Networking Named Content
By Van Jacobson, Diana K. Smetters, James D. Thornton, Michael Plass, Nick Briggs, and Rebecca Braynard
Current network use is dominated by content distribution and retrieval yet current networking protocols are
designed for conversations between hosts. Accessing
content and services requires mapping from the what
that users care about to the network’s where. We present
Content-Centric Networking (CCN) which uses content
chunks as a primitive—decoupling location from identity,
security and access, and retrieving chunks of content by
name. Using new approaches to routing named content,
derived from IP, CCN simultaneously achieves scalability,
security, and performance. We describe our implementation of the architecture’s basic features and demonstrate
its performance and resilience with secure file downloads
and VoIP calls.
The engineering principles and architecture of today’s
Internet were created in the 1960s and 1970s. The problem networking aimed to solve was resource sharing—
remotely using scarce and expensive devices like card
readers or high-speed tape drives or even supercomput-ers. The communication model that resulted is a conversation between exactly two machines, one wishing to use
the resource and one providing access to it. Thus IP packets contain two identifiers (addresses), one for the source
and one for the destination host, and almost all the traffic on the Internet consists of TCP conversations between
pairs of hosts.
In the 50 years since the creation of packet networking, computers and their attachments have become cheap,
ubiquitous commodities. The connectivity offered by the
Internet and low storage costs enable access to a staggering
amount of new content—500EB were created in 2008 alone. 7
People value the Internet for what content it contains, but
communication is still in terms of where.
We see a number of issues that affect users arising from
this incompatibility between models.
•;Availability: Fast, reliable content access requires awkward, pre-planned, application-specific mechanisms
like CDNs and P2P networks, and/or imposes excessive
•;Security: Trust in content is easily misplaced, relying on
untrustworthy location and connection information.
•;Location-dependence: Mapping content to host locations
complicates configuration as well as implementation of
The direct, unified way to solve these problems is to
replace where with what. Host-to-host conversations are a
networking abstraction chosen to fit the problems of the
1960s. We argue that named data is a better abstraction
for today’s communication problems than named hosts.
We introduce Content-Centric Networking (CCN), a communications architecture built on named data. CCN has no
notion of host at its lowest level—a packet “address” names
content, not location. However, we preserve the design
decisions that make TCP/IP simple, robust, and scalable.
Figure 1 compares the IP and CCN protocol stacks.
Most layers of the stack reflect bilateral agreements; e.g., a
layer 2 framing protocol is an agreement between the two
ends of a physical link and a layer 4 transport protocol is
an agreement between some producer and consumer. The
only layer that requires universal agreement is layer 3, the
network layer. Much of IP’s success is due to the simplicity
of its network layer (the IP packet—the thin “waist” of the
stack) and the weak demands it makes on layer 2, namely:
stateless, unreliable, unordered, best-effort delivery. CCN’s
network layer (Section 3) is similar to IP’s and makes fewer
demands on layer 2, giving it many of the same attractive
properties. Additionally, CCN can be layered over anything,
including IP itself.
CCN departs from IP in a number of critical ways. Two
of these, strategy and security, are shown as new layers in
its protocol stack. CCN can take maximum advantage of
multiple simultaneous connectivities (e.g., ethernet and
3G and bluetooth and 802.11) due to its simpler relationship with layer 2. The strategy layer (Section 3. 3) makes
the fine-grained, dynamic optimization choices needed
to best exploit multiple connectivities under changing
conditions. CCN secures content itself (Section 4), rather
than the connections over which it travels, thereby avoiding many of the host-based vulnerabilities that plague IP
figure 1. CCn moves the universal component of the network stack
from iP packets to named content chunks.
email WWW phone ...
SMTP HTTP RTP ...
TCP UDP ...
Ethernet PPP ...
Copper fiber radio ...
CSMA async sonet ...
File Stream ...
Browser chat ...
Copper fiber radio ...
IP UDP P2P BCast ...
A previous version of this paper was published in
Proceedings of ACM's CoNEXT Conference 2009 (Rome,
Italy, Dec. 1–4, 2009), ACM, NY.