Computing in Network Research Group P. Liu
Internet-Draft H. Yao
Intended status: Informational L. Geng
Expires: January 11, 2021 China Mobile
July 10, 2020
Differential Computing Resource Reservation
draft-liu-coin-differential-reservation-00
Abstract
Computing in the network may require the embedded computing
capability in the network device, such as gateway, switch, etc, and
there might be so much distributed computing task in the network.
Some new applications like AR/VR, motion control put forward higher
demand of network than before, and AI is also considered to be used
in the app and network.
In order to satisfy their demands, network may not only need to
reserve bandwidth resource, but also reserve computing resource.
This document analyzes the requirements of Serial distributed
computing model and give some reference solutions.
Requirements Language
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].
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 https://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 January 11, 2021.
Liu, et al. Expires January 11, 2021 [Page 1]
Internet-Draft Computing in network July 2020
Copyright Notice
Copyright (c) 2020 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
(https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
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.
Table of Contents
1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Existing Protocol . . . . . . . . . . . . . . . . . . . . . . 3
2.1. Resource Reservation Protocol . . . . . . . . . . . . . . 3
2.2. Path Computation Element Protocol . . . . . . . . . . . . 3
3. Problems of Resource Reservation . . . . . . . . . . . . . . 4
4. Reference Method . . . . . . . . . . . . . . . . . . . . . . 5
4.1. Distributed Resource Reservation . . . . . . . . . . . . 5
4.2. Centralized Resource Reservation . . . . . . . . . . . . 6
4.2.1. PCEP . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2.2. Netconf/Yang . . . . . . . . . . . . . . . . . . . . 7
5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . 8
6. Security Considerations . . . . . . . . . . . . . . . . . . . 8
7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8
8. Normative References . . . . . . . . . . . . . . . . . . . . 8
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9
1. Overview
From cloud computing to edge computing, computing power is
distributed to the customer side. In the future network and
computing convergence system, computing power will be distributed as
ubiquitous endogenous resources in each node of the network. The
user's request can be satisfied by calling the nearest node resource,
which is no longer limited to a specific node.
Resource reservation is usually used to guarantee the QoS of specific
application traffic. The reservation of network resources is same in
an end-to-end path, which means the reserved bandwidth resources will
not change from the client to the server, but computing is different.
Distributed computing will bring different computing power, and
different resources need to be reserved for different nodes. For
Liu, et al. Expires January 11, 2021 [Page 2]
Internet-Draft Computing in network July 2020
example, AI algorithm now has a model of step-by-step iteration at
multiple nodes. The previous iteration will affect the next
calculation results, and the computing resources required for each
iteration are not the same. From the perspective of network
standard, we hope to regard computing resources as the dimensions to
measure network performance, such as the same bandwidth, path, etc.,
while the traditional technologies of resource reservation have not
considered the reservation of computing resources, and have not
considered the differentiated resource reservation model.
2. Existing Protocol
Existing resource reservation protocols, such as Resource ReSerVation
Protocol(RSVP) and Path Computation Element Protocol (PCEP) , can be
used to reserve bandwidth resources. RSVP is a traditional protocol,
which only focuses on how to initiate the reservation of resources,
not the establishment of path. Later, RSVP-TE protocol was developed
for MPLS. PCEP was designed to separate the path calculation and
path establishment functions of RSVP-TE firstly, which means that the
path calculation part before resource reservation can be realized.
Therefore, RSVP and PCEP can be used together or separately.
2.1. Resource Reservation Protocol
Resource reservation is currently regarded as the key technology
configuration scheme to guarantee network QoS. In order to solve the
problem of bandwidth competition caused by the simultaneous arrival
of specific data flow and common data flow on the network node, the
bandwidth reservation management of data from the source node to the
destination node end-to-end is realized, so as to ensure the real-
time data flow QoS and delay requirements. The general process is as
follows:
The sender client initiates the request of resource reservation by
the path message. After determining the path, the sender sends the
request along the path, carrying the network requirements (latency,
etc.) to the receiver.
The receiver calculates the bandwidth and other resources that need
to be reserved for the network according to the request of the
sender. Then it returns according to the original path, and informs
the equipment to reserve resources one by one.
2.2. Path Computation Element Protocol
Path Computation Element Protocol (PCEP) is a centralized
configuration technology, which is usually used in software defined
network (SDN) as the South interface calculation and configuration
Liu, et al. Expires January 11, 2021 [Page 3]
Internet-Draft Computing in network July 2020
path information. PCE can improve the agility of the network. Any
change in network can be programmed using PCECC to learn the change
and react to it quickly and efficiently.
PEC can initiate resource reservation application to each device in
the path by the PCLRResv message. This message is sent by Path
Computation Element (PCE) to Path Computation Client (PCC) to sent
reserved label range for the network. The objects supported in this
message are stateful PCE request parameters objects, setting the
unique identifier for mapping request/response between PCEP and PCC.
3. Problems of Resource Reservation
In the model of computing in the network, the computing resource may
be distributed in multiple nodes. A task may be divided into several
parts to be executed by multiple nodes, including serial distribution
and parallel distribution. Parallel distribution can reserve
resources separately. However, in the serial computing model, the
calculation process of serial distribution algorithm is sequential,
and the results of the previous calculation need to be used in the
later calculation, so it will bring the following two problems:
Different computing nodes on the same path need different reserved
computing resources.
The bandwidth resources to be reserved maybe different after the
previous calculations in the same path.
A typical example is the artificial intelligence algorithm, which
involves the multi-layer convolution iterative process and can be
completed by multiple computing device in serial. As shown in the
figure, 20%, 30% and 50% tasks are calculated on network device 1, 3
and server respectively, and the calculation results of device 1 will
affect the subsequent calculation of device 3 and server. Then,
Network device 1, 3 and server need to reserve corresponding
computing resources respectively.
Since devices 1 and 3 calculated, the traffic will change after
passing through devices 1 and 3, so the bandwidth resources to be
reserved are different.
Traditional RSVP and other protocols do not consider the calculation
attribute, so the reserved value of bandwidth resource along the path
is unchanged, and the calculation resource cannot be reserved. PCEP
also dosen't consider about the comuputing resource.
Liu, et al. Expires January 11, 2021 [Page 4]
Internet-Draft Computing in network July 2020
+------+ +--------+
|Client| ->| Server |
+------+ \ +--------+ +--------+ +--------+ / +--------+
\->|network | |network | |network |->/ 50% of
|device 1|-->|device 2|-->|device 3| computing
+--------+ +--------+ +--------+ tasks
20% of 30% of
computing computing
tasks tasks
Serial distributed computing model
4. Reference Method
This scheme provides distributed and centralized resource reservation
reference scheme. It should be noted that for serial distributed
computing, we assume that the application side implements the
following functions:
The number of steps are involved in the calculation.
The computing proportion of calculation required at each node.
For bandwidth changes after each step of calculation, if this item
cannot be implemented, the same bandwidth resources will be reserved
by default.
4.1. Distributed Resource Reservation
Distributed resource reservation can be implemented by extending RSVP
or RSVP-TE protocol. The server receives the client's service
request, calculating the resource reservation strategy and return it.
The process is as follows:
1. The client sends the service request, carrying the service
requirements and the collected resource status of each node on the
path. They will be collected and added to the information that
carried by the service request.
2. The server receives the client's service request, then generates
the resource reservation strategy for target nodes on the path based
on the the service requirements and the resource status of each node,
and return the resource reservation strategy to each target node
along the path to reserve the resource.
The resource status at least includes the computing resource status
such as the catergery of chip, algorithm, etc. It can also includes
the network resource status such as bandwidth, delay, etc.
Liu, et al. Expires January 11, 2021 [Page 5]
Internet-Draft Computing in network July 2020
The resource reservation strategy at least includes the computing
resource reservation information of target nodes, which is as
follows:
1. Determine the serial distributed computing subtasks and computing
resources required by each computing subtask based on the service
request.
2. Select the target nodes for each computing subtask and generate
the computing resources reservation information to inform each target
node to reserve resource based on the computing resource status of
each node and the computing resources required by each computing
subtask.
Moreover, if the bandwidth change after each subtask can be
calculated, the resource reservation strategy can also carrying the
bandwidth resources reservation information.
It can be realized by defining new object of RSVP or RSVP-TE to
reserve different resources in each target nodes. The object can be
customized and extended with variable length. For example,
redefining a new class num as 30, carries the following message body:
[L = 0, IPv4, 64, IP address1, bandwidth 1, computing resource 1]
[L = 0, IPv4, 64, IP address2, bandwidth 2, computing resource 2]
[L = 0, IPv4, 64, IP address3, bandwidth 3, computing resource 3]
[L = 0, IPv4, 64, IP address4, bandwidth 4, computing resource 4]
......
It should be noted that the extended object can not only carry the
collected resources status of each node in the PATH message, but also
return the resource reservation strategy in the RESV message.
4.2. Centralized Resource Reservation
Centralized resource reservation can be realized by the network
manager. The manager receives the service request, calculates the
network and computing resources needed, and initiates resource
reservation configuration for the target nodes along the path.The
process is as follows:
The client sends a service request to the network manager.
Liu, et al. Expires January 11, 2021 [Page 6]
Internet-Draft Computing in network July 2020
Network manager selects the path according to the service request and
get the resource status of each node on the path.
Network manager generates the resource reservation strategy based on
the client's service request and resource status of each node.
Network manager sends resource reservation strategy to target nodes
to reserve the resource.
The resource status at least includes the computing resource status.
The resource reservation strategy at least includes the computing
resource reservation information of each target node. Which are the
same with chapter 4.1.
If at least one node in the selected path does not meet the resource
reservation requirements, it is necessary to re-select at least one
node in the path and get the resource status of the re-selected node
until the path meets the requirements of the resource reservation
strategy.
4.2.1. PCEP
By adding calculation force resource reservation field to resource
reservation object in PECP message, each calculation force flow has a
dynamic resource range based on the minimum reserved resource.
+---------+---------+-----------+----------+--------+
| Object | Label | Reserverd |Interface | In/ |
| Type | ID | Bandwidth |IP Address| Out |
+---------+---------+-----------+----------+--------+
PCEP extension
4.2.2. Netconf/Yang
It can also send resource reservation configuration to the target
nodes by netconf and defining the Yang structure. The reference Yang
module is as follows.
Liu, et al. Expires January 11, 2021 [Page 7]
Internet-Draft Computing in network July 2020
module: rs-computing-network
+--rw rs-computing-network
+--rw added-device[id]
| +--rw service id string
| +--rw user id string
| +--rw bandwitdh mbps
| +--rw computing resource tbd
+--rw deleted-device[id]
Yang Module
5. Conclusion
The draft proposes a method of differential reservation of computing
power and bandwidth resources. Because the traditional network does
not include computing power, the reservation of network resources is
the same on the path. This scheme can accurately reserve computing
power and network resources for the serial distributed computing
services. It also present the reference methods to realize different
resource reservation.Of course, there may be more and more
appropriate methods to achieve serial distributed computing power and
network resource reservation, which may require more analysis and
discussion.
6. Security Considerations
TBD.
7. IANA Considerations
TBD.
8. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[RFC5440] Vasseur, JP., Ed. and JL. Le Roux, Ed., "Path Computation
Element (PCE) Communication Protocol (PCEP)", RFC 5440,
DOI 10.17487/RFC5440, March 2009,
<https://www.rfc-editor.org/info/rfc5440>.
Liu, et al. Expires January 11, 2021 [Page 8]
Internet-Draft Computing in network July 2020
Authors' Addresses
Peng Liu
China Mobile
Beijing 100053
China
Email: liupengyjy@chinamobile.com
Huijuan Yao
China Mobile
Beijing 100053
China
Email: yaohuijuan@chinamobile.com
Liang Geng
China Mobile
Beijing 100053
China
Email: gengliang@chinamobile.com
Liu, et al. Expires January 11, 2021 [Page 9]