Network Working Group P. Saint-Andre
Internet-Draft XMPP Standards Foundation
Intended status: Informational October 2, 2007
Expires: April 4, 2008
Interdomain Presence Scaling Analysis for the Extensible Messaging and
Presence Protocol (XMPP)
draft-saintandre-xmpp-presence-analysis-01
Status of this Memo
By submitting this Internet-Draft, each author represents that any
applicable patent or other IPR claims of which he or she is aware
have been or will be disclosed, and any of which he or she becomes
aware will be disclosed, in accordance with Section 6 of BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
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."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on April 4, 2008.
Copyright Notice
Copyright (C) The IETF Trust (2007).
Abstract
This document analyzes the scalability of presence sharing between
domains that federate using the Extensible Messaging and Presence
Protocol (XMPP). This analysis is provided as a source of comparison
with a similar analysis being performed regarding the presence
extensions to the Session Initiation Protocol (SIP).
Saint-Andre Expires April 4, 2008 [Page 1]
Internet-Draft XMPP Presence Analysis October 2007
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . 4
3. Protocol Flows . . . . . . . . . . . . . . . . . . . . . . . . 5
4. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.1. Constants . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2. Initial Messages . . . . . . . . . . . . . . . . . . . . . 7
4.3. Steady State Messages . . . . . . . . . . . . . . . . . . 8
4.4. Termination Messages . . . . . . . . . . . . . . . . . . . 8
4.5. Bottom Line . . . . . . . . . . . . . . . . . . . . . . . 9
5. Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1. Basic . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.2. Widely Distributed Inter-Domain Presence . . . . . . . . . 11
5.3. Very Large Network Peering . . . . . . . . . . . . . . . . 13
5.4. Intra-Domain Peering . . . . . . . . . . . . . . . . . . . 15
6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7. Security Considerations . . . . . . . . . . . . . . . . . . . 17
8. Informative References . . . . . . . . . . . . . . . . . . . . 17
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 18
Intellectual Property and Copyright Statements . . . . . . . . . . 19
Saint-Andre Expires April 4, 2008 [Page 2]
Internet-Draft XMPP Presence Analysis October 2007
1. Introduction
Presence is information about the network availability of an
individual (or, more precisely, of a presence address of the kind
that is often but not necessarily associated with an individual). As
typically designed and deployed, presence is shared only with
authorized entities, where the authorization takes the form of a
subscription. (In this document, we employ the term "user" to
signify an account that generates presence information and the term
"contact" to signify an annount that is subscribed to the user's
presence.)
The sharing of presence information can result in a large volume of
traffic as users log on or off throughout the life of a presence
session, especially for users with large numbers of contacts (e.g.,
the author of this document has over 1,500 contacts in his list of
presence subscribers). The volume is increased by communication of
information beyond basic on-off network availability, such as
availability substates (e.g., "away" and "do not disturb"). The
volume is further increased if the presence "transport" is used to
communicate information such as geolocation, mood, activity, even the
music to which an individual is listening. Such traffic may be a
concern even in a standalone presence domain. However, when presence
is shared across domain boundaries then such traffic may introduce a
more significant impact on the functioning of the Internet as a
whole. Therefore it is important to analyze the traffic generated
during interdomain communication of presence information.
There are several standardized technologies for sharing presence
information. One is a set of extensions (commonly called "SIMPLE")
to the Session Initiation Protocol (SIP), where the base protocol is
defined in [SIP] and the extensions are defined in [SIP-EVENT] and
[SIP-PRES]. Another is the Extensible Messaging and Presence
Protocol (XMPP) as defined in [XMPP-CORE] and [XMPP-IM].
[PROBLEM] analyzes several factors regarding the scalability of
interdomain communication of presence information using SIP/SIMPLE
technologies. For the sake of comparison, this document aims to
provide a similar analysis regarding XMPP technologies. In
particular, this document focuses on traffic load exclusively since
bandwidth usage has the greatest potential impact on the Internet.
By contrast, issues such as state management and server processing of
presence information are implementation-specific. This document also
briefly mentions existing methods for improving the scalability of
XMPP presence (and presence-like) communications.
Saint-Andre Expires April 4, 2008 [Page 3]
Internet-Draft XMPP Presence Analysis October 2007
2. Assumptions
The model for XMPP presence subscriptions is different from that of
SIP. In particular, XMPP presence subscriptions are long-lived, and
once established last until cancelled. Thus XMPP does not have
subscription timeouts and refresh periods as SIP presence does. In
addition, this document does not include presence subscriptions in
its protocol flows since in XMPP they are preconditions for the
exchange of presence notifications (in any case, the number of XML
stanzas exchanged in the process of establishing a presence
subscription is negligible compared to the volume of presence
notifications).
XMPP presence subscriptions are typically bidirectional (i.e., the
contact has a subscription to the user's presence and the user has a
subscription to the contact's presence). However, because [PROBLEM]
assumes that subscriptions are uni-directional (i.e., the contact has
a subscription to the user's presence but not vice-versa), the same
assumption is made herein.
Although an XMPP user or contact may have multiple connected
"resources" (e.g., client or device) at any one time, for the sake of
simplification this document assumes that each entity has only one
simultaneous resource.
Note that, unlike in SIP, XMPP packets are not typically acknowledged
with the equivalent of a 200/OK message.
[PROBLEM] assumes that presence notification packets will typically
be on the order of 3.5 kilobytes in size (not including TCP or UDP
overhead). XMPP presence notification packets tend to be much
smaller than SIP presence notification packets; in this document we
assume (based on deployment experience) that they are typically 200
bytes in size for basic on-off presence. However, some XMPP
applications may include additional information in a presence
notification packet, such as entity capabilities as described in
[XEP-0115].
Both [XMPP-CORE] and [XMPP-IM] strongly recommend that XMPP presence
notifications should include only information that is relevant to a
user's willingness or ability to communicate using real-time methods
such as instant messaging. However, some XMPP applications include
information that is not communications-relevant, such as the hash of
a user's avatar icon (see [XEP-0153] and even metadata about the
music to which a user is listening (see [XEP-0118]). Although it is
recommended to communicate such information using the XMPP publish-
subscribe extension (see [XEP-0060]) and appropriate profiles thereof
(e.g., [XEP-0163]), some existing XMPP clients send non-
Saint-Andre Expires April 4, 2008 [Page 4]
Internet-Draft XMPP Presence Analysis October 2007
communications-relevant information using presence notifications
instead of dedicated publish-subscribe nodes. Such behavior
marginally increases notification size but can drastically increase
the number of notifications sent (e.g., one notification every 3 or 4
minutes when the user begins listening to a new music track). This
document does not discuss such usage, since it is actively
discouraged and borders on abusive.
This document does not discuss various optimizations for SIMPLE (for
which see [PROBLEM]) or XMPP. The primary deployed optimization for
XMPP is stream compression, implemented either at the TLS level via
native TLS compression or at the XMPP level where TLS compression is
not available (see [XEP-0138]). Because XMPP communications occur
over long-lived TCP connections and associated long-lived XML
streams, such compression has been found to yield significant
bandwidth savings, up to 90% or even 95%. Stream compression is
therefore the recommended method for reducing bandwidth consumption
in XMPP systems.
3. Protocol Flows
When a contact (in these examples romeo@example.net) becomes
available, the contact's server sends an XMPP presence stanza of type
"probe" to the user (in these examples juliet@example.com) on behalf
of the contact, as shown in the following example (this can be seen
as similar to the initial SUBSCRIBE in SIP presence):
Contact's server sends presence probe to user (82 bytes):
If the user's server determines that the contact is authorized to see
the user's presence, the user's server return's the user's current
presence state to the contact (this is equivalent to the "Initial
NOTIFY" in SIP presence).
Saint-Andre Expires April 4, 2008 [Page 5]
Internet-Draft XMPP Presence Analysis October 2007
User's server sends presence to contact (170 bytes):
away
be right back
0
If the user subsequently changes her presence, the user's server
sends an updated presence notification to the contact.
User's server sends updated presence to contact (160 bytes):
xa
bbiab
0
A presence session can include any number of presence changes.
When the user goes offline, the user's server sends a presence stanza
of type "unavailable" to the contact.
User's server sends unavailable presence to contact (96 bytes):
Naturally, similar protocol flows are generated by the contact during
the life of his presence session.
4. Analysis
To enable valid comparison between SIMPLE and XMPP with regard to
interdomain presence scaling, this document adheres as closely as
possible to the analysis presented in [PROBLEM], with appropriate
modifications given differences between the two technologies. In
particular, traffic calculations are based on the following inputs
Saint-Andre Expires April 4, 2008 [Page 6]
Internet-Draft XMPP Presence Analysis October 2007
and formulae, where the numbering follows that in [PROBLEM] and the
terminology is adjusted to conform to XMPP.
4.1. Constants
o (C01) Presence session lifetime in hours -- assumed to be 8 hours.
o (C02) Presence state changes per hour -- assumed to be 3 times per
hour.
o (C03) Subscription refresh interval per hour -- does not apply
since XMPP presence subscriptions are long-lived.
o (C04) Total federated contacts per user -- varies based on the
scenario under discussion.
o (C05) Number of dialogs to maintain per watcher -- does not apply
to XMPP since XMPP presence subscriptions are long-lived, for
purposes of calculation treated as equal to C04.
o (C06) Number of federated users -- varies based on the scenario
under discussion.
o (C07) Subscription request size in bytes -- 100 bytes.
o (C08) Subscription approval size in bytes -- 100 bytes.
o (C09) Presence notification size in bytes -- 200 bytes.
o (C10) Presence notification ack size in bytes -- does not apply
since XMPP presence notifications are not acked.
o (C11) Presence document size in bytes -- does not apply since XMPP
presence notifications do not include presence documents.
4.2. Initial Messages
o (I01) Number of initial subscribe requests per presence session --
the XMPP equivalent is a presence probe, of which there is 1 per
contact ( = C04 ), equal in size to a presence subscription
request or approval (100 bytes).
o (I02) Number of initial subscription approvals per presence
session -- does not apply since XMPP presence probes are not
acked.
o (I03) Number of initial presence notifications -- 1 per contact (
= C04 ).
o (I04) Number of initial presence acks -- does not apply since XMPP
presence notifications are not acked.
o (I05) Total number and bytes of initial subscribe messages --
Number = (I01*C06) and Bytes = (I01*C06*C07).
o (I06) Total number and bytes of initial subscribe acks -- Number =
(I02*C06) and Bytes = (I02*C06*C07), both equal to zero since I02
equals zero for XMPP.
o (I07) Total number and bytes of initial notifications -- Number =
(I03*C06) and Bytes = (I03*C06*C09), note that this formula does
not take into account optimizations of the kind discussed in
[PROBLEM].
Saint-Andre Expires April 4, 2008 [Page 7]
Internet-Draft XMPP Presence Analysis October 2007
o (I08) Total number and bytes of initial notification acks --
Number = (I04*C10) and Bytes = (I04*C06*C10), both equal to zero
since I02 equals zero for XMPP.
o (I09) Total number and bytes of initial messages per presence
session -- Number = (numbers in I05+I06+I07+I08) and Bytes =
(bytes in I05+I06+I07+I08).
4.3. Steady State Messages
o (S01) Presence notifications caused by state changes -- (C02*(C01
- 2)).
o (S02) Notification acks for state change notifications -- zero
since XMPP presence notifications are not acked.
o (S03) Number and size of presence notifications caused by state
changes -- Number = (S01*C06*C04), Bytes = (S01*C06*C04)*(C09+C10+
C11).
o (S04) Subscription refreshes -- zero since XMPP presence
subscriptions are long-lived.
o (S05) Acks for subscription refreshes -- zero since there are no
subscription refreshes in XMPP.
o (S06) Notify messages caused by subscription refreshes -- zero
since there are no subscription refreshes in XMPP.
o (S07) Acks for notify messages caused by subscription refreshes --
zero since there are no subscription refreshes in XMPP.
o (S08) Number and size of messages caused by subscription refreshes
-- zero since there are no subscription refreshes in XMPP.
o (S09) Total number and bytes of steady-state messages per session
-- Number = (numbers in S03+S08), Bytes = (bytes in S03+S08).
4.4. Termination Messages
o (T01) Number of terminating subscribe messages -- zero since XMPP
presence subscriptions are long-lived.
o (T02) Number of acks for terminating subscribe messages -- zero
since there are no terminating subscribe messages in XMPP.
o (T03) Number of terminating notifications -- in XMPP this is
unavailable presence sent when user goes offline = C05.
o (T04) Number of acks for terminating notifications -- zero since
XMPP presence notifications are not acked.
o (T05) Total number and size of terminating subscribe messages --
zero since there are no terminating subscribe messages in XMPP.
o (T06) Total number and size of acks for terminating subscribe
messages -- zero since there are no terminating subscribe messages
in XMPP.
o (T07) Total number and size of terminating notifications -- Number
= (numbers in T03*C06), Bytes = (T03*C06*(C09+C11)).
Saint-Andre Expires April 4, 2008 [Page 8]
Internet-Draft XMPP Presence Analysis October 2007
o (T08) Total number and size of acks for terminating notifications
-- zero since XMPP presence notifications are not acked.
o (T09) Total number and size of terminating messages per session --
Number = (numbers in T05+T06+T07+T08), Bytes = (bytes in T05+T06+
T07+T08).
4.5. Bottom Line
o (B01) Total number and bytes per presence session = (I09+S09+T09).
o (B02) Total number and bytes per second = (B01/(C01*3600)).
5. Scenarios
5.1. Basic
This scenario assumes two domains, each with 20,000 users, where each
user has 4 contacts in the other domain and changes presence 3 times
per hour during an 8-hour presence session. The calculations are as
follows.
CONSTANTS
(C01) Presence session lifetime (hours) ....................... 8
(C02) Presence state changes per hour ......................... 3
(C03) Subscription refresh interval per hour ................ N/A
(C04) Total federated contacts per user ....................... 4
(C05) Number of dialogs to maintain per watcher ............... 4
(C06) Number of federated users .......................... 40,000
(C07) Subscription request size in bytes .................... 100
(C08) Subscription approval size in bytes ................... 100
(C09) Presence notification size in bytes ................... 200
(C10) Presence notification ack size in bytes ............... N/A
(C11) Presence document size in bytes ....................... N/A
INITIAL MESSAGES
(I01) Initial subscribe requests per presence session ......... 4
(I02) Initial subscription approvals per presence session ..... 0
(I03) Number of initial presence notifications ................ 4
(I04) Number of initial presence acks ......................... 0
(I05) Total number and bytes of initial subscribe messages
Number ...................................... 160,000
Bytes ................................... 16,000,000
(I06) Total number and bytes of initial subscribe acks
Number ............................................ 0
Bytes ............................................. 0
(I07) Total number and bytes of initial notifications
Number ...................................... 160,000
Bytes ................................... 32,000,000
Saint-Andre Expires April 4, 2008 [Page 9]
Internet-Draft XMPP Presence Analysis October 2007
(I08) Total number and bytes of initial notification acks
Number ............................................ 0
Bytes ............................................. 0
(I09) Total number and bytes of initial messages
Number ...................................... 320,000
Bytes ................................... 48,000,000
STEADY STATE MESSAGES
(S01) Presence notifications caused by state changes ......... 18
(S02) Notification acks for state change notifications ........ 0
(S03) Number and size of steady-state presence notifications
Number .................................... 2,880,000
Bytes ................................... 576,000,000
(S04) Subscription refreshes .................................. 0
(S05) Acks for subscription refreshes ......................... 0
(S06) Notify messages caused by refreshes ..................... 0
(S07) Acks for notify messages caused by refreshes ............ 0
(S08) Number and size of messages caused by refreshes ......... 0
(S09) Total number and bytes of steady-state messages
Number .................................... 2,880,000
Bytes ................................... 576,000,000
TERMINATION MESSAGES
(T01) Number of terminating subscribe messages ................ 0
(T02) Number of acks for terminating subscribe messages ....... 0
(T03) Number of terminating notifications ..................... 4
(T04) Number of acks for terminating notifications ............ 0
(T05) Total number and size of terminates
Number ............................................ 0
Bytes ............................................. 0
(T06) Total number and size of acks for terminates
Number ............................................ 0
Bytes ............................................. 0
(T07) Total number and size of terminating notifications
Number ...................................... 160,000
Bytes ................................... 16,000,000
(T08) Total number and size of acks for terminating notifications
Number ............................................ 0
Bytes ............................................. 0
(T09) Total number and size of terminating messages per session
Number ...................................... 160,000
Bytes ................................... 16,000,000
BOTTOM LINE
(B01) Total number and bytes per presence session
Number .................................... 3,360,000
Bytes .................................. 640,000,000
(B02) Total number and bytes per second
Saint-Andre Expires April 4, 2008 [Page 10]
Internet-Draft XMPP Presence Analysis October 2007
Number .......................................... 116
Bytes ....................................... 22,222
For the bottom-line figures, the comparable numbers for SIMPLE in a
non-optimized state (see [PROBLEM]) are 12,800,000 total messages,
49,756,800,000 bytes, 444 messages per second, and 1,727,667 bytes
per second; thus for this scenario XMPP uses 26% of the messages and
1.3% of the bytes used by SIMPLE.
5.2. Widely Distributed Inter-Domain Presence
This scenario assumes two domains, each with 20,000 users, where each
user has 20 contacts in the other domain and changes presence 3 times
per hour during an 8-hour presence session. The calculations are as
follows.
CONSTANTS
(C01) Presence session lifetime (hours) ....................... 8
(C02) Presence state changes per hour ......................... 3
(C03) Subscription refresh interval per hour ................ N/A
(C04) Total federated contacts per user ...................... 20
(C05) Number of dialogs to maintain per watcher .............. 20
(C06) Number of federated users .......................... 40,000
(C07) Subscription request size in bytes .................... 100
(C08) Subscription approval size in bytes ................... 100
(C09) Presence notification size in bytes ................... 200
(C09) Presence notification ack size in bytes ............... N/A
(C11) Presence document size in bytes ....................... N/A
INITIAL MESSAGES
(I01) Initial subscribe requests per presence session ........ 20
(I02) Initial subscription approvals per presence session ..... 0
(I03) Number of initial presence notifications ............... 20
(I04) Number of initial presence acks ......................... 0
(I05) Total number and bytes of initial subscribe messages
Number ...................................... 800,000
Bytes ................................... 80,000,000
(I06) Total number and bytes of initial subscribe acks
Number ............................................ 0
Bytes ............................................. 0
(I07) Total number and bytes of initial notifications
Number ...................................... 800,000
Bytes .................................. 160,000,000
(I08) Total number and bytes of initial notification acks
Number ............................................ 0
Bytes ............................................. 0
(I09) Total number and bytes of initial messages
Number .................................... 1,600,000
Saint-Andre Expires April 4, 2008 [Page 11]
Internet-Draft XMPP Presence Analysis October 2007
Bytes .................................. 240,000,000
STEADY STATE MESSAGES
(S01) Presence notifications caused by state changes ......... 18
(S02) Notification acks for state change notifications ........ 0
(S03) Number and size of steady-state presence notifications
Number ................................... 14,400,000
Bytes ................................. 2,880,000,000
(S04) Subscription refreshes .................................. 0
(S05) Acks for subscription refreshes ......................... 0
(S06) Notify messages caused by refreshes ..................... 0
(S07) Acks for notify messages caused by refreshes ............ 0
(S08) Number and size of messages caused by refreshes ......... 0
(S09) Total number and bytes of steady-state messages
Number ................................... 14,400,000
Bytes ................................. 2,880,000,000
TERMINATION MESSAGES
(T01) Number of terminating subscribe messages ................ 0
(T02) Number of acks for terminating subscribe messages ....... 0
(T03) Number of terminating notifications .................... 20
(T04) Number of acks for terminating notifications ............ 0
(T05) Total number and size of terminates
Number ............................................ 0
Bytes ............................................. 0
(T06) Total number and size of acks for terminates
Number ............................................ 0
Bytes ............................................. 0
(T07) Total number and size of terminating notifications
Number ...................................... 800,000
Bytes .................................. 160,000,000
(T08) Total number and size of acks for terminating notifications
Number ............................................ 0
Bytes ............................................. 0
(T09) Total number and size of terminating messages per session
Number ...................................... 800,000
Bytes .................................. 160,000,000
BOTTOM LINE
(B01) Total number and bytes per presence session
Number ................................... 16,800,000
Bytes ................................ 3,280,000,000
(B02) Total number and bytes per second
Number .......................................... 583
Bytes ...................................... 113,889
For the bottom-line figures, the comparable numbers for SIMPLE in a
non-optimized state (see [PROBLEM]) are 65,000,000 total messages,
Saint-Andre Expires April 4, 2008 [Page 12]
Internet-Draft XMPP Presence Analysis October 2007
248,784,000,000 bytes, 2,222 messages per second, and 8,638,333 bytes
per second; thus for this scenario XMPP uses 26% of the messages and
1.3% of the bytes used by SIMPLE.
5.3. Very Large Network Peering
This scenario assumes two domains, each with 10,000,000 users, where
each user has 10 contacts in the other domain and changes presence 6
times per hour during an 8-hour presence session. The calculations
are as follows.
CONSTANTS
(C01) Presence session lifetime (hours) ....................... 8
(C02) Presence state changes per hour ......................... 3
(C03) Subscription refresh interval per hour ................ N/A
(C04) Total federated contacts per user ...................... 10
(C05) Number of dialogs to maintain per watcher .............. 10
(C06) Number of federated users ...................... 20,000,000
(C07) Subscription request size in bytes .................... 100
(C08) Subscription approval size in bytes ................... 100
(C09) Presence notification size in bytes ................... 200
(C09) Presence notification ack size in bytes ............... N/A
(C11) Presence document size in bytes ....................... N/A
INITIAL MESSAGES
(I01) Initial subscribe requests per presence session ........ 10
(I02) Initial subscription approvals per presence session ..... 0
(I03) Number of initial presence notifications ............... 10
(I04) Number of initial presence acks ......................... 0
(I05) Total number and bytes of initial subscribe messages
Number .................................. 200,000,000
Bytes ............................... 20,000,000,000
(I06) Total number and bytes of initial subscribe acks
Number ............................................ 0
Bytes ............................................. 0
(I07) Total number and bytes of initial notifications
Number .................................. 200,000,000
Bytes ............................... 40,000,000,000
(I08) Total number and bytes of initial notification acks
Number ............................................ 0
Bytes ............................................. 0
(I09) Total number and bytes of initial messages
Number .................................. 400,000,000
Bytes ............................... 60,000,000,000
STEADY STATE MESSAGES
(S01) Presence notifications caused by state changes ......... 18
(S02) Notification acks for state change notifications ........ 0
Saint-Andre Expires April 4, 2008 [Page 13]
Internet-Draft XMPP Presence Analysis October 2007
(S03) Number and size of steady-state presence notifications
Number ................................ 3,600,000,000
Bytes ............................... 720,000,000,000
(S04) Subscription refreshes .................................. 0
(S05) Acks for subscription refreshes ......................... 0
(S06) Notify messages caused by refreshes ..................... 0
(S07) Acks for notify messages caused by refreshes ............ 0
(S08) Number and size of messages caused by refreshes ......... 0
(S09) Total number and bytes of steady-state messages
Number ................................ 3,600,000,000
Bytes ............................... 720,000,000,000
TERMINATION MESSAGES
(T01) Number of terminating subscribe messages ................ 0
(T02) Number of acks for terminating subscribe messages ....... 0
(T03) Number of terminating notifications .................... 10
(T04) Number of acks for terminating notifications ............ 0
(T05) Total number and size of terminates
Number ............................................ 0
Bytes ............................................. 0
(T06) Total number and size of acks for terminates
Number ............................................ 0
Bytes ............................................. 0
(T07) Total number and size of terminating notifications
Number .................................. 200,000,000
Bytes ............................... 20,000,000,000
(T08) Total number and size of acks for terminating notifications
Number ............................................ 0
Bytes ............................................. 0
(T09) Total number and size of terminating messages per session
Number .................................. 200,000,000
Bytes ............................... 20,000,000,000
BOTTOM LINE
(B01) Total number and bytes per presence session
Number ................................ 4,200,000,000
Bytes ............................... 800,000,000,000
(B02) Total number and bytes per second
Number ...................................... 145,833
Bytes ................................... 27,777,777
For the bottom-line figures, the comparable numbers for SIMPLE in a
non-optimized state (see [PROBLEM]) are 25,600,000,000 total
messages, 99,348,000,000,000 bytes, 888,889 messages per second, and
3,449,583,333 bytes per second; thus for this scenario XMPP uses 16%
of the messages and 0.8% of the bytes used by SIMPLE.
Saint-Andre Expires April 4, 2008 [Page 14]
Internet-Draft XMPP Presence Analysis October 2007
5.4. Intra-Domain Peering
This scenario assumes two domains, each with 60,000 users, where each
user has 10 contacts in the other domain and changes presence 3 times
per hour during an 8-hour presence session. The calculations are as
follows.
CONSTANTS
(C01) Presence session lifetime (hours) ....................... 8
(C02) Presence state changes per hour ......................... 3
(C03) Subscription refresh interval per hour ................ N/A
(C04) Total federated contacts per user ...................... 10
(C05) Number of dialogs to maintain per watcher .............. 10
(C06) Number of federated users ......................... 120,000
(C07) Subscription request size in bytes .................... 100
(C08) Subscription approval size in bytes ................... 100
(C09) Presence notification size in bytes ................... 200
(C09) Presence notification ack size in bytes ............... N/A
(C11) Presence document size in bytes ....................... N/A
INITIAL MESSAGES
(I01) Initial subscribe requests per presence session ........ 10
(I02) Initial subscription approvals per presence session ..... 0
(I03) Number of initial presence notifications ............... 10
(I04) Number of initial presence acks ......................... 0
(I05) Total number and bytes of initial subscribe messages
Number .................................... 1,200,000
Bytes .................................. 120,000,000
(I06) Total number and bytes of initial subscribe acks
Number ............................................ 0
Bytes ............................................. 0
(I07) Total number and bytes of initial notifications
Number .................................... 1,200,000
Bytes .................................. 240,000,000
(I08) Total number and bytes of initial notification acks
Number ............................................ 0
Bytes ............................................. 0
(I09) Total number and bytes of initial messages
Number .................................... 2,400,000
Bytes .................................. 360,000,000
STEADY STATE MESSAGES
(S01) Presence notifications caused by state changes ......... 18
(S02) Notification acks for state change notifications ........ 0
(S03) Number and size of steady-state presence notifications
Number ................................... 21,600,000
Bytes ................................. 4,320,000,000
(S04) Subscription refreshes .................................. 0
Saint-Andre Expires April 4, 2008 [Page 15]
Internet-Draft XMPP Presence Analysis October 2007
(S05) Acks for subscription refreshes ......................... 0
(S06) Notify messages caused by refreshes ..................... 0
(S07) Acks for notify messages caused by refreshes ............ 0
(S08) Number and size of messages caused by refreshes ......... 0
(S09) Total number and bytes of steady-state messages
Number ................................... 21,600,000
Bytes ................................. 4,320,000,000
TERMINATION MESSAGES
(T01) Number of terminating subscribe messages ................ 0
(T02) Number of acks for terminating subscribe messages ....... 0
(T03) Number of terminating notifications .................... 10
(T04) Number of acks for terminating notifications ............ 0
(T05) Total number and size of terminates
Number ............................................ 0
Bytes ............................................. 0
(T06) Total number and size of acks for terminates
Number ............................................ 0
Bytes ............................................. 0
(T07) Total number and size of terminating notifications
Number .................................... 1,200,000
Bytes .................................. 240,000,000
(T08) Total number and size of acks for terminating notifications
Number ............................................ 0
Bytes ............................................. 0
(T09) Total number and size of terminating messages per session
Number .................................... 1,200,000
Bytes .................................. 240,000,000
BOTTOM LINE
(B01) Total number and bytes per presence session
Number ................................... 25,200,000
Bytes ................................. 4,920,000,000
(B02) Total number and bytes per second
Number .......................................... 875
Bytes ...................................... 170,833
For the bottom-line figures, the comparable numbers for SIMPLE in a
non-optimized state (see [PROBLEM]) are 96,000,000 total messages,
373,176,000,000 bytes, 3,333 messages per second, and 12,957,500
bytes per second; thus for this scenario XMPP uses 26% of the
messages and 1.3% of the bytes used by SIMPLE.
6. Conclusion
With respect to presence scaling, the differences between XMPP
systems and SIP-based systems are startling. In particular, this
Saint-Andre Expires April 4, 2008 [Page 16]
Internet-Draft XMPP Presence Analysis October 2007
analysis indicates that XMPP requires only about 1.5% of the
bandwidth required by SIMPLE. There are two primary causes for this
disparity: (1) XMPP requires only about 25% of the packets required
by SIMPLE and (2) XMPP packets are about 5% of the size of SIMPLE
packets for presence notifications and 20% of the size for subscribe
packets. Together, these two factors appear to result in a
significant disparity with respect to the scalability of presence
technologies. Naturally, real-world studies of deployed systems will
be necessary to determine if these theorized differences occur in
reality.
7. Security Considerations
This document introduces and addresses no security concerns above and
beyond those already defined in [XMPP-CORE] and [XMPP-IM].
8. Informative References
[PROBLEM] Houri, A., "Presence Interdomain Scaling Analysis for SIP/
SIMPLE", draft-ietf-simple-interdomain-scaling-analysis-01
(work in progress), July 2007.
[SIP] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston,
A., Peterson, J., Sparks, R., Handley, M., and E.
Schooler, "SIP: Session Initiation Protocol", RFC 3261,
June 2002.
[SIP-EVENT]
Roach, A., "Session Initiation Protocol (SIP)-Specific
Event Notification", RFC 3265, June 2002.
[SIP-PRES]
Rosenberg, J., "A Presence Event Package for the Session
Initiation Protocol (SIP)", RFC 3856, August 2004.
[XEP-0060]
Millard, P., Saint-Andre, P., and R. Meijer, "Publish-
Subscribe", XSF XEP 0060, September 2007.
[XEP-0115]
Hildebrand, J., Saint-Andre, P., and R. Troncon, "Entity
Capabilities", XSF XEP 0115, August 2007.
[XEP-0118]
Saint-Andre, P., "User Tune", XSF XEP 0118, June 2007.
Saint-Andre Expires April 4, 2008 [Page 17]
Internet-Draft XMPP Presence Analysis October 2007
[XEP-0138]
Hildebrand, J. and P. Saint-Andre, "Stream Compression",
XSF XEP 0138, September 2007.
[XEP-0153]
Saint-Andre, P., "vCard-Based Avatars", XSF XEP 0153,
August 2006.
[XEP-0163]
Saint-Andre, P. and K. Smith, "Personal Eventing via
Pubsub", XSF XEP 0163, September 2007.
[XMPP-CORE]
Saint-Andre, P., "Extensible Messaging and Presence
Protocol (XMPP): Core", RFC 3920, October 2004.
[XMPP-IM] Saint-Andre, P., "Extensible Messaging and Presence
Protocol (XMPP): Instant Messaging and Presence",
RFC 3921, October 2004.
Author's Address
Peter Saint-Andre
XMPP Standards Foundation
P.O. Box 1641
Denver, CO 80201
USA
Email: stpeter@jabber.org
URI: https://stpeter.im/
Saint-Andre Expires April 4, 2008 [Page 18]
Internet-Draft XMPP Presence Analysis October 2007
Full Copyright Statement
Copyright (C) The IETF Trust (2007).
This document is subject to the rights, licenses and restrictions
contained in BCP 78, and except as set forth therein, the authors
retain all their rights.
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Intellectual Property
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org.
Acknowledgment
Funding for the RFC Editor function is provided by the IETF
Administrative Support Activity (IASA).
Saint-Andre Expires April 4, 2008 [Page 19]