NSF/Mideast Workshop
Future Internet Architectures Panel
Convener: Zhi-Li Zhang University of Minnesota
Panelists
• Jeff Chase, Duke University • Sonia Fahmy, Purdue University • George Kesidis, Penn State University • Taieb Znati, Pittsburgh University • Zhi-Li Zhang, University of Minnesota
Internet: Past & Now • From the original 4-node ARPANet (in 1969)
– underwent a few transformations
• to today’s “hourglass” Internet architecture – based on TCP/IP (+ DNS & BGP) as the core networking protocols
• Original Internet Design Goals: David Clark [Sigcomm88] In the order of importance:
0 Connect existing networks 1. Survivability 2. Support multiple types of services 3. Must accommodate a variety of networks 4. Allow distributed management 5. Allow host attachment with a low level of effort 6. Be cost effective 7. Allow resource accountability
What Has Become of Internet • Information Service Platform
– deliver all kinds of information (web, iTune, YouTube, Netflix, …)
• Global Information Repository – store and search for all kinds of information (e.g., Dropbox)
• Cyberspace and Virtual Communities – keep in touch with friends and strangers (e.g., Facebook, Twitter)
• Enormous Super-Computer – cloud & mobile computing and services
• What’s coming: Internet of Things Ø … we increasingly depend on it!
Diverging Trends … • Internet Core: concentration
– high bandwidth, dense connectivity – data centers: computing, storage, networking, …
• Internet Edges: diversification – “smart” to “dumb” devices
• PCs with significant processing and storage capacities • small or mobile devices with limited computing, memory, power, …
– broadband to narrowband – “always on” to intermittent connectivity
Challenges and Opportunities! • overcome heterogeneity, seamlessly integrate • new services & “disruptive” technologies
Within the Internet Core • Large ISPs with large
geographical span and • Large content providers
with huge data centers
• High capacity, dense and rich topology
• Cloud Computing/Services and Mobile Computing
On the Internet Edge … • Large number of mobile
users • Large number of “dumb”
or “smart” devices and appliances, some resource constrained
• Intermittent connectivity with varying bandwidth
• Diverse applications and services
• Heterogeneous technologies
Internet Home users
Banking & e-commerce
dumb & smart phones
POTS
VoIP
Multimedia Streaming
Games
Surveillance & Security
Online TV Web/emails
Challenges Facing Today’s Internet • Scalability: capability to connect tens of thousands, millions or
more users and devices – routing table size, constrained by router memory, lookup speed
• Availability & Reliability: must be resilient to failures – need to be “proactive” instead of reactive; need to localize effect of failures
• Mobility: users and hosts/servers are more mobile – need to separate location (“addressing”) and identity (“naming”)
• Manageability: ease of deployment, “plug-&-play” – need to minimize manual configuration – self-configure, self-organize, while ensuring security and trust
• Security & Privacy: – in addition to encryption, etc, how to distinguish “good” guys from “bad”
guys à need a “social, behavioral & economic” perspectives!
• Economic Viability – various stakeholders, often with shared but also competing interests
Challenges Facing Today’s Internet • Scalability: capability to connect tens of thousands, millions or
more users and devices – routing table size, constrained by router memory, lookup speed
• Availability & Reliability: must be resilient to failures – need to be “proactive” instead of reactive; need to localize effect of failures
• Mobility: users and hosts/servers are more mobile – need to separate location (“addressing”) and identity (“naming”)
• Manageability: ease of deployment, “plug-&-play” – need to minimize manual configuration – self-configure, self-organize, while ensuring security and trust
• Security & Privacy: – in addition to encryption, etc, how to distinguish “good” guys from “bad”
guys à need a “social, behavioral & economic” perspectives!
• Economic Viability – various stakeholders, often with shared but also competing interests
Internet: critical global information infrastructure,
big, complex, massively distributed, and changing!
US NSF “Future Internet Architectures” Initiatives
Started circa 2006, two phases • Phase I: FIND (Future Internet Network Design) Initiative
– A number of small and medium-size projects funded – See http://www.nets-find.net
• Phase 2: FIA (Future Integrative Architectures) Initiative – Four large multi-institution projects funded
• eXpressive Internet Architecture (PI: Peter Steenkiste, CMU)
• MobilityFirst (PI: Dipankar Raychaudhuri, Rutgers U.)
• Named Data Networking (PI: Lixia Zhang, UCLA)
• NEBULA (PI: Jonathan Smith, U. of Pennsylvania)
– See http://www.nets-fia.net • Separately, GENI Initiative (serving as testbed?)
Why Research on “Future/New Internet Architectures”
My personal perspective:
• Many short-term “fixes/patches” have been developed/applied – fix some problems but introduce others; e.g., NAT, firewalls – also make things more complex and error-prone (esp. net config.)
• Certain limitations of the Internet architecture require radical changes and long-term solutions – need “out-of-the-box” re-thinking of network architectures – where the (academic) research community can play a significant role!
• “Clean-slate” (re-)designs of Internet architectures – unconstrained by the current Internet’s “idiosyncrasies” – unencumbered by “conventional wisdoms”
Panelists
• Jeff Chase, Duke University • Sonia Fahmy, Purdue University • George Kesidis, Penn State University • Taieb Znati, Pittsburgh University • Zhi-Li Zhang, University of Minnesota
NSF/Mideast Workshop New Internet Architectures Panel
VIRO:
Scalable, Robust & Name-Independent Virtual Id Routing
for (future) Large-scale, Dynamic Networks
Zhi-Li Zhang Qwest Chair Professor
Department of Computer Science and Engineering University of Minnesota
Email: [email protected]
Designed to Meet Challenges posed by Large, Dynamic Networks (e.g., Data Center Networks)
• Scalability: capability to connect tens of thousands, millions or more users and devices – routing table size, constrained by router memory, lookup speed
• Mobility: hosts are more mobile – need to separate location (“addressing”) and identity (“naming”)
• Availability & Reliability: must be resilient to failures – need to be “proactive” instead of reactive – need to localize effect of failures
• Manageability: ease of deployment, “plug-&-play” – need to minimize manual configuration – self-configure, self-organize, while ensuring security and trust – Agility: dynamically adapt to demand
• ......
Pros & Cons of Existing Technologies
• (Layer-2) Ethernet/Wireless LANs u Pluses:
• plug-&-play, minimal configuration, better mobility
u Minuses: • (occasional) data plane
flooding, sub-optimal routing (using spanning tree), not robust to failures
• Not scalable to large (& wide-area) networks
– IETF TRILL
q (Layer-3) IPv4/IPv6 ¤ Pluses:
• better data plane scalability, more “optimal” routing, …
¤ Minuses: • control plane flooding, global effect of
network failures • poor support for mobility • difficulty/complexity in “network
renaming” • Esp., changing addressing schemes
(IPv4 -> IPv6 transition) requires modifications in routing and other network protocols
Meeting the Challenges: VIRO: A Scalable, Robust, Namespace-
Independent, “Plug-&-Play” Routing Architecture • Decoupling routing from naming/”addressing”
– “native” naming/address-independent • “future-proof” & capable of supporting multiple namespaces
• Introduce a “self-organizing” virtual id (vid) layer – a layer 2 (LLC)/layer-3 convergence layer – subsume layer-2/layer-3 routing/forwarding functionality
• except for first/last hop: host to switch or switch to host • layer-3 addresses (or higher layer names): global addressing or naming for
inter-networking and “persistent” identifiers
l DHT-style routing using a topology-aware, structured vid space • highly scalable and robust: going beyond shortest-path routing, with built-
in multi-path & fast rerouting capabilities, – O(log N) routing table size, localize failures, enable fast rerouting
• support multiple topologies or virtualized network services
Virtual ID layer and VID space • Topology-aware, structured virtual id (vid) space
– embed physical topology in a Kademlia-like “virtual” binary tree – virtual id’s (vid’s): encode location of nodes/switches, i.e., “locators” – self-configurable and self-organizing – support (interoperability of) multiple namespaces & multiple virtual nets
Layer 2 Physical Network Topology
IPv4/IPv6
Virtual ID Layer
Other Namespaces DNS Names
M N H
G J L K
C
F E B
D A 1
1
1
1 1
0
0
0 0 0 1 1
0
0
1
1 0
0
0
0
1
1
1
1
1
0 0
-‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ -‐ F E H G B A D C N M J L K
1
0
0 1
0 0 1
1
0 0 0 0 0 1 1 1 1 1 1
VIRO: Three Core Components • Virtual id space construction and vid assignment
– performed most at the bootstrap process (i.e., network set up): • a vid space “skeleton” is created
– once network is set up/vid space is constructed: • a new node (a “VIRO switch”) joins: assigned based on neighbors’ vid’s • end-host/device: inherits a vid (prefix) from “host switch” (to which it is attached), plus
a randomly assigned host id; host may be agnostic of its vid
• VIRO routing algorithm/protocol: – DHT-style, but needs to build end-to-end connectivity/routes
• a bottom-up, round-by-round process, no network-wide control flooding • O(log N) routing entries per node, N: # of VIRO switches
l (Persistent) layer-2/3 address/name resolution and vid look-up – DHT directory services built on top of the same vid space
• “persistent” identifier (e.g., MAC/IP address) hashed to a “vid” key, which is then used for (pid, vid) mapping registration, look-up, etc.
l Data forwarding among VIRO switches using vid only
Summary • VIRO provides a scalable & robust substrate for future networks • Enables (nearly) configuration-free networks • Support for multiple namespaces • Support mobility, multiple topologies, virtualized network
services, security • Backward compatibility: compatible with current host protocols
(such as ARP etc) • Ongoing & Future work:
– prototyping using Click and Openflow – virtualized services, inter-domain routing issues
Please visit http://networking.cs.umn.edu/newsite/veil-wiro
for: demo videos, List of related publications, source code, or simply search online for “VIRO VEIL”