< draft-lee-network-stratum-query-problem-01.txt		draft-lee-network-stratum-query-problem-02.txt >
	Network Working Group Young Lee (Huawei)		Network Working Group Young Lee (Huawei)
	Internet Draft Dave McDysan (Verizon)		Internet Draft Dave McDysan (Verizon)
	Intended Status: Informational Ning So (UTD)		Intended Status: Informational Ning So (UTD)
	Greg Bernstein (Grotto)		Greg Bernstein (Grotto)
	Tae Yeon Kim (ETRI)		Tae Yeon Kim (ETRI)
	Kohei Shiomoto (NTT)		Kohei Shiomoto (NTT)
	Oscar Gonzalez de Dios (Telefonica)		Oscar Gonzalez de Dios (Telefonica)
	October 20, 2010		April 20, 2011
	Problem Statement for Network Stratum Query		Problem Statement for Network Stratum Query
	draft-lee-network-stratum-query-problem-01.txt		draft-lee-network-stratum-query-problem-02.txt
	Status of this Memo		Status of this Memo
	This Internet-Draft is submitted to IETF in full conformance with the		This Internet-Draft is submitted to IETF in full conformance with the
	provisions of BCP 78 and BCP 79.		provisions of BCP 78 and BCP 79.
	Internet-Drafts are working documents of the Internet Engineering		Internet-Drafts are working documents of the Internet Engineering
	Task Force (IETF), its areas, and its working groups. Note that		Task Force (IETF), its areas, and its working groups. Note that
	other groups may also distribute working documents as Internet-		other groups may also distribute working documents as Internet-
	Drafts.		Drafts.
	skipping to change at page 2, line 14 ¶		skipping to change at page 2, line 14 ¶
	publication of this document. Please review these documents		publication of this document. Please review these documents
	carefully, as they describe your rights and restrictions with respect		carefully, as they describe your rights and restrictions with respect
	to this document.		to this document.
	Abstract		Abstract
	This document describes the general problem of network stratum query		This document describes the general problem of network stratum query
	for application optimization. Network Stratum query is an ability to		for application optimization. Network Stratum query is an ability to
	query the network from an application controller such as those used		query the network from an application controller such as those used
	in Data Centers so that application controller decisions such as		in Data Centers so that application controller decisions such as
	server assignment or virtual machine instantiation/migration can be		server assignment or virtual machine instantiation/migration could be
	performed with better knowledge of the underlying network conditions.		performed with better knowledge of the underlying network conditions.
	As application servers are distributed geographically across Data		As application servers are distributed geographically across Data
	Centers, many application-related decisions such as which server to		Centers, many application-related decisions such as which server to
	assign a new client or where to instantiate/migrate virtual machines		assign a new client or where to instantiate/migrate virtual machines
	will suffer from sub-optimality unless the underlying network		will suffer from sub-optimality unless the underlying network
	conditions are factored in the decision process. The lack of network		conditions are factored in the decision process. The lack of network
	awareness may result in not meeting the end-user service objective		awareness may result in not meeting the end-user service objective
	for some key applications like video gaming/conferencing that require		for some key applications like video gaming/conferencing that require
	stringent latency and bandwidth requirement.		stringent latency and bandwidth requirement.
	Table of Contents		Table of Contents
	1. Introduction...................................................3		1. Introduction......................................... 2
	2. Network and Application Contexts...............................4		2. Network Contexts..................................... 4
	3. Problem Statement..............................................7		3. Problem Statement .................................... 7
	3.1. Limitation of existing probing schemes....................7		4. High-level requirements................................ 9
	3.2. Lack of vertical query schemes............................8		4.1. Data Center-Network Stratum Communication (NS Query) Error!
	3.3. Limitation of SNMP MIB network monitoring techniques......8		Bookmark not defined.
	3.4. Lack of abstraction mechanisms............................8		4.1.1. Application Profile........................... 9
	4. High-level requirements........................................9		4.1.2. Network Load Data to be queried................ 10
	4.1. Application Profile.......................................9		4.1.3. Responses to NS Query from network to application. 10
	4.2. Network Load Data to be queried..........................10		5. Security Considerations............................... 11
	4.3. A Whole Network Query capability.........................10		6. References......................................... 11
	4.4. Data Synchronization Mechanism...........................10		Author's Addresses..................................... 13
	4.5. Responses to NS Query from network to application........11		Intellectual Property Statement .......................... 13
	5. Security Considerations.......................................11		Disclaimer of Validity.................................. 14
	6. IANA Considerations...........................................11
	7. References....................................................12
	7.1. Informative References...................................12
	Author's Addresses...............................................13
	Intellectual Property Statement..................................13
	Disclaimer of Validity...........................................14
	1. Introduction		1. Introduction
	Cross Stratum Optimization is a joint optimization effort in		Cross Stratum Optimization is a joint optimization effort in
	allocating resources to end-users that involves both the Application		allocating resources to end-users that involves both the Application
	Stratum and Network Stratum.		Stratum and Network Stratum.
	The application stratum is the functional block which manages and		The application stratum is the functional block which manages and
	controls application resources and provides application resources to		controls application resources and provides application resources to
	a variety of clients/end-users. Application resources are non-network		a variety of clients/end-users. Application resources are non-network
	resources critical to achieving the application service		resources critical to achieving the application service
	functionality. Examples include: application specific servers,		functionality. Examples include: application specific servers,
	storage, content, large data sets, and computing power. Data Centers		storage, content, large data sets, and computing power. Data Centers
	are regarded as a tangible realization of the application stratum		are regarded as tangible realization of the application stratum
	architecture.		architecture.
	The network stratum is the functional block which manages and		The network stratum is the functional block which manages and
	controls network resources and provides transport of data between		controls network resources and provides transport of data between
	clients/end-users to and among application resources. Network		clients/end-users to and among application resources. Network
	Resources are resources of layer 3 or below (L1/L2/L3) such as		Resources are resources of any layer 3 or below (L1/L2/L3) such as
	bandwidth, links, paths, path processing (creation, deletion, and		bandwidth, links, paths, path processing (creation, deletion, and
	management), network databases, path computation, admission control,		management), network databases, path computation, admission control,
	and resource reservation capability.		and resource reservation capability.
	Application services by their very nature utilize application		Application services offered by Data Centers by their very nature
	resources (e.g., servers, storage, memory, etc...) in Data Centers,		utilize application resources (e.g., servers, storage, memory,
	and the underlying network resources provided by LANs, MANs, and		etc...) in Data Centers, and the underlying network resources
	carrier's transport networks. By "data center" is any location in the		provided by LANs, MANs, and carrier's transport networks.
	network where applications resources, such as servers or storage, are
	aggregated. We include in this definition extremely large data
	centers with 10,000 or more servers, all the way down to small points
	of presence co-located in network carrier facilities.
	As the application servers are distributed geographically across many		As the application servers are distributed geographically across many
	Data Centers, decisions such as server assignment or new virtual		Data Centers, decisions such as server assignment or new virtual
	machine instantiation/migration will suffer from sub-optimality		machine instantiation/migration will suffer from sub-optimality
	unless the underlying network conditions are factored in the decision		unless the underlying network conditions are factored in the decision
	process. The lack of network awareness may result in not meeting the		process. The lack of network awareness may result in not meeting the
	end-user service objective for some key applications like video		end-user service objective for some key applications like video
	gaming/conferencing that require stringent latency and bandwidth		gaming/conferencing that require stringent latency and bandwidth
	requirement.		requirement.
	This document describes the general problem of network stratum query		This document describes the general problem of network stratum query
	(NS Query) in Data Center environments. Network Stratum query is an		(NS Query) in Data Center environments. Network Stratum query is an
	ability to query the network from application controllers in Data		ability to query the network from application controller in Data
	Centers so that application server assignment or virtual machine		Centers so that application server assignment or virtual machine
	instantiation/migration decision would be jointly performed based on		instantiation/migration decision would be jointly performed based on
	both the application resource/load status and the network		both the application resource/load status and the network
	resource/load status.		resource/load status.
	The NS query is different from what is called "horizontal network		The NS query is different from typical "horizontal" query
	queries" performed as part of network management. These horizontal		capabilities in the network. The horizontal query in the network is
	queries are carried out by an entity (network management systems)		carried by the head end (i.e., data source) that would "probe" the
	within the network and would have fairly complete access to network		network to test the capabilities for data flows to/from particular
	information.		point in the network. This is a horizontal scheme.
	NS Query can be thought of as a two-stage query that consists of:		NS Query is a two-stage query that consists of two stages:
	. First Stage: A vertical query from an application entity (i.e.,		. A vertical query capability where an external point (i.e., the
	the Application Control Gateway (ACG) in Data Center) to an		Application Control Gateway (ACG) in Data Center) will query
	entity representing the network (i.e., the Network Control		the network (i.e., the Network Control Gateway (NCG)); and
	Gateway (NCG))for highly summarized and abstracted network
	related information; and
	. Second Stage: Internal "horizontal queries" at the network work		. A horizontal query capability where the NCG to gather the
	layer along with summarization and abstraction of the network		collective information of a variety of horizontal schemes
	information in a form that preserves network confidentiality		(IPPM, IGP, RIB, etc.) implemented in the network stratum.
	and significantly reduces the amount of information that needs
	to be transferred. The raw information needed to perform this
	summarization/abstraction is defined in existing and emerging
	network management standards and protocols (SNMP, Netflow,
	sFlow, IPPM, IGP, RIB, etc...).
	NS Query would not necessarily standardize how the above "internal		NS Query does not re-invent the wheel on existing network
	horizontal queries" and summarization would be performed but would		capabilities but tries to reuse them where possible.
	illustrate how such processes can be accomplished via standards,
	emerging standards or common commercial practices.
	2. Network and Application Contexts		2. Network Contexts
	Figure 1 shows a typical application architecture where an end-user		Figure 1 shows a typical data center architecture where an end-user
	(the point of consuming resource) needs to be connected for its		(the point of consuming resource) needs to be connected for its
	application (e.g., gaming) to a server located in one of the		application (e.g., gaming) to a server located in one of the data
	geographically distributed data centers.		centers geographically spread.
			---------------
			---------- | DC 1 |
			| End-user |. . . . .>| o o o |
			| | | \|/ |
			---------- | O |
			| ----- --|------
			| |
			| |
			| -----------------|-----------
			| / | \
			| / ..........O PE1 \ --------------
			| | . | | o o o DC 2 |
			| | PE4 . PE2 | | \|/ |
			----|---O.........................O---|---|---O |
			| . | | |
			| . PE3 | --------------
			\ ..........O Carrier /
			\ | Network /
			---------------|-------------
			|
			--------|------
			| O |
			| /|\ |
			| o o o |
			| DC 3 |
			---------------
	,-----. ---------------
	---------- / App \ | DC 1 |
	| End-user |. . .>( Control ) | o o o |
	| | \ / | \|/ |
	---------- `-----' | O |
	| ----- --|------
	| |
	| |
	| --------------------------|--
	| / PE1 | \
	| / ...................O \ --------------
	| | . | | o o o DC 2 |
	| | PE4 . PE2 | | \|/ |
	----|---O.........................O---|---|---O |
	| . | | |
	| . PE3 | --------------
	\ ..........O Carrier /
	\ | Network /
	---------------|-------------
	|
	--------|------
	| O |
	| /|\ |
	| o o o |
	| DC 3 |
	---------------
	Figure 1. Data Center Architecture		Figure 1. Data Center Architecture
	Figure 1 shows that the user application can be served by any of the		Figure 1 shows that the user application can be served by any of the
	servers in DC1, DC2 or DC3. When the initial request arrives to the		servers in DC1, DC2 or DC3. When the initial request arrives to the
	application controller, the controller (aka, a global load balancer)		proxy server in DC1, the proxy server (aka, the load balancer) would
	would ideally assign an "optimal" server based on both server		ideally assign an "optimal" server based on both server resource/load
	resource/load status and the network resources/load status. This		status and the network resources/load status. This server assignment
	server assignment decision today, however, is limited due to the lack		decision today, however, is limited due to the lack of network
	of network awareness in this decision making process in the		awareness in this decision making process in the application.
	application.
	For example, the application controller needs to find a good server		For example, the server close to the user in Data Center 1 may find a
	that can serve the client. Assume that this particular application		good server that can serve the application. Assume that this
	requires x amount of minimum bandwidth guarantee and with less than y		particular application requires x amount of minimum bandwidth
	ms of latency limit. The route that serves Data Center 1 traffic to		guarantee and with less than y ms of latency limit. The route that
	the end-user (PE1 - PE4) may not have enough capacity at a moment of		serves Data Center 1 traffic to the end-user (PE1 - PE4) may not have
	service instantiation and therefore the service objective of the end-		enough capacity at a moment of service instantiation and therefore
	user may not be satisfied had such route been taken.		the service objective of the end-user may not be satisfied had such
			route been taken.
	On the other hand, there may be good servers available in Data		On the other hand, there may be good servers available in Data
	Centers 2 and 3 and their routes (PE2-PE4 and PE3-PE4) may have		Centers 2 and 3 and their routes (PE2-PE4 and PE3-PE4) may have
	enough capacity to meet the service requirement.		enough capacity to meet the service requirement.
	This example illustrates the benefit of and the need for the joint		This example illustrates the benefit of and the need for the joint
	optimization across the application and network strata. NS Query is		optimization across the application and network strata. NS Query is
	the ability to query the network from an application controller to		the ability to query the network from an application to collect a
	collect a certain level of network information. No such mechanisms		certain level of network information. No such mechanisms exist in the
	exist in the today's Internet Protocol technologies.		today's Internet Protocol technologies.
	Figure 2 shows the context of NS Query in a more detail within the		Figure 2 shows the context of NS Query in a more detail within the
	overarching data center architecture shown in Figure 1.		overarching data center architecture shown in Figure 1.
	+--------------------------------------------+		--------------------------------------------
	| +-------------+ |		| Application Overlay |
	| | Application | |		| (Data Centers) |
	| | Controller | Application Overlay |		| |
	| +------|------+ (Data Centers) |		---------- | -------------- -------------- |
	| | |
	---------- | ------|------- -------------- |
	| End-User | | | Application |. . . .| Application | |		| End-User | | | Application |. . . .| Application | |
	| |. . . >| | Control | | Processes | |		| |. . . >| | Control | | Processes | |
	---------- | | Gateway (ACG)| -------------- |		---------- | | Gateway (ACG)| -------------- |
	| | | -------------- |		| | | -------------- |
	| ------------- . . . . | Application | |		| ------------- . . . . | Application | |
	| /\ | Related Data | |		| /\ | Related Data | |
	| || -------------- |		| || -------------- |
	----------||--------------------------------		----------||--------------------------------
	||		||
	|| Network Stratum Query (First Stage)		|| Network Stratum Query (First Stage)
	||		||
	+----------||--------------------------------+		----------||--------------------------------
	| \/ Network Underlay |		| \/ Network Underlay |
	| |		| |
	| -------------- ---------------- |		| -------------- ---------------- |
	| | Network |. . . | Network | |		| | Network |. . . | Network | |
	| | Control | | Processes | |		| | Control | | Processes | |
	| | Gateway (NCG)| ---------------- |		| | Gateway (NCG)| ----------------
	| | | ---------------- |		| | | ---------------- |
	| ------------- | Network | |		| ------------- | Network | |
	| |------------->| Related Data | |		| |------------->| Related Data | |
	| (Second Stage) ---------------- |		| (Second Stage) ---------------- |
	+--------------------------------------------+		-------------------------------------------
	Figure 2. NS Query Architecture		Figure 2. NS Query Architecture
	Figure 2 shows key architectural components that enable NS Query		Figure 2 shows key architectural components that enable NS Query
	capability. The application controller, e.g., global load balancer,		capability. The Application Control Gateway (ACG) is the proxy
	utilizes the Application Control Gateway (ACG) to interface with the		gateway that interfaces with network and generate queries to network.
	network and generate queries to network. The ACG can query various		The ACG can query various metric values that may contribute to
	metric values that may contribute to meeting the overall service		meeting the overall service objective of an application. This is a
	objective of an application. This is a vertical query (Stage 1).		vertical query (Stage 1).
	In the network stratum, the Network Control Gateway (NCG) serves the		In the network stratum, the Network Control Gateway (NCG) serves as
	interface to the network stratum. The NCG receives the query request		the proxy gateway to the network. The NCG receives the query request
	from the ACG, makes use of current and/or historical network		from the ACG, probes the network to test the capabilities for data
	information (IPPM, IGP, MIB, TED, etc...), abstracts and summaries		flow to/from particular point in the network, and gather the
	this information implemented in the network stratum. This is the		collective information of a variety of horizontal schemes (IPPM,
	horizontal query and summarization stage (Stage 2).		IGP, MIB, TED, etc.) implemented in the network stratum. This is a
			horizontal query (Stage 2).
	Further, the NCG provides the responses to the original query sent		Further, the NCG provides the responses to the original query sent
	from the ACG. The data collected by the NCG needs to be abstracted.		from the ACG. The data collected by the NCG needs to be abstracted.
	This abstraction is needed on two grounds.		This abstraction is needed on two grounds.
	First, the network does not usually reveal its details to the		First, the network does not usually reveal its details to the
	outside entity. Although the Data Center providers and the carriers		outside entity. Although the Data Center providers and the carriers
	are business partners in providing application services to the end-		are business partners in providing application services to the end-
	users and to the application providers (e.g., gaming providers),		users and to the application providers (e.g., gaming providers),
	detail network data may not be leaked to the Data Centers, and vice		detail network data may not be leaked to the Data Centers, and vice
	skipping to change at page 7, line 36 ¶		skipping to change at page 7, line 25 ¶
	Secondly, detail network data may not be understood by the		Secondly, detail network data may not be understood by the
	application. Link or node level data in and of themselves may not		application. Link or node level data in and of themselves may not
	help the application to process the detail data. For instance,		help the application to process the detail data. For instance,
	latency or bandwidth on a link level is too detail for application		latency or bandwidth on a link level is too detail for application
	to handle. Instead, latency or bandwidth on a route level (i.e., PE1		to handle. Instead, latency or bandwidth on a route level (i.e., PE1
	- PE4 in Figure 1) will help the application make its server		- PE4 in Figure 1) will help the application make its server
	selection/instantiation decision.		selection/instantiation decision.
	The abstraction function needs to be provided by the NCG. Note that		The abstraction function needs to be provided by the NCG. Note that
	NCG plays gate keeper role to information concerning the network. a		NCG plays a head end role within the network probing/collecting
	Within the network the NCG collects and or probes network		network performance/management data (e.g., IPPM, MIB, etc.) or
	performance/management data (e.g., IPPM, MIB, etc.) or routing data		routing data [MRT] (e.g., LSDB, TED, BGP-RIB, etc.) and others. Once
	[MRT] (e.g., LSDB, TED, BGP-RIB, etc.) and others. Once the basic		the basic data is collected, the NCG will need to abstract/summary
	data is collected, the NCG will need to abstract/summary before it		before it sends to the application.
	sends to the application.
	3. Problem Statement		3. Problem Statement
	3.1. Limitation of existing probing schemes		3.1. Limitation of existing probing schemes
	The current state-of-the art application network awareness schemes		The current state-of-the art probing schemes from an external point
	for an entity external to the network are based on ping, trace route,		are based on ping or trace route like mechanisms based on the
	or vendor specific probing mechanisms based on the assumption that		assumption that the underlying transport network is L3 network and
	the underlying transport network is L3 network and that the routing		that the routing is simple IP forwarding.
	is simple IP forwarding.
	In reality, the carrier's routing schemes are likely to include IP		In reality, the carrier's routing schemes are likely to include IP
	tunneling or MPLS tunneling on top of or in place of IP forwarding.		tunneling or MPLS tunneling on top of or in place of IP forwarding.
	In some cases, the actual network may be VPN, MPLS-TE or GMPLS-TE		In some cases, the actual network may be VPN, MPLS-TE or GMPLS-TE
	networks where trace route does not work.		networks where trace route does not work.
	This implies that network status estimation technique made from		This implies that network status estimation technique made from
	application stratum may have limited accuracy. Thus, application		application stratum cannot be accurate. Thus, application resource
	resource allocation to end-users can suffer sub-optimality and fail		allocation to end-users can suffer sub-optimality and fail to meet
	to meet performance objective for the application.		performance objective for the application.
	3.2. Lack of vertical query schemes		3.2. Lack of vertical query schemes
			Currently, the query in the network is carried by the head end (i.e.,
			data source) that would "probe" the network to test the capabilities
			for data flows to/from particular point in the network. This is a
			horizontal scheme.
	There is no standard "vertical" query scheme that allows an		There is no standard "vertical" query scheme that allows an
	application control gateway in a Data Center to query network stratum		application control gateway in Data Center to query network stratum
	in a way suitable for a third party (i.e. an entity "outside" the		in a way suitable for a third party (i.e. an entity "outside" the
	network).		network).
	Due to the lack of standard vertical query scheme, there is a		Due to the lack of standard vertical query scheme, there is a
	limitation on exchanging information between application and network		limitation on exchanging information between application and network
	that would increase efficiency of joint optimization across		that would increase efficiency of joint optimization across
	application to network. For instance, the ability to exchange the		application to network. For instance, the ability to exchange the
	application profile information (defined in Section 4.1) or network		application profile information (defined in Section 4.1) or network
	capability information between application and network would increase		capability information between application and network would increase
	efficiency of resource allocation across application to network.		efficiency of resource allocation across application to network.
	3.3. Limitation of SNMP MIB network monitoring techniques		3.3. Limitation of SNMP MIB network monitoring techniques
	SNMP MIB Network Monitoring lacks a whole network query capability. A		SNMP MIB monitoring techniques as defined in [RFC2261] and [RFC2265]
	whole network query is a query to gather information across many		do not provide mechanisms to guarantee synchronization of the data
	boxes simultaneously under the control of a single administration		collection. This higher level of synchronization is necessary to
	domain (AD) as defined in RFC 1136. A single AD means the single AS		service: a) application with stringent QoS and Bandwidth, or to b)
	or multiple ASes under the control of a single AD.		better schedule massive quantities of small data flows.
			In addition, SNMP MIB Network Monitoring lacks a whole network query
			capability. A whole network query is a query to gather information
			across many boxes simultaneously under the control of a single
			administration domain (AD) as defined in RFC 1136. A single AD means
			the single AS or multiple ASes under the control of a single AD.
	3.4. Lack of abstraction mechanisms		3.4. Lack of abstraction mechanisms
	Most of the information needed to provide NS Query is currently		Most of the information needed to provide NS Query is currently
	available from the network; however, it is not aggregated into a form		available from the network; however, it is not aggregated into a form
	suitable for use by the application stratum. For example from		suitable for use by the application stratum. For example from
	commonly monitored SNMP based link statistics and current routing		commonly monitored SNMP based link statistics and current routing
	tables one can easily compute average available bandwidth and many		tables one can easily compute average available bandwidth and many
	other statistical performance measures such as packet loss, latency,		other statistical performance measures such as packet loss, latency,
	etc.		etc.
	However, neither the raw SNMP nor routing table data should be		However, neither the raw SNMP nor routing table data should be
	delivered to the application stratum since (a) this reveals too much		delivered to the application stratum since (a) this reveals too much
	information concerning the carriers network, (b) presents too much		information concerning the carriers network, (b) presents too much
	information to transfer to each application. This warrants some work		information to transfer to each application. This warrants some works
	on abstraction from network side to preserve the privacy of network		on abstraction from network side to preserve the privacy of network
	stratum details from the application stratum.		stratum details from the application stratum.
	4. High-level requirements		4. High-level requirements
	This section discusses high-level requirements to support NS Query in		This section discusses high-level requirements to support NS Query in
	the Data Center environments.		the Data Center environments.
	The ACG plays the key role functioning as an application gateway to		The ACG plays the key role functioning as an application gateway to
	network and runs the NS Query. The ACG has access to the end-user		network and runs the NS Query. The ACG has access to the end-user
	profile for the application and the candidate servers' locations		profile for the application and the candidate servers' locations
	locally and remotely located. How the ACG access these information is		locally and remotely located. How the ACG access these information is
	beyond the scope of this work.		beyond the scope of this work.
	4.1. Application Profile		4.1. Application Profile
	The application Stratum needs to provide the application profile to		The application Stratum needs to provide the application profile to
	network as part of a query.		network.
	Example service profile information that can be useful to network to		Example service profile information that can be useful to network to
	understand is as follows:		understand is as follows:
	. End user IP address;		. End user IP address;
	. User access router IP address;		. User access router IP address;
	. Authentication Profile: Authentication Key;		. Authentication Profile: Authentication Key;
	skipping to change at page 10, line 5 ¶		skipping to change at page 10, line 5 ¶
	. Connectivity Profile: P-P, P-MP, Anycast (Multi-destination);		. Connectivity Profile: P-P, P-MP, Anycast (Multi-destination);
	. Directionality of the connectivity: unidirectional, bi-		. Directionality of the connectivity: unidirectional, bi-
	directional;		directional;
	. Path Estimation Objective Function: Min latency, etc.		. Path Estimation Objective Function: Min latency, etc.
	Additional profile information can be added depending on the network		Additional profile information can be added depending on the network
	capability.		capability.
	4.2. Network Load Data to be queried		4.2. Network Load Data to be queried (First Satge)
	For a given location mapping information (i.e., from the server		For a given location mapping information (i.e., from the server
	location to end-user location), the query from an application can ask		location to end-user location), the query from an application can ask
	the following network load data:		the following network load data:
	. Type of networks and the technical capabilities of the networks;		. Type of networks and the technical capabilities of the networks;
	. Bandwidth capabilities and availability;		. Bandwidth capabilities and availability;
	. latency;		. latency;
	. jitter;		. jitter;
	. packet loss;		. packet loss;
	skipping to change at page 10, line 27 ¶		skipping to change at page 10, line 27 ¶
	section 5 of [ITU-T Y.1541].		section 5 of [ITU-T Y.1541].
	Note that this can be asked in a different way. For example, the		Note that this can be asked in a different way. For example, the
	query can simply ask:		query can simply ask:
	. Can you give me a route with x amount of b/w (from server to		. Can you give me a route with x amount of b/w (from server to
	end-user) within y ms of latency?		end-user) within y ms of latency?
	. Can you give me a route with x amount of b/w (from server to		. Can you give me a route with x amount of b/w (from server to
	end-user) with no packet loss?		end-user) with no packet loss?
	4.3. A Whole Network Query capability		4.3. A Whole Network Query capability (Second Stage)
	Upon the request from application (specifically, the ACG in Figure		Upon the request from application (specifically, the ACG in Figure
	2), the network (specifically the NCG in Figure 2) should perform "a		2), the network (specifically the NCG in Figure 2) should perform "a
	whole network query" of information.		whole network query" of information.
	A whole network query is a query to gather information across many		A whole network query is a query to gather information across many
	boxes simultaneously under the control of a single administration		boxes simultaneously under the control of a single administration
	domain (AD) as defined in RFC 1136. A single AD means the single AS		domain (AD) as defined in RFC 1136. A single AD means the single AS
	or multiple ASes under the control of a single AD.		or multiple ASes under the control of a single AD.
	The scope of a whole network query can include the topology of the		The scope of a whole network query can include the topology of the
	network, the bandwidth availability for the routes of interest, the		network, the bandwidth availability for the routes of interest, the
	capabilities and congestion of links and routes, and an indication of		capabilities and congestion of links and routes, and an indication of
	the contribution to delay and jitter that each link and route will		the contribution to delay and jitter that each link and route will
	contribute and so on.		contribute and so on.
	4.4. Data Synchronization Mechanism		4.4. Data Synchronization Mechanism
	When querying the network there are two general categories of		The ability to capture the data at the same instant should be
	information that are of interest (a) current data, and (b) historical		provided.
	data. Current data information tells us information about the current
	state of the network, this data represent current network "state" and
	over time is subject to change. Application servers would use
	abstracted versions of this information to make decisions regarding
	current application activities.
	Historical data can be used by the application controller to schedule
	application events at future times. Such historical data must be time
	stamped so that inferences can be drawn from the data. For example,
	consider the backup and synchronization of large application
	databases between two data centers. Not only would we like to know
	how much bandwidth is available, and importantly when is the best
	time to perform such synchronization (given there is flexibility on
	when to perform the backup).
	4.5. Responses to NS Query from network to application		4.5. Responses to NS Query from network to application
	Given the network query from application, the network should provide		Given the network query from application, the network should provide
	the following mechanisms:		the following mechanisms:
	- For a given location mapping information from application (i.e.,		- For a given location mapping information from application (i.e.,
	from the server location to end-user location) and the gathered		from the server location to end-user location) and the gathered
	information by the second stage query discussed in section 4.3.,		information by the second stage query discussed in section 4.3.,
	the network needs to present the requested information in a		the network needs to present the requested information in a
	skipping to change at page 12, line 16 ¶		skipping to change at page 11, line 39 ¶
	7.1. Informative References		7.1. Informative References
	[RFC2261] D. Harrington, et al., "An Architecture for Describing SNMP		[RFC2261] D. Harrington, et al., "An Architecture for Describing SNMP
	Management Frameworks," January, 1998.		Management Frameworks," January, 1998.
	[RFC2265] B. Wijnen, et al., "View-based Access Control Model (VACM)		[RFC2265] B. Wijnen, et al., "View-based Access Control Model (VACM)
	for the Simple Network Management Protocol (SNMP),"		for the Simple Network Management Protocol (SNMP),"
	January, 1998.		January, 1998.
	[Y.1541] Network performance objectives for IP-based services,
	February, 2002.
	[Y.2011] General principles and general reference model for Next		[Y.2011] General principles and general reference model for Next
	Generation Networks, October, 2004.		Generation Networks, October, 2004.
	[Y.2012] Functional Requirements and architecture of the NGN, April,		[Y.2012] Functional Requirements and architecture of the NGN, April,
	2010.		2010.
	[MRT] L. Blunk, M. Karir, and C. Labovitz, "MRT routing		[MRT] L. Blunk, M. Karir, and C. Labovitz, "MRT routing
	information export format," draft-ietf-grow-mrt, work in		information export format," draft-ietf-grow-mrt, work in
	progress.		progress.
End of changes. 40 change blocks.
	169 lines changed or deleted		142 lines changed or added
This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/