Internet DRAFT - draft-nordman-classification
draft-nordman-classification
Network Working Group B. Nordman
Internet-Draft Lawrence Berkeley National
Intended status: Standards Track Laboratory
Expires: April 26, 2012 October 24, 2011
Basic Device Classification
draft-nordman-classification-00
Abstract
This specification addresses how to communicate a basic sense of the
type of each device present on the network. This is particularly
important as the range of devices with connectivity greatly expands,
and as indirect interests in device characteristics increase, such as
energy consumption. Many applications will benefit from a single
standard enumeration of basic device types. This draft does not
address detailed device characteristics or subtypes.
The draft also discusses related identify information. It is an
initial discussion document to generate feedback and improvement.
Status of this Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
This Internet-Draft will expire on April 26, 2012.
Copyright Notice
Copyright (c) 2011 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
Nordman Expires April 26, 2012 [Page 1]
Internet-Draft Device Classification October 2011
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
1. Overview
As the number of types of devices that are connected to the Internet
increases, it will be increasingly necessary for devices to make
decisions based on the types of devices they encounter. In
particular, devices may be able to discover what other devices share
the same physical space, so that the range of devices that they find
may be large. Basic decisions about whether or not to respond to a
device can be informed by understanding the fundamental nature of
each device. Detailed understanding likely requires device-specific
information; this draft does not attempt to do this; it only covers a
first layer of classification.
Classification is proposed to be represented as a 2-byte value that
corresponds to an IANA registry of devices.
Other information also contributes to the identity of a device, such
as brand and model, and would benefit from consistency in naming and
availability.
2. Terminology
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [RFC2119].
3. Other Relevant Standards
There are many standards that touch on topics related to this one,
though generally they have a different specific purpose, or address
just a small subset of the devices in scope for this standard. Thus,
while they should inform the discussion, none are suitable as a
replacement for or substantial basis for this one.
Some example other standards and classifications are:
Nordman Expires April 26, 2012 [Page 2]
Internet-Draft Device Classification October 2011
* The Distributed Management Task Force (DMTF) defines a set of
entity times in its Common Information Model (CIM). These include
both entire devices and components of devices (components are out
of scope of this proposal).
* Systems for labeling retail products (e.g. Universal Product
Code, UPC) embody a detailed listing of product types.
* The United Nations publishes a Standard Products and Services
Code, for use in facilitating electronic commerce.
* There are efforts to develop standard taxonomies for heating and
cooling systems in large commercial buildings. This is
particularly needed since that industry has been long dominated by
proprietary protocols, each with their own internal taxonomy.
* Some companies have implemented information systems to describe
devices. An example is the Cisco Products MIB, which provides for
a device to report its model number.
4. Classification and Identity
There are many standards that speak to identity on the network, or
service discovery. This document does not address those topics.
Rather, the classification addressed is comparable to the first
impression a human being might have on a device, to recognize its
core function or type.
4.1. General Issues
An analogy can be drawn to the ASCII character coding system. It
specifies a single simple and compact numeric correspondence for
letters, numbers, and symbols. It makes no attempt to cover
characteristics other than the character's identity such as font,
size, etc. Because of its simplicity, it can be utilized
universally, or nearly so.
The concept of identity is usually for information which is unique to
the entity at hand, rather than data which puts it into a group of
entities.
Classification of populations of entities is commonly done with a
taxonomy; a system of organization or categories. Taxonomies often
have multiple layers of organization with groupings having one or
more common characteristics; the most widely known of these is the
biological classification system which has seven layers or grouping.
4.2. Registry Proposal
Classification should be a characteristic of a device that never
changes, though it may be unnecessary to prohibit this.
Nordman Expires April 26, 2012 [Page 3]
Internet-Draft Device Classification October 2011
Classification of a device is self-determined. It will be desirable
to have standard translations of the classification code into all
major languages.
A possible implementation of this is an IANA registry of 2-byte
values for class. Each device would be a member of at most one
class. The class for a particular device would be set by the
manufacturer.
This proposal is for a single listing of device types, as the choice
of criteria for grouping might change with the particular
application. As the number of device types anticipated is not so
large, an application can readily impose its own categorization
system on top of the basic classes.
Some devices have more than one major function (e.g. a combination
television and DVD player). While it would be possible to allow more
than one classification, this draft is crafted on the assumption that
this is unnecessary.
This mechanism is not intended to solve all problems with an
application understanding characteristics of devices it encounters.
There are many existing mechanisms for service discovery and the like
which enable much more detailed information, but none of these are
universal.
Categories would be drawn broadly. For example, a likely one is
"refrigerator" which would encompass any device that cools its inside
space regardless of size, technology, features, or intended market
(e.g. residential or commercial). Obtaining more detailed
information would require a different mechanism.
4.3. Application to Energy Management
One application that would benefit from such a classification system
is Energy Management. A management system for energy will gather up
data about all devices in a building, and have no functional
relationship to many of them. The energy information collected is
generic to any device type; the data are unrelated to the services
that the device provides functionally. Classification information
may be the only data that the management system has about the device
other than its identity on the network. Thus, any classification
(and identity) data should be universal. Also, many buildings will
be partly or entirely unmanaged, with some devices that come and go
asynchronously so that classification information should be gathered
automatically.
With classification data, a management system can readily and
Nordman Expires April 26, 2012 [Page 4]
Internet-Draft Device Classification October 2011
automatically provide a breakdown of energy use by major category of
device. A person may be investigating energy use in a building after
some devices have left the network, or in a situation where they lack
direct access to the network (an example of the latter is where the
analysis is outsourced to a third party). Thus, it is desirable that
the most critical information be available in what is reported during
energy management queries.
5. Related Needs
Beyond the basic class, it is often necessary or desirable to have
the manufacturer's brand name and model number. These can be used in
reports, or to help gather additional information about a device.
These are text strings of some modest length. A human-readable name
is often associated with such information. There are some existing
MIBs which have these variables.
Related to this, manufacturers often have a web page for information
about each model. It would be helpful if a device could report a URL
for such a page. Additionally, the page could have both human-
readable and machine-readable portions, with the latter specified
according to some standard format. This would enable useful
information about devices to be gathered automatically by a
management system.
Many applications that use classification will also want to know the
current location of a device, either geographically, or within a room
or building. This draft does not speak to location but it is
important.
6. Issues
Should there be a repository somewhere of standard translations from
IANA code to human languages? Where?
What issues are introduced when classification is proxied for non-IP
devices?
7. Security Considerations
In general none. It is worth noting that a device may report a type
other than what it is.
Nordman Expires April 26, 2012 [Page 5]
Internet-Draft Device Classification October 2011
8. Privacy Considerations
None.
9. Acknowledgement
We would like to thank <get your name here>.
10. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
Author's Address
Bruce Nordman
Lawrence Berkeley National Laboratory
1 Cyclotron Road
Berkeley, CA 94720
USA
Email: BNordman@LBL.gov
Nordman Expires April 26, 2012 [Page 6]
