6TiSCH D. Dujovne, Ed.
Internet-Draft Universidad Diego Portales
Intended status: Informational LA. Grieco
Expires: January 5, 2015 Politecnico di Bari
MR. Palattella
University of Luxembourg
N. Accettura
University of California Berkeley
July 4, 2014
6TiSCH On-the-Fly Scheduling
draft-dujovne-6tisch-on-the-fly-03
Abstract
This document describes the environment, problem statement, and goals
of On-The-Fly (OTF) scheduling, a Layer-3 mechanism for 6TiSCH
networks. The purpose of OTF is to dynamically adapt the aggregate
bandwidth, i.e., the number of reserved soft cells between neighbor
nodes, based on the specific application constraints to be satisfied.
When using OTF, softcell and bundle reservation is distributed:
through the 6top interface, neighbor nodes negotiate the cell(s) to
be (re)allocated/deleted, with no intervention needed of a
centralized entity. This document aims at defining a module which
uses the functionalities provided by the 6top sublayer to (i) extract
statistics and (ii) determine when to reserve/delete soft cells in
the schedule. The exact reservation and deletion algorithm, and the
number and type of statistics to be used in the algorithm are out of
scope. OTF deals only with the number of softcells to be reserved/
deleted; it is up to 6top to select the specific soft cells within
the TSCH schedule.
Status of This Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
Dujovne, et al. Expires January 5, 2015 [Page 1]
Internet-Draft 6tisch-on-the-fly July 2014
This Internet-Draft will expire on January 5, 2015.
Copyright Notice
Copyright (c) 2014 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
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. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Allocation policy . . . . . . . . . . . . . . . . . . . . . . 3
3. Allocation methods . . . . . . . . . . . . . . . . . . . . . 5
4. Cell and Bundle Reservation/Deletion . . . . . . . . . . . . 6
5. Getting statistics and other information about cells through
6top . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
6. Events triggering algorithms in OTF . . . . . . . . . . . . . 8
7. Bandwidth Estimation Algorithms . . . . . . . . . . . . . . . 9
8. OTF external CoAP interface . . . . . . . . . . . . . . . . . 10
9. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 11
10. References . . . . . . . . . . . . . . . . . . . . . . . . . 11
10.1. Informative References . . . . . . . . . . . . . . . . . 11
10.2. External Informative References . . . . . . . . . . . . 12
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 12
1. Introduction
The IEEE802.15.4e standard [IEEE802154e] was published in 2012 as an
amendment to the Medium Access Control (MAC) protocol defined by the
IEEE802.15.4-2011 [IEEE802154] standard. The Timeslotted Channel
Hopping (TSCH) mode of IEEE802.15.4e is the object of this document.
On-The-Fly (OTF) scheduling is a 1-hop protocol with which a node
negotiates the number of soft cells scheduled with its neighbors,
without requiring any intervention of a centralized entity (e.g., a
PCE). This document describes the OTF allocation policies and
methods used by two neighbors to allocate one or more softcells in a
distribution fashion. It also proposes an algorithms for estimating
the required bandwidth (BW). This document defines the interface
Dujovne, et al. Expires January 5, 2015 [Page 2]
Internet-Draft 6tisch-on-the-fly July 2014
between OTF and the 6top sublayer ([I-D.wang-6tisch-6top]), to
collect and retrieve statistics, or allocate/delete cells and
bundles. This document defines a framework; the algorithm and
statistics used are out of scope. This draft follows the terminology
defined in [I-D.ietf-6tisch-terminology] and addresses the open issue
related to the scheduling mechanisms raised in
[I-D.ietf-6tisch-tsch].
2. Allocation policy
OTF is a distributed scheduling protocol which increases/decreases
the bandwidth between two neighbor nodes (i.e., adding/deleting
sofcells) by interacting with the 6top sublayer. It retrieves
statistics from 6top, and uses that information to trigger 6top to
add/delete softcells to a particular neighbor. The algorithm which
decides when to add/delete softcells is out of scope. For example,
6top might decide to add a cell if some queue of outbound frames is
overflowing. Similarly, OTF can delete cells when the queue has been
empty for some time. OTF only triggers 6top to add/delete the soft
cells, it is the responsibility of the 6top sublayer to determine the
exact slotOffset/channelOffset of those cells. In this document, the
term "cell" and "soft cell" are used interchageably.
All the softcells allocated are part of best effort track, i.e. with
TrackID=00, as defined i [I-D.wang-6tisch-6top]. These cells can be
used for forwarding any packet in the queue, regardless of the
specific track it belongs to. OTF manages the global bandwidth
requirements between two neighbor nodes; per-track management is
currently out of scope.
OTF is prone to schedule collisions. Nodes might not be aware of the
cells allocated by other pairs of nodes. A schedule collision occurs
when the same cell is allocated by different pairs in the same
interference space. The probability of having allocation collision
may be kept low by grouping cells into chunks (see
[I-D.ietf-6tisch-terminology] and [I-D.ietf-6tisch-architecture] for
more details). The use of chunks is outside the scope of this
current version of the OTF draft.
The "allocation policy" is the algorithm used by OTF to decide when
to increase/decrease the bandwidth allocated between two neighbor
nodes in order to satisfy the traffic requirements. These
requirements can be expressed in terms of throughput, latency or
other constraints.
An OTF allocation policy MAY be defined according to a combination of
two different approaches: reactive and proactive.
Dujovne, et al. Expires January 5, 2015 [Page 3]
Internet-Draft 6tisch-on-the-fly July 2014
In a reactive approach, the allocation policy follows the increased/
decreased need for bandwidth. Upon reception of a bandwidth request,
OTF sends softcell allocation requests to the 6top sublayer. OTF
estimates the number of cells to be allocated per neighbor. If the
traffic exchanged between two neighbors reduces, OTF asks 6top to de-
allocate one or more cells. Once the cells are deleted, 6top
notifies OTF, which updates its internal state.
In a proactive approach, the allocation policy over-provisions the
number of cells reserved in a bundle between two neighbors, i.e.,
cells are scheduled in advance. When OTF issues a bundle allocation
request to 6top, it indicates the desired size of the bundle and the
TrackID=00. 6top selects the cells belonging to the bundle on the
best effort track. Based on the network traffic conditions (e.g.,
queue utilization), some portion of those cells are used for
communication. In any case, allocated cells within a bundle are
consecutive, starting from the first cell in the block. The cells
which are not currently used, are still reserved for that pair of
nodes, for possible future use.
It is up to the implementor to select the approach most appropriate
for the application. The reactive approach is, in general, be more
energy-efficient (it allocates only the cells needed), at the expense
of increased cell allocation latency (negotiating to add/delete cells
takes some time).
The proactive approach compared to the reactive one reduces the cell
allocation latency. Cells within a bundle are over-provisioned, and
a priori scheduled. When needed, the 6top sublayer of the node can
allocate them, without going through any negotiation phase with the
6top layer of the neighbor node. Thus, the proactive approach
provides a low-delay response after a surge in bandwidth usage. In
fact, soft cells within a bundle are already scheduled and become
immediately available, upon bandwidth request, without the need of a
negotiation phase. The use of bundles does force the receiver module
of the node to be active during the whole length of the bundle,
resulting in increased power consumption.
This document introduces the following parameters to accomplish both
approaches previously described:
SCHEDULEBW: The amount of cells scheduled in a bundle on the best
effort track between two neighbors.
REQUIREDBW: Bandwidth requested by OTF to 6top, a non-negative
number. How this is computed is out of the scope. It MAY be the
an instantaneous bandwidth request, or a value averaged on several
measurement, or an over-provisioned value.
Dujovne, et al. Expires January 5, 2015 [Page 4]
Internet-Draft 6tisch-on-the-fly July 2014
PROACTIVETHRESH: Threshold parameter to introduce pro-activity in
the allocation policy, described below. It is a non-negative
bandwidth value. What value to use is application specific and
out of scope. The maximum acceptable value for this parameter is
equal to the current SCHEDULEBW.
The OTF allocation policy compares the required bandwidth against the
scheduled one, using the pro-activity threshold for bounding the
signaling overhead due to negotiations of new cells. In details:
1. If REQUIREDBW is greater than SCHEDULEBW, OTF asks 6top to add
REQUIREDBW-SCHEDULEBW cells to the bundle on the best effort
track.
2. If REQUIREDBW is greater or equal than SCHEDULEBW-
PROACTIVETHRESH, and it is lower than or equal to SCHEDULEBW, OTF
does not perform any bundle resizing, since the scheduled
bandwidth is sufficient for managing the current traffic
conditions.
3. If REQUIREDBW is lower than SCHEDULEBW-PROACTIVETHRESH, OTF asks
6top to delete SCHEDULEBW-PROACTIVETHRESH-REQUIREDBW from the
bundle on the best effor track.
A purely reactive approach uses PROACTIVETHRESH=0. In this case, OTF
does not perform any allocating/deallocating operation when the
required bandwidth is equal to the scheduled one.
A purely proactive approach uses PROACTIVETHRESH=SCHEDULEBW. In this
case, OTF resizes the bundle only when the required bandwidth is
greater than the scheduled one.
3. Allocation methods
Beyond the allocation policies that describe the approach used by OTF
for fulfilling the node bandwidth requests, the OTF framework also
includes Allocation Methods that specify how OTF issues commands to
the 6top sublayer. In other words, the allocation methods represent
the mechanisms that are used by the allocation policies.
In detail, OTF includes two distinct allocation methods: soft cell
and bundle allocation methods. Each Allocation Policy can use either
one or both allocation methods. As specified in
[I-D.wang-6tisch-6top], 6top provides a set of commands that allows
OTF to allocate/delete soft cells. The same set of commands can be
used for reserving bundles.
Dujovne, et al. Expires January 5, 2015 [Page 5]
Internet-Draft 6tisch-on-the-fly July 2014
With the soft cell allocation method, OTF has 6top reserve a single
soft cell on the best effort track, for allowing a given node to
exchange traffic with a specific neighbor. The 6top layer allocates
and maintains this cell. If a bundle is already reserved between the
same pair of neighbors, on the same track, this request translates
into a bundle resize request. The newly allocated cell increase the
size of the already existing bundle. Similarly, when OTF realizes
there is a reduction of traffic exchanged between the two neighbors,
it may asks 6top to delete a softcell from the best effort track,
i.e. to decrease the size of the bundle on the best effort track. If
no bundle with TrackID=00 exist, the 6top softcell create command
generates a new bundle of size 1.
With the bundle allocation method, OTF sends bundle allocation
requests to 6top sublayer, specifying the bundle size (the number of
soft cells) and the TrackID=00. Scheduling N softcells is equivalent
to asking for a bundle of size N. The cells within the bundle are
allocated by 6top (and thus, used for traffic exchange) only
afterwards, according to the nodes bandwidth need.
4. Cell and Bundle Reservation/Deletion
In order to reserve/delete softcells, OTF interacts with 6top
sublayer. To this aim OTF uses the following set of commands offered
by 6top: CREATE.softcell, and DELETE.softcell. When creating
(deleting) a softcell, OTF specified the track the cell belongs to
(i.e., best effort track, TrackID=00), but not its slotOffset and
channelOffset. If at least one cell on the best effort track already
exists, the CREATE.softcell and DELETE.softcell, translate into
INCREASE and DECREASE the bundle size, respectively. 6top is
responsible for picking the specific cell to be added/deleted within
the bundle. Before being able to do so, source and destination nodes
go through a cell negotiation process. This process is out of scope
of 6top and OTF. In order to reserve a best effort bundle, OTF uses
the CREATE.softcell command, set TrackID=00, but asks 6top for
multiple softcells. Following OTF request, 6top either (i) creates a
new bundle, if no cells were reserved already on the best effort
track, or (ii) increases the bundle size of the already existing
best-effort bundle. By using the DELETE.softcell command, and asking
for deleting multiple softcells, OTF has 6top delete the entire best
effort bundle.
OTF provides a policy for 6top to generate CREATE/DELETE.softcells
commands, policy that is out of 6top scope [I-D.wang-6tisch-6top].
Such policy is not the only one that can be used by 6top. Others may
be defined in the future.
Dujovne, et al. Expires January 5, 2015 [Page 6]
Internet-Draft 6tisch-on-the-fly July 2014
5. Getting statistics and other information about cells through 6top
Statistics are kept in 4 data structure of 6top MIB: CellList,
MonitoringStatusList, NeighborList, and QueueList.
CellList provides per-cell statistics. From this list, an upper
layer can get per-bundle statistics. OTF may have access to the
CellList, by using the CoAP-YANG Model, but actually cell-specific
statistics are not significant to OTF, since softcells can be re-
allocated in time by 6top itself, based on network conditions.
MonitoringStatusList provides per-neighbor and slotframe statistics.
From it an upper layer (e.g., OTF) can get per bundle overview of
scheduling and its performance. Such list contains information about
the number of hard and soft cells reserved to a given node with a
specific neighbor, and the QoS (that can be expressed in form of
different metrics: PDR, ETX, RSSI, LQI) on the actual bandwidth, and
the over-provisioned bandwidth (which includes the over-provisioned
cells). 6top can use such list to operate 6top Monitoring Functions,
such as re-allocating cells (by changing their slotOffset and/or
channelOffset) when it finds out that the link quality of some
softcell is much lower than average. Unlike 6top, OTF does not
operate any re-allocation of cells. In fact, OTF can ask for more/
less bandwidth, but cannot move any cell within the schedule. Thus,
the 6top Monitoring function is useful to OTF, because it can provide
better cells for a given bandwidth requirement, specified by OTF.
For instance, OTF may require some additional bandwidth (e.g. 2 cells
in a specific slotframe) with PDR = 75%; then, 6top will reserve 3
slots in the slotframe to meet the bandwidth requirement. In
addition, when the link quality drop to 50%, 6top will reserve 4
slots to keep meeting the bandwidth requirement. Given that OTF
operates on the global bandwidth between two neighbor nodes, it does
not need to be informed from 6top about cells' re-allocation.
NeighborList provides per-neighbor statistics. From it, an upper
layer can understand the connectivity of a pair of nodes. Based on
the quality of the link, e.g., LQI under threshold, OTF may ask 6top
to delete some cells, in order to reserve them for better-connected
links.
QueueList provides per-Queue statistics. From it, an upper layer can
know the traffic load. OTF, based on such queue statistics (e.g.,
average length of the queue, average age of the packet in queue,
etc.) may trigger a 6top CREATE.softcell (DELETEsoftcell) command for
increasing (decreasing) the bandwidth and be able to better serve the
packets in the queue.
Dujovne, et al. Expires January 5, 2015 [Page 7]
Internet-Draft 6tisch-on-the-fly July 2014
6. Events triggering algorithms in OTF
The Algorithms running within OTF MUST be event-oriented. As a
consequence, OTF requires to connect the algorithms with external
events to trigger their execution. The algorithm also generates one
or more events when it is executed, such as a new softcell
allocation. Both type of events, the one which trigger the algorithm
and the ones which are generated by the execution of the algorithm
are called OTF events.
The following notation is used on the definition of OTF events:
BW <- BWA(B,T,S(T)) where:
BWA: Bandwidth allocation algorithm
BW: Bandwidth
T: Best Effort Track
B: Bundle
S(B,T) Statistics for bundle B on track T
M(B,T): Actual bundle size for bundle B on track T
The OTF events are defined as:
Event A: A new bundle B on track T is created. The OTF events
generated by the algorithm are:
1. Add a new entry in the storage M for bundle B on track T.
2. Ask 6top for S(B,T).
3. BW <- BWA(B,T,S(T)).
4. Ask 6top to allocate a bundle of size BW.
5. M(B,T)<-BW.
Event B: A packet is waiting to be transmitted on any track, but no
cell is available (i.e., saturation). The OTF events generated by
the algorithm are:
1. Collect stats S from 6top.
2. BW <- BWA(B,T,S(T))).
Dujovne, et al. Expires January 5, 2015 [Page 8]
Internet-Draft 6tisch-on-the-fly July 2014
3. Ask 6top to increase the bundle size up to BW.
4. If (allocation successful) then M(B,T)<-BW.
Event C: The usage of a bundle B on track T is too low, below a pre-
established threshold. The OTF events generated by the algorithm
are:
1. Collect stats S from 6top.
2. BW <- BWA(B,T,S(T)).
3. Ask 6top to decrease the bundle size to BW.
4. If (allocation successful) then M(B,T)<-BW.
Event D: The usage of a bundle B on track T is too high, above a pre-
established threshold. The OTF events generated by the algorithm
are:
1. Collect stats S from 6top.
2. BW <- BWA(B,T,S(T)).
3. Ask 6top to increase the bundle size to BW.
4. If (allocation successful) then M(B,T)<-BW.
Event E: Bundle B on track T is deleted. The OTF events generated by
the algorithm are:
1. purge M(B,T).
7. Bandwidth Estimation Algorithms
OTF supports different bandwidth estimation algorithms that can be
used by a node in a 6TiSCH network for checking the current traffic
condition and thus the actual bandwidth usage. By doing so, one can
adapt (increase or increase) the number of scheduled cells/bundles
for a given pair of neighbors (e.g., parent node and its child),
according to their needs. OTF supports several bandwidth estimation
algorithms numbered 0 to 255 in the OTF implementation. The first
algorithm (0) is reserved to the default algorithm that is described
below. By using SET and GET commands, one can set the specific
algorithm to be used, and get information about which algorithm is
implemented.
Dujovne, et al. Expires January 5, 2015 [Page 9]
Internet-Draft 6tisch-on-the-fly July 2014
Steps of the default bandwidth estimation algorithm, running over a
parent node:
Step 1: Collect the bandwidth requests from child nodes (incoming
traffic).
Step 2: Collect the node bandwidth requirement from the application
(self/local traffic).
Step 3: Collect the current outgoing scheduled bandwidth (outgoing
traffic).
Step 4: If (outgoing < incoming + self) then SCHEDULE soft cells/
bundles to satisfy bandwidth requirements.
Step 5: If (outgoing > incoming + self) then DELETE the soft cells
that are not used.
Step 6: Return to step 1.
The default bandwidth estimation algorithm introduced in this
document adopts a reactive allocation policy; it is possible to
configure proactivity by using a given PROACTIVETHRESH value. In
this case, at Step 4, new soft cells will be scheduled, using the
cell allocation method, only if there are no free cells in the bundle
that can satisfied the current bandwidth request. The node asks 6top
for increasing the bundle size by using the bundle allocation method.
8. OTF external CoAP interface
In order to select the current OTF algorithm and provide functional
parameters from outside OTF, this module uses CoAP with YANG as the
data model. The algorithm number and the parameters MUST be invoked
in different CoAP calls.
The path to select the algorithm is '6t/e/otf/alg' with A as the
algorithm number.
+------------------------------------------+
Header | POST |
+------------------------------------------+
Uri-Path| /6t/e/otf/alg |
+------------------------------------------+
Options | CBOR( {AlgNo: 123} ) |
+------------------------------------------+
Figure 1: Algorithm number POST message
Dujovne, et al. Expires January 5, 2015 [Page 10]
Internet-Draft 6tisch-on-the-fly July 2014
To obtain the current algorithm number:
+------------------------------------------+
Header | GET |
+------------------------------------------+
Uri-Path| /6t/e/otf/alg |
+------------------------------------------+
Options | Accept: application/cbor |
+------------------------------------------+
Figure 2: Algorithm number GET message
An example is: 'coap://[aaaa::1]/6t/e/otf/alg'
The current algorithm parameter path is '6t/e/otf/alg/par'.
+------------------------------------------+
Header | POST |
+------------------------------------------+
Uri-Path| /6t/e/otf/alg/par |
+------------------------------------------+
Options | CBOR( {Par: 0x1234} ) |
+------------------------------------------+
Figure 3: Algorithm number POST message
An example follows: 'coap://[aaaa::1]/6t/e/otf/alg/par'
9. Acknowledgments
Special thanks to Prof. Kris Pister for his valuable contribution in
designing the default Bandwidth Estimation Algorithm, and to Prof.
Qin Wang for her support in defining the interaction between OTF and
6top sublayer.
Thanks to the Fondecyt 1121475 Project, to INRIA Chile "Network
Design" group and to the IoT6 European Project (STREP) of the 7th
Framework Program (Grant 288445).
10. References
10.1. Informative References
[I-D.ietf-6tisch-terminology]
Palattella, M., Thubert, P., Watteyne, T., and Q. Wang,
"Terminology in IPv6 over the TSCH mode of IEEE
802.15.4e", draft-ietf-6tisch-terminology-01 (work in
progress), February 2014.
Dujovne, et al. Expires January 5, 2015 [Page 11]
Internet-Draft 6tisch-on-the-fly July 2014
[I-D.ietf-6tisch-architecture]
Thubert, P., Watteyne, T., and R. Assimiti, "An
Architecture for IPv6 over the TSCH mode of IEEE
802.15.4e", draft-ietf-6tisch-architecture-02 (work in
progress), June 2014.
[I-D.ietf-6tisch-tsch]
Watteyne, T., Palattella, M., and L. Grieco, "Using
IEEE802.15.4e TSCH in an LLN context: Overview, Problem
Statement and Goals", draft-ietf-6tisch-tsch-00 (work in
progress), November 2013.
[I-D.wang-6tisch-6top]
Wang, Q., Vilajosana, X., and T. Watteyne, "6TiSCH
Operation Sublayer (6top)", draft-wang-6tisch-6top-00
(work in progress), October 2013.
10.2. External Informative References
[IEEE802154e]
IEEE standard for Information Technology, "IEEE std.
802.15.4e, Part. 15.4: Low-Rate Wireless Personal Area
Networks (LR-WPANs) Amendament 1: MAC sublayer", April
2012.
[IEEE802154]
IEEE standard for Information Technology, "IEEE std.
802.15.4, Part. 15.4: Wireless Medium Access Control (MAC)
and Physical Layer (PHY) Specifications for Low-Rate
Wireless Personal Area Networks", June 2011.
Authors' Addresses
Diego Dujovne (editor)
Universidad Diego Portales
Escuela de Informatica y Telecomunicaciones
Av. Ejercito 441
Santiago, Region Metropolitana
Chile
Phone: +56 (2) 676-8121
Email: diego.dujovne@mail.udp.cl
Dujovne, et al. Expires January 5, 2015 [Page 12]
Internet-Draft 6tisch-on-the-fly July 2014
Luigi Alfredo Grieco
Politecnico di Bari
Department of Electrical and Information Engineering
Via Orabona 4
Bari 70125
Italy
Phone: 00390805963911
Email: a.grieco@poliba.it
Maria Rita Palattella
University of Luxembourg
Interdisciplinary Centre for Security, Reliability and Trust
4, rue Alphonse Weicker
Luxembourg L-2721
LUXEMBOURG
Phone: (+352) 46 66 44 5841
Email: maria-rita.palattella@uni.lu
Nicola Accettura
University of California Berkeley
Berkeley Sensor & Actuator Center
490 Cory Hall
Berkeley, California 94720
USA
Email: nicola.accettura@eecs.berkeley.edu
Dujovne, et al. Expires January 5, 2015 [Page 13]