802.1CF Max Riegel, NSN (Chair OmniRAN TG)
PlanetLab Architecture Larry Peterson Princeton University.
The Case for Enterprise Ready Virtual Private Clouds Timothy Wood, Alexandre Gerber *, K.K. Ramakrishnan *, Jacobus van der Merwe *, and Prashant Shenoy.
GIMI I&M and Monitoring Mike Zink, Max Ott, Ilya Baldine University of Massachusetts Amherst GEC 18, Brooklyn, October 27 st 1.
PlanetLab Operating System support* *a work in progress.
ExoGENI Racks Ilia Baldine
ORCA Overview LEARN Workshop Ilia Baldine, Anirban Mandal Renaissance Computing Institute, UNC-CH.
GEC21 Experimenter/Developer Roundtable (Experimenter) Paul Ruth RENCI / UNC Chapel Hill
ORCA-BEN Spiral 1 Status Yufeng Xin, Ilia Baldine Renaissance Computing Institute Jeff Chase Duke University
Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.
© 2007 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.1 Computer Networks and Internets with Internet Applications, 4e By Douglas.
1 GENI: Global Environment for Network Innovations Jennifer Rexford Princeton University
Network Rspecs in PlanetLab and VINI Andy Bavier PL Developer's Meeting May 13-14, 2008.
A Survey of Wireless Sensor Network Data Collection Schemes by Brett Wilson.
Data Center Basics (ENCS 691K – Chapter 5)
©Silberschatz, Korth and Sudarshan18.1Database System Concepts Centralized Systems Run on a single computer system and do not interact with other computer.
Using the jFed tool to experiment from zero to hero Brecht Vermeulen FGRE, July 7 th, 2015.
Microsoft Virtual Academy Module 4 Creating and Configuring Virtual Machine Networks.
Additional SugarCRM details for complete, functional, and portable deployment.
Ilya Baldin 2.
Introduction to the Mobile Security (MD)  Chaitanya Nettem  Rawad Habib  2015.
SILO: A novel framework for flexible protocol composition
National Science Foundation Arlington, Virginia January 7-8, 2013 Tom Lehman University of Maryland Mid-Atlantic Crossroads.
Resource Representations in GENI Rob Sherwood, OpenFlow Hongwei Zhang, Wireless sensor network description language Ilia Baldine, Yufeng Xin, Semantic.
Sponsored by the National Science Foundation Campus/Experiment Topics in Monitoring and I&M GENI Engineering Conference 15 Houston, TX Sarah Edwards Chaos.
Common Devices Used In Computer Networks
Sponsored by the National Science Foundation Research & Experiments on GENI GENI CC-NIE Workshop NSF Mark Berman, Mike Zink January 7,
Chapter 6 Operating System Support. This chapter describes how middleware is supported by the operating system facilities at the nodes of a distributed.
Sponsored by the National Science Foundation Programmable Networks and GENI Marshall Brinn, GPO GEC October 25, 2012.
GEC3 – October 28-30, 20081www.geni.net1 Substrate WORKING GROUP System Engineering Report John Jacob SWG System Engineer groups.geni.net GENI working.
GEC3www.geni.net1 GENI Spiral 1 Control Frameworks Global Environment for Network Innovations Aaron Falk Clearing.
GEC 15 Houston, Texas October 23, 2012 Tom Lehman Xi Yang University of Maryland Mid-Atlantic Crossroads (MAX)
Sponsored by the National Science Foundation GEC14 Session: SDN * in GENI Marshall Brinn, GPO July 11, 2012 * Software-Defined Networking.
Sponsored by the National Science Foundation GENI Exploring Networks of the Future
Sponsored by the National Science Foundation GENI Exploring Networks of the Future Sarah Edwards, GPO
OIF NNI: The Roadmap to Non- Disruptive Control Plane Interoperability Dimitrios Pendarakis
Sponsored by the National Science Foundation ExoGENI
Ilia Baldine, Jeff Chase, Mike Zink, Max Ott.  14 GPO-funded racks ◦ Partnership between RENCI, Duke and IBM ◦ IBM x3650 M3/M4 servers  1x146GB 10K.
Sponsored by the National Science Foundation Cluster D Working Meetings GENI Engineering Conference 5 Seattle, WA July ,
Sponsored by the National Science Foundation GENI Exploring Networks of the Future Sarah Edwards, GPO
LAMP: Bringing perfSONAR to ProtoGENI Martin Swany.
Resource representations in GENI workshops (GEC[78]) Ilia Baldine.
Sponsored by the National Science Foundation Meeting Introduction: Integrating GENI Networks with Control Frameworks Aaron Falk GENI Project Office June.
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 Based upon slides from Jay Lepreau, Utah Emulab Introduction Shiv Kalyanaraman
Sponsored by the National Science Foundation GENI Aggregate Manager API Tom Mitchell March 16, 2010.
Sponsored by the National Science Foundation 1 Nov 4, 2010 Cluster-D Mtg at GEC9 Tue, Nov 2, 12noon – 4:30pm Meeting Chair: Ilia Baldine (RENCI) –System.
SOFTWARE DEFINED NETWORKING/OPENFLOW: A PATH TO PROGRAMMABLE NETWORKS April 23, 2012 © Brocade Communications Systems, Inc.
Sponsored by the National Science Foundation GENI SDN Offering Marshall Brinn, GPO GEC18: October 28, 2013.
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
GEC22 Experimenter/Developer Roundtable (Developer) Victor Orlikowski Duke University
Virtualization One computer can do the job of multiple computers, by sharing the resources of a single computer across multiple environments. Turning hardware.
Web Technologies Lecture 13 Introduction to cloud computing.
Sponsored by the National Science Foundation GENI Exploring Networks of the Future
Mid-Atlantic Crossroads (MAX) GENI Facility Status Update March 16, 2010 Tom Lehman Xi Yang Peter O'Neil Abdella Battou.
© 2012 Eucalyptus Systems, Inc. Cloud Computing Introduction Eucalyptus Education Services 2.
Trusted Virtual Machine Images the HEPiX Point of View Tony Cass October 21 st 2011.
SEMINAR ON.  OVERVIEW -  What is Cloud Computing???  Amazon Elastic Cloud Computing (Amazon EC2)  Amazon EC2 Core Concept  How to use Amazon EC2.
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING CLOUD COMPUTING
Using the jFed tool to experiment from zero to hero
Stitching: the ORCA View
GGF15 – Grids and Network Virtualization
Network+ Guide to Networks 6th Edition
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Concept of VLAN (Virtual LAN) and Benefits
Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience Lab
GENI Exploring Networks of the Future
Presentation transcript:

Resource Representations in GENI: A path forward Ilia Baldine, Yufeng Xin Renaissance Computing Institute, UNC-CH

Slicing of a network

Link slivering

Agreements – resource representation cycle ExperimenterCF Substrate provider/AM CF 4 Possibly fuzzy request More specific request to substrate provider(s) Detailed manifest from substrate provider Collective possibly filtered manifest Ads from substrate providers

PG Rspec V1 GENI Resource representation mechanisms ‘Traditional’ Network resources –Ethernet links, tunnels, vlans Edge compute/storage resources Measurement resources –Including collected measurement data objects Wireless resources –WiFi, WiMax, motes etc Lack of agreement on what resources represent will be a significant impediment to interoperability Agreement on a format is important, but can be dealt with on the engineering level 5 PL RSpec PG Rspec V2 ORCA NDL-OWL OMF

Network resources Key distinctions –Number of layers –Describing adaptations between layers –Syntax –Tools 6 PL PG NDL-OWL

Aside: why adaptations are critical? Network adaptations are part of the description of stitching capability Needed for properly computing paths between aggregates connected by network providers at different layers –E.g. if a host has an interface that has an Ethernet to VLAN adaptation, this interface is capable of stitching to vlans –Consistent way to describe connectivity across layers (tunnels, DWDM, optical) Also –Important for matching requests to substrate capabilities E.g. creating a VM is a process of ‘adaptation’ of real hardware to virtualization layer 7

Network resources: a practical solution Stay primarily within Ethernet layer Accept one format to be used between control frameworks Perform bi-directional format conversion –Only partial may be possible Hosted services that perform conversion on demand –E.g. NS2/RSpec v1 and v2 request converter to NDL-OWL – Works well for several types of links, nodes, interfaces, ip addresses and simple link attributes 8

Agreeing on wire format RSpec v2 with extensions is a viable possibility Conversion from RSpec v2 is relatively straightforward pending agreement on edge resources Conversion to RSpec v2 is likely to be partial but probably sufficient for the time being –Nothing below Layer2 is visible to experimenter 9

Next challenge: Edge compute resources Ads: –Aggregates can produce (adapt to) different types of nodes –E.g. PL VServer, PG bare hardware node types, ORCA’s Xen/KVM virtual machines (and hardware nodes) –Constraints are possible on disk, memory size, number and type of CPU cores –Properties may include location, ownership etc. –Note: internal topology may or may not be part of the site ad. E.g. clouds have no inherent topology that needs to be advertized Requests: –Based on constraints on type of node, disk, memory size, core type and count, location 10

Advertising edge resources A server can be an individual node or a cloud of servers A site may choose to advertise individual servers/nodes or server clouds –Clouds have no inherent topology, just constraints on the type of topology they can produce and adaptations for nodes Servers/nodes or server clouds are adaptable to different types of nodes distinguished by –Virtualization (Xen, KVM, VServer, None etc) –Possibly memory, disk sizes, core types and counts, OS 11

Requesting edge resources A request for a node specifies multiple constraints on that node –Type of virtualization preferred –Memory, disk size –CPU Type –Core count –OS Allows policy to pick best sites based on request and resource availability. 12

Semantic Shortcut examples Semantic shortcuts – PL node Virtualiztion: Vserver – Simple PG node Virtualization: None CPU type: x86 or ?? or ?? – EC2M1Small Virtualization: KVM or XEN CPU count: 1 Memory size: 128M – EC2M1Large Virtualization: KVM or XEN CPU count: 2 Memory size: 512M – PlanetLabCluster Produces PL Nodes – ProtoGeniCluster Produces PG nodes 13

Other considerations Emerging standards: –OVF – portable appliance images across heterogeneous environments –CF should be able to generate OVF based on combination of request data and substrate information Higher-level programming environments: –Google App Engine, AppScale –Distributed map/reduce –… 14

Next steps Align edge compute resource descriptions Enable conversion as a GENI-wide service Test full interoperation PG ORCA by GEC11 Get the conversation started on aligning abstraction models for for –Wireless resources Max Ott, Hongwei Zhang, ? –Storage (physical and cloud) Mike Zink, ? 15