Transcript Virtualization as Architecture

Virtualization as Architecture GENI
CSC/ECE 573, Sections 001, 002
Fall, 2012
Some slides from Harry Mussman, GPO
Network Integration

Vision of integrated services network
–

But – requirements are very different
–
–

Integrating networks requires making “greatest of all
networks” (ATM) rather than “least of all networks”
Raises barrier to entry
Separate networks are good
–

Single network infrastructure which carries traffic for
various types of use
For banking and videochat and telesurgery, e.g.
But frustrating that “solved” problems reappear,
old solutions cannot be easily applied
Motivation for Virtualization

Approach similar to compute virtualization
 A substrate that provides basic capabilities
 A method to identify smallest units (“slivers”) of
–
–
–

Resources that make up substrate must each be
sliverable
–

Easiest when slivering is along physical lines (NICs, switches)
Collection of slivers makes up a virtual network (“slice”)
–

Bandwidth
Switching
???
Similar to a virtual machine
Advantage of integrated network without (some of) the
drawbacks
GENI




In late 2000’s, an NSF initiative to create a nationalscale sharable network testbed
Allow researchers to experiment with a national “atscale” footprint
Allow experimentation with different architectures,
fundamentally incompatible
Virtualized underlying infrastructure indispensable for
such a testbed
–
–

Different experiments would be completely isolated
Would use completely different stacks, hops
Also the thought: maybe virtualization is the next
architecture
Global networks are creating
extremely important new challenges
Science Issues
We cannot currently
understand or predict the
behavior of complex,
large-scale networks
Innovation Issues
Substantial barriers to
at-scale experimentation with
new architectures, services,
and technologies
Society Issues
Credit: MONET Group at UIUC
We increasingly
rely on
the Internet but are unsure
that can trust its security,
privacy or resilience
Sponsored by the National Science Foundation
March 31, 2009
5
GENI Conceptual Design
Infrastructure to support at-scale experimentation
Virtualized
Deeply programmable
Programmable & federated, with end-to-end virtualized “slices”
Sensor Network
Federated
International
Infrastructure
Mobile Wireless Network
Sponsored by the National Science Foundation
March 31, 2009
Edge Site
Heterogeneous,
and evolving over time via
spiral development
6
Federation
GENI grows by “gluing together” heterogeneous infrastructure
My experiment runs across
the evolving GENI federation.
Wireless
#1
Corporate
GENI suites
Backbone #1
Compute
Cluster
#1
Compute
Cluster
#2
My GENI Slice
Access
#1
Backbone #2
NSF parts of GENI
Other-Nation
Projects
Other-Nation
Projects
This approach looks
remarkably familiar . . .
Wireless
#2
Goals: avoid technology “lock in,” add new technologies as they mature, and potentially
grow quickly by incorporating existing infrastructure into the overall “GENI ecosystem”
Sponsored by the National Science Foundation
March 31, 2009
7
GENI System Decomposition (simplified)
Engineering analysis drives Spiral 1 integration
GENI Admin and Ops Org
GENI Clearinghouse
<
Register
<
Admin and Account
Trust
<
Slice
Create
Principal
Registry
<
Register
Slice
Registry
EU Clearinghouse
(federated
network
example)
(Aggr)
Comp
Registry
EU
Admin
and
Ops
Org
< Admin
=<
Ticket
Broker
<
Authen
< Operator
<
Ops and Mgmt
Ticket
Log
=<
View
Help
Desk
=<
View
=<
View
Trust
Research
Org A
< Slice
Admin
= PI
= Research
=
Researcher
Helper
Tools
GENI Programmable
Host Cluster A
GENI Programmable Ntwk
Routing (Switch) Node B
GENI Metro (Sensor)
Wireless Ntwk C
GENI Enterprise (Resident)
Access Ntwk D
GENI Regional (National)
Optical Ntwk E
< Com
Admin
< Com
Admin
< Com
Admin
< Com
Admin
< Com
Admin
Comp
Operator
Comp
Operator
Comp
Operator
Comp
Operator
=<
Component Mgr
=<
Component Mgr
=<
Component Mgr
=<
Component Mgr
<
Ops Portal
<
Ops Portal
<
Ops Portal
<
Ops Portal
<
Ops Portal
Host
Ax
Node B
Ntwk C
Ntwk D
Ntwk E
PoP
Research
Org B
Comp
Operator
=<
Component Mgr
Host
A1
EU
Comp
AA
PoP
Experiment Plane
Measurement Plane
= Control Plane
< Ops and Mgmt Plane
Sponsored by the National Science Foundation
March 31, 2009
8
Resource discovery
Aggregates publish resources, schedules, etc., via clearinghouses
What resources can I use?
GENI
Clearinghouse
These
Researcher
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Computer Cluster
Backbone Net
Metro Wireless
Sponsored by the National Science Foundation
March 31, 2009
9
Slice creation
Clearinghouse checks credentials & enforces policy
Aggregates allocate resources & create topologies
Create my slice
GENI
Clearinghouse
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Computer Cluster
Backbone Net
Metro Wireless
Sponsored by the National Science Foundation
March 31, 2009
10
Experimentation
Researcher loads software, debugs, collects measurements
Experiment – Install my software,
debug, collect data, retry, etc.
GENI
Clearinghouse
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Computer Cluster
Backbone Net
Metro Wireless
Sponsored by the National Science Foundation
March 31, 2009
11
Slice growth & revision
Allows successful, long-running experiments to grow larger
Make my slice bigger !
GENI
Clearinghouse
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Computer Cluster
Backbone Net
Metro Wireless
Sponsored by the National Science Foundation
March 31, 2009
12
Federation of Clearinghouses
Growth path to international, semi-private, and commercial GENIs
Make my slice even bigger !
GENI
Clearinghouse
Federated
Clearinghouse
Components
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Aggregate D
Computer Cluster
Backbone Net
Metro Wireless
Non-NSF Resources
Sponsored by the National Science Foundation
March 31, 2009
13
Operations & Management
Always present in background for usual reasons
Will need an ‘emergency shutdown’ mechanism
Stop the experiment
immediately !
GENI
Clearinghouse
Oops
Federated
Clearinghouse
Components
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Aggregate D
Computer Cluster
Backbone Net
Metro Wireless
Non-NSF Resources
Sponsored by the National Science Foundation
March 31, 2009
14
Spiral 1 integration and trial operations
Five competing control frameworks, wide variety of substrates
Reference
Design
Cluster A
Cluster B
Components
Components
Components
Aggregate A
Aggregate B
Aggregate C
Computer
Cluster
Backbone Net
Metro Wireless
Components
Components
Components
Aggregate
A1
Aggregate
A2
Aggregate
A3
Computer
Cluster
Optical Network
Metro Wireless
Components
Cluster C
Components
Components
Cluster D
Components
Cluster E
Components
Aggregate
C1
Aggregate
C2
Aggregate
D1
Aggregate
D2
Computer
Cluster
Programmable
Switches
Optical Network
Sensor Network
Sponsored by the National Science Foundation
Components
Components
Components
Aggregate
B1
Aggregate
B2
Optical Network
Sensor Network
Components
Components
Aggregate
E1
Aggregate
E2
Aggregate
E3
Aggregate
E4
Computer
Cluster
Optical Network
Sensor Network
Programmable
Switches
March 31, 2009
15
World-class expertise in GENI Partners
Internet2 and National Lambda Rail
Internet2
10 Gbps dedicated bandwidth
National Lambda Rail
Up to 30 Gbps nondedicated bandwidth
40 Gbps capacity for GENI prototyping on two national footprints
to provide Layer 2 Ethernet VLANs as slices (IP or non-IP)
Sponsored by the National Science Foundation
March 31, 2009
16
Summary

GENI has completed Spirals 1 and 2, and is starting
Spiral 3
 Original thinking and positioning has been questioned
and revisited
 GENI research council has been set up
 Architectural vision also evolved – common Aggregate
Manager API
 Overall broad goal remains to enable isolated
experiments deep into the network stack
 Future Internet architectural insights and/or partial
realizations might emerge