Internet DRAFT - draft-mjsraman-panet-ospf-power-topo

draft-mjsraman-panet-ospf-power-topo



 



PANET Working Group                                        Shankar Raman
INTERNET-DRAFT                                Balaji Venkat Venkataswami
Intended Status: Experimental RFC                           Gaurav Raina
Expires: September 28, 2013                                  Vasan Srini
                                                              IIT Madras
                                                          March 27, 2013


       Power Based Topologies and TE-Shortest Power Paths in OSPF
                draft-mjsraman-panet-ospf-power-topo-02


Abstract

   In a Interior Gateway Protocol like OSPF (Open Shortest Path First)
   the computation of the Constrained shortest path to destinations is
   computed for an area say a backbone or a non-backbone area using the
   TE-metrics advertised in the area. With importance given to the
   reduction of power within a network it becomes important to provide a
   solution that reduces the power consumed amongst routers and links
   that make up the network (in this case an area or a collection of
   areas including the backbone and non-backbone areas). This proposal
   aims at providing such a solution by producing a power topology of
   the area / areas. This power topology is constructed by assigning
   metrics to links based on the power consumed by the linecards (and
   hence their respective ports in an indirect way) of adjacent routers
   that are interconnected by each such link.



Status of this Memo

   This Internet-Draft is submitted to IETF in full conformance with the
   provisions of BCP 78 and 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/1id-abstracts.html

 


Shankar Raman et.al,     Expires September 2013                 [Page 1]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html


Copyright and License Notice

   Copyright (c) 2012 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  . . . . . . . . . . . . . . . . . . . . . . . . .  3
     1.1  Terminology . . . . . . . . . . . . . . . . . . . . . . . .  3
     1.2 Low-power routers and switches . . . . . . . . . . . . . . .  3
     1.3 Power reduction using routing and traffic engineering  . . .  3
   2. Methodology . . . . . . . . . . . . . . . . . . . . . . . . . .  4
       2.0.1 ERRATA . . . . . . . . . . . . . . . . . . . . . . . . .  8
     2.1 Power Bias . . . . . . . . . . . . . . . . . . . . . . . . .  8
       2.1.1 Advertising Available POWER  . . . . . . . . . . . . . .  8
     2.2 ECMP links . . . . . . . . . . . . . . . . . . . . . . . . .  9
     2.3 Dampening the side effects of constant change  . . . . . . .  9
     2.4 Calculating power shortest paths in an Area  . . . . . . . .  9
       2.4.1 Power profiles of Routers and Switches . . . . . . . . . 10
         2.4.1.3 Need to advertise both available power and
                 consumed power . . . . . . . . . . . . . . . . . . . 14
       2.4.2 Power to Available Bandwidth ratio in a TLV  . . . . . . 14
   3. Conclusion  . . . . . . . . . . . . . . . . . . . . . . . . . . 16
   3  Security Considerations . . . . . . . . . . . . . . . . . . . . 17
   4  IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 17
   5  References  . . . . . . . . . . . . . . . . . . . . . . . . . . 17
     5.1  Normative References  . . . . . . . . . . . . . . . . . . . 17
     5.2  Informative References  . . . . . . . . . . . . . . . . . . 17
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 18



 


Shankar Raman et.al,     Expires September 2013                 [Page 2]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


1. Introduction

   Estimates of power consumption for the Internet predict a 300%
   increase, as access speeds increase from 10 Mbps to 100 Mbps [3],
   [8]. Access speeds are likely to increase as new video, voice and
   gaming devices get added to the Internet. Various approaches have
   been proposed to reduce the power consumption of the Internet such as
   designing low-power routers and switches, and optimizing the network
   topology using traffic engineering methods [2].


1.1  Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC 2119 [RFC2119].

1.2 Low-power routers and switches

   Low-power router and switch design aim at reducing the power consumed
   by hardware architectural components such as transmission link,
   lookup tables and memory. In [4] it is shown that the router's link
   power consumption can vary by 20 Watts between idle and traffic
   scenarios. Hence the authors suggest having more line cards and
   running them to capacity: operating the router at full throughput
   will lead to less power per bit, and hence larger packet lengths will
   consume lower power. The two important components in routers that
   have received attention for high power consumption are buffers and
   TCAMs. Buffers are built using dynamic RAM (DRAM) or static RAM
   (SRAM). SRAMs are limited in size and consume more power, but have
   low access times. Guido [1] states that a 40Gb/s line card would
   require more than 300 SRAM chips and consume 2.5kW. DRAM access times
   prevent them from being used on high speed line cards. Sometimes the
   buffering of packets in DRAM is done at the back end, while SRAM is
   used at the front end for fast data access. But these schemes cannot
   scale with increasing line speeds. Some variants of TCAMs have been
   proposed for increasing line speeds and for reduced power consumption
   [7].

1.3 Power reduction using routing and traffic engineering

   At the Internet level, creating a topology that allows route
   adaptation, capacity scaling and power-aware service rate tuning,
   will reduce power consumption. In [8] the author has proposed a
   technique to traffic engineer the data packets in such a way that the
   link capacity between routers is optimized. Links which are not
   utilized are moved to the idle state. Power consumption can be
   reduced by trading off performance related measures like latency. For
 


Shankar Raman et.al,     Expires September 2013                 [Page 3]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   example, power savings while switching from 1 Gbps to 100 Mbps is
   approximately 4 W and from 100 Mbps to 10 Mbps around 0:1 Watts.
   Hence instead of operating at 1 Gbps the link speed could be reduced
   to a lower bandwidth under certain conditions for reduced power
   consumption.

   Multi layer traffic engineering based methods make use of parameters
   such as resource usage, bandwidth, throughput and QoS measures, for
   power reduction. In [6] an approach for reducing Intra-AS power
   consumption for optical networks that uses Djikstra's shortest path
   algorithm is proposed. The input to this method assumes the existence
   of a network topology using which an auxiliary graph is constructed.
   Power optimization is done on the auxiliary graph and traffic is
   routed through the low-power links. However, the algorithm expects
   the topology to be available for getting the auxiliary graph. While
   [6] handles optical networks and their corresponding power
   consumption, it does not take into account other link layer
   technologies. It is specialized for optical and not for heterogenous
   links that will exist in common OSPF domains. 

   The proposal we make in this document indicates ways to solve the
   power reduction problem, by calculating a POWER metric whose
   importance is highlighted in the below mentioned sections. This POWER
   metric is obtained by including the factors such as power consumed by
   a linecard on a single chassis or multi-chassis router and
   consequently a port on that linecard by proportionally calculating
   power consumed for that port and hence for the link. The other factor
   that is taken into account is the Available Bandwidth on that port
   and hence on that link.



2. Methodology

   For each router / switch there exist linecards and each linecard has
   a set of ports or sometimes just one port of high capacity. This
   usually applies on routers and switches that are either single
   chassis or multi-chassis in their characterisation. By single chassis
   we mean that there exists a single chassis and slots for the Route
   Processor Card (one or more of these) typically upto to two of them,
   and one or more slots for linecards each having their respective
   characteristics such as number of ports (port density), type of such
   ports (SONET, ethernet, ATM etc..) usually depending on the link
   layer technology they support. Links are connections between ports on
   these linecards to other ports on linecards of other single chassis
   or multi-chassis system. A multi-chassis system is one that has
   multiple such chassis interconnceted amongst each other to form a
   single logical view of the system. Both single and multi-chassis have
 


Shankar Raman et.al,     Expires September 2013                 [Page 4]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   linecards and respective ports on these linecards. Multi-chassis
   typically have a switch fabric chassis which connects each of these
   chassis to each other or to chassis of other multi-chassis or single
   chassis systems.

   Consider the following topology...

   Router A               Router B              Router C
   +---+---+             +---+---+             +-------+
   |   |   |             |   |   |             |   |   |
   |LC1|LC2|             |LC1|LC2|             |LC1|LC2|
   |   |   |             |   |   |    L11      |   |   |
   | P1| P1|             | P1| P1|-------------- P1| P1|---+
   | P2| P2|--+          | P2| P2|    L12      | P2| P2|   |
   | P3| P3|  |   L4     | P3| P3|-------------- P3| P3|   |
   | P4| P4|--+----------- P4| P4|         +---- P4| P4|   |
   | P5| P5|  |       +----P5| P5--+    L5 |   | P5| P5|   |
   | | | | |  |       |  |   |   | |       |   |   | | |   |
   +-|-+-|-+  |L3     |  +---+---+ |       |   +---+-|-+   | L13
     |   |    |       +------------+-------+         |     |
     |   |L2  |                L5  |                 |     |
     |   +----+------------+       |                 |     |
     |        |            |       |                 |     |
     |L1      |            |       |L6               |     |
     |        | Router D   |       |   Router E   L12|     | Router F
     |        | +---+---+  |       |  +---+---+      |     |+-------+
     |        | |   |   |  |L2     |  |   |   |      |     ||   |   |L
     |        | |LC1|LC2|  |       |  |LC1|LC2|      |     ||LC1|LC2|1
     |        | |   |   |  |       |  |   |   |      |     ||   |   |4..
     |        +-| P1| P1---+       |  | P1| P1|------+     || P1| P1|-> 
     |          | P2| P2|     L7   +--- P2| P2|            +--P2| P2|-> 
     |          | P3| P3|-------------- P3| P3|   L10       | P3| P3|-> 
     +----------| P4| P4|         +---- P4| P4|-------------- P4| P4|
                | P5| P5|         | +-- P5| P5|        +----- P5| P5|
                | | |   |         | | |   |   |        |    |   |   |
                +-|-+---+     L8  | | +---+---+  L9    |    +---+---+
                  +---------------+ +------------------+











 


Shankar Raman et.al,     Expires September 2013                 [Page 5]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   The table of links between the various routers (which are assumed to
   be single chassis systems) is as follows...

   +--------+----------+-----------+-----------+-----------+----------+
   | Links  |  Routers |  LC <> LC | Port Conn.| Capacity  |Available |
   |        |          |           |           |           |Bandwidth |
   +--------+----------+-----------+-----------+-----------+----------+
   |  L1    |  A <> D  |  LC1<>LC1 |  P5<>P4   |   10G      |  7.5    |
   |  L2    |  A <> D  |  LC2<>LC2 |  P5<>P1   |   10G      |  6.0    |
   |  L3    |  A <> D  |  LC2<>LC1 |  P2<>P1   |   10G      |  4.0    |
   |  L4    |  A <> B  |  LC2<>LC1 |  P4<>P4   |   10G      |  3.0    |
   |  L5    |  B <> C  |  LC1<>LC1 |  P5<>P4   |   10G      |  3.5    |
   |  L6    |  B <> E  |  LC1<>LC1 |  P6<>P2   |   10G      |  1.0    |
   |  L7    |  D <> E  |  LC2<>LC1 |  P3<>P3   |   10G      |  6.0    |
   |  L8    |  D <> E  |  LC1<>LC1 |  P5<>P4   |   10G      |  1.5    |
   |  L9    |  E <> F  |  LC1<>LC2 |  P5<>P5   |  100G      | 20.0    |
   |  L10   |  E <> F  |  LC2<>LC1 |  P4<>P4   |   10G      |  2.5    |
   |  L11   |  B <> C  |  LC2<>LC1 |  P1<>P1   |   10G      |  3.0    |
   |  L12   |  E <> C  |  LC2<>LC2 |  P1<>P5   |   10G      |  2.0    |
   |  L13   |  C <> F  |  LC2<>LC1 |  P1<>P2   |   10G      |  1.0    |
   |  L14   |  F <> OA |  LC2<>    |  P1<>     |            |         |
   |        |          |           |           |            |         |
   +--------+----------+-----------+-----------+------------+---------+


   In the above topology assume all point-to-point links between the
   routers. For now we will deal with P2P links alone and not venture
   into Broadcast Multi-access links or Non-Broadcast Multi-access links
   etc.. It is suffice to show how the scheme works for P2P links and
   then move more specifically to other types of networks to demonstrate
   this method of calculating the power topology of the network in the
   figure.

   Each linecard consumes a certain amount of power and it is vendor
   dependent as to how the power consumed relates to the Available
   Bandwidth on any of the links to which the linecard connects to. It
   is possible that the said topology of routers come from one vendor or
   from multiple vendors. It is assumed that the algorithm proposed will
   have the power consumed by a linecard available as a readable value
   in terms of W or kW or whichever measurable metric that is provided
   by the vendor. 

   It is possible that some of the Linecards are more capable than the
   others. Consider that Router A is a more capable router with more
   powerful linecards with higher port density. This is not shown in the
   figure, but assume so. LC1, LC2 on Router A could be consuming more
   power than the other Linecards on other routers. The main reason
   could be that LC1 and LC2 may have higher port density or higher
 


Shankar Raman et.al,     Expires September 2013                 [Page 6]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   speed ports than the other routers. In order to calculate the power
   consumed on a link by a linecard it is important that we normalize
   the power as power consumed per port. Here the ports are normalized
   to lowest common denominator. If all links in the topology have 10G
   port capacity then the power calculated should be in terms power
   consumed per 10G port.

   Assuming we have done this normalization we go on to calculate the
   POWER metric for each of the ports involved in a link which is
   derived as follows...

   POWER metric    = Power consumed per XG (normalized bandwidth) port
   for a given       -------------------------------------------------
   Port on a LC              Available Bandwidth on that port

   Assume link L1. The ports concerned are both 10G and the ports are P5
   on Router A and P4 on Router D. For calculating the POWER metric for
   a link which we will call PWRLINK we calculate the POWER metric for
   each side of the link and average the two to get PWRLINK.

   So PWRLINK for L1 =  POWER for P5 on LC1  +  Power for P4 on LC1
                         on Router A               on Router D
                        ============================================
                                            2

   The above can also be weighted if there is a multi-capacity port on
   one side of the link and not on the other. A multi-capacity link is
   one which provides multiple bandwidth capabilities such (1G/10G/100G)
   for example but auto-negotiates with other end to provide a lesser
   than highest capacity service.

   The PWRLINK metrices once calculated are flooded in already defined
   OSPF-TE-LSA as an adapted TE-metric and is typically flooded as a
   link characteristic. 

   It is important to note that the denominator for POWER metric is
   Available Bandwidth instead of Available Bandwidth on that port. The
   Available Bandwidth is measured in terms of intervals and not as
   discrete quantities. This is in order not to flood PWRLINK metrics
   into the OSPF area in LSAs very frequently as Bandwidth may
   constantly change. The same applies to POWER metric as well. 

   Once the LSAs have been flooded the Routers run CSPF on the graph of
   the topology with PWRLINKs assigned to the links and calculate the
   PWRLINK based paths which consume the least power. The shortest power
   paths based on this topology can be used for forwarding high
   bandwidth streams and to optimally use power within the area.

 


Shankar Raman et.al,     Expires September 2013                 [Page 7]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   The Available Bandwidth column shows the Available bandwidth of the
   link corresponding to the row and column intersection. This figure is
   used as the numerator in the POWER metric computation for that port.

2.0.1 ERRATA 

   ERRATA : Previously the experiments were carried out with Available
   Utilization since only 10G and 100G ports were considered. This
   baselines the metric to 10G ports and proportionality thereof. But in
   reality the actual Available Bandwidth needs to be considered for
   real world experiments. Hence this draft has been changed to reflect
   the Available Bandwidth to be taken as the denominator of the formula
   thereof.

   In our previous experiments the 100G link if it showed a utilization
   of 0.2 would end up as a high POWER metric and hence would be totally
   avoided. In reality this link may have been a more power optimal link
   given that if it had a first power profile (Please refer section on
   Power Profiles). Dividing the Power consumed or Available Power by
   the Available Bandwidth gives a better picture of how much power cost
   per Gb is consumed and normalizes the metric amongst links of varying
   bandwidth.

2.1 Power Bias

   Assume in the figure that there exist Routers A and D and that there
   is a bias on the link L1 in such a way that Router D computes a POWER
   metric of 10 and the Router D computes a POWER metric of 2 on the
   ports P5 and P4 respectively. Now the PWRLINK would be 6 for that
   link L1. Thus even if one side is excessively power guzzling then the
   PWRLINK moves up and thus is less preferred in the CSPF algorithm and
   path computation based on the Power topology. 

   If there is no bias and both the sides of the link are optimal in
   their power usage then the metric stays low even if more streams are
   sent on it. This is the main objective that is set out for router and
   switch manufacturers in the single chassis and multi-chassis world,
   in that they are incentivised to manufacture linecards that are not
   power hungry even if the number of packets flowing through them is
   high and thus the Bandwidth Available is also reasonably on the
   higher side compared to other routers. 

   For those manufacturers who set a high power value for even minimal
   traffic, the vendors that dont would win out in the end. 

2.1.1 Advertising Available POWER 

   Please see section 2.4.1 for more information on why Available POWER
 


Shankar Raman et.al,     Expires September 2013                 [Page 8]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   plays a crucial role in determining the choice of routers based on
   the Power metric.  

2.2 ECMP links

   It is possible that multiple links would have the same PWRLINK metric
   after a computation cycle. In such a case load-balancing techniques
   can be used to keep the ECMP links in a steady state with respect to
   each other. Depending on the Available Bandwidth thereafter it is
   possible that the ECMP links may no longer be Equal cost but UCMP or
   Unequal Cost Paths.

2.3 Dampening the side effects of constant change

   It is recommended in this draft that the implementation of the
   proposal be adaptive, infrequent in computation to the extent
   possible without sacrificing adapting to the dynamism and also reduce
   any frequent oscillations. The actual methods to adopt for this
   computation are outside the scope of this document.

2.4 Calculating power shortest paths in an Area


   Assume the following topology where A,B,C etc.. are routers and
   corresponding labelled edges with weights are the links. These
   weights are the current values of the PWRLINK attribute that has been
   flooded in the LSAs through the Area concerned. Assume B is the ABR
   for Area 1 and the routers A and C are the Area 0 core routers. The
   rest of the routers are assumed to be in Area 1. Once the power
   topology of the Area 1 has been calculated as shown below with the
   PWRLINK attributes being assigned to the links, Constrained shortest
   path can be run from the ABR to any of the other routers say H, E , X
   etc.. The CSPF algorithm takes the constraint in terms of the PWRLINK
   attributes along with other attributes to construct a power shortest
   path from say router B to other routers in Area 1.

                       0.5
             (C)  +----------------+
           0.5|  /                 |
              | /                  |
         0.05 V/ 0.1   0.03   0.2  V
      (A)--->(B)--->(D)--->(G)--->(H)
              |             |      |
              |          0.5|      | 0.1
              |             V      V
              +----------->(E)--->(X)
                 0.5           0.3

 


Shankar Raman et.al,     Expires September 2013                 [Page 9]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   Once the path has been computed it is possible to use RSVP-TE to
   construct the power shortest path with the TE-LSP being instantiated
   with the labels appropriately placed in the routers on the power
   shortest path. In this topology, assume one would want to construct a
   path from B to X then the dotted path shows the path constructed and
   to be used by a set of flows or streams of packets belonging to
   multiple flows as seen fit by the router B. If the PWRLINK metrics
   change after due course of time then another power shortest path that
   possibly traverses the same path (if the SUM of PWRLINKs doesnt
   exceed any other path's metrics' SUM) or some other path would be
   constructed. Specifically this method makes use of traffic-
   engineering signalling protocols as the method to place the streams
   from point X to point Y (where X and Y are routers).

                       0.5
             (C)  +----------------+
           0.5|  /                 |
              | /                  |
         0.05 V/ 0.1   0.03   0.2  V
      (A)--->(B)...>(D)...>(G)...>(H)
              |             |      .
              |          0.5|      . 0.1
              |             V      V
              +----------->(E)--->(X)
                 0.5           0.3

2.4.1 Power profiles of Routers and Switches

   It has been experimented and from several sources found that there
   exist routers which have different power profiles. The power profile
   of a router is the curve of power consumption to available bandwidth.
   Mentioned below are a few of these prominent ones that have to be
   taken into consideration.

   The first profile that we will consider is the flattening curve. The
   power consumed to available bandwidth curve takes the shape of a
   steep one initially and then tapers off to a plateau. The point at
   which it begins to give a delta-C (delta in Power Consumed) to delta-
   B (Available Bandwidth exhausted) is the inflection point that tapers
   off to a plateau. Here the delta-C/delta-B begins to slow down or
   decrease rapidly. The more the traffic that is added onto the device
   the lesser it draws power.






 


Shankar Raman et.al,     Expires September 2013                [Page 10]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


        ^
        |
      P |                             .
      o |                       .
      w |                   .
      e |               .
      r |           .
        |        .
      c |      .
      o |     .
      n |    .
      s |   .
      u.|  .
      ------------------------------------>
        |    Available Bandwidth exhausted


   The second profile that we will consider is the exponential curve.
   The power consumed to available bandwidth curve takes the shape of an
   ever increasing steep curve as shown below. Here the delta-C/delta-B
   begins to increase as more traffic is thrown onto it as the Available
   bandwidth exhausted increases. This power curve beyond a point is
   intolerable with respect to power guzzling.

        ^
        |
      P |                             .
      o |                            .
      w |                           .
      e |                          .
      r |                        .
        |                      .
      c |                    .
      o |                 .
      n |             .
      s |        .
      u.|  .
      ------------------------------------>
        |    Available Bandwidth exhausted









 


Shankar Raman et.al,     Expires September 2013                [Page 11]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   The third profile that we will consider is a linear curve. In other
   words just a straight line. Here delta-C/delta-B is a constant.

        ^
        |
      P |                      .
      o |                    .
      w |                  .
      e |                .
      r |              .
        |            .
      c |          .
      o |        .
      n |      .
      s |    .
      u.|  .
      ------------------------------------>
        |    Available Bandwidth exhausted

   2.4.1.1 Concave and Convex power curves

   Given that there are 3 kinds of major profiles in the router power
   consumption, what line would we like to pick. This is an important
   point when choosing the metric to pick the low power paths. 

   (a) If the confrontation is between 2 first profile routers the lower
   of the 2 would be considered as shown below. The lower curve offers
   better power savings for each GB of bandwidth transported.

        ^
        |
      P |                             .
      o |                       .
      w |                   .          .
      e |               .       .
      r |           .      .
        |        .     .
      c |      .    .
      o |     .  .
      n |    . .
      s |   . .
      u.|  .
      ------------------------------------>
        |    Available Bandwidth exhausted




 


Shankar Raman et.al,     Expires September 2013                [Page 12]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   (b) If the confrontation is between 2 second profile routers the
   upper curve offers more power savings per GB of bandwidth.

        ^
        |
      P |                          .  .
      o |                         .  .
      w |                       .   .
      e |                     .    .
      r |                   .    .
        |                .     .
      c |             .      .
      o |         .       .
      n |   .         .
      s |        .
      u.|  .
      ------------------------------------>
        |    Available Bandwidth exhausted

   (c) When the confrontation is between a first profile curve and a
   second profile curve, it would be optimal to pick (as shown below)
   the lower of the curves because it gives us lesser power consumed for
   every GB of traffic routed / switched. Here the exponential curve is
   the one that offers lesser amount of power consumed per GB of traffic
   is chosen. But when it gets to a point that the two curves intersect
   it would be more optimal to pick the tapering curve. Thus at the
   meeting point of the 2 curves the exponential curve becomes more
   costly and the tapering one gives us more GB for the power buck. Thus
   this switchover from one curve to the other (in other words from the
   exponential curve to the tapering one) does the trick in terms of
   finding an optimal solution.

















 


Shankar Raman et.al,     Expires September 2013                [Page 13]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


        ^                         .
        |                        .
      P |                       .      .
      o |                     (*)
      w |                   . .   
      e |               .    .
      r |           .      .
        |        .       .
      c |      .       .
      o |     .     .
      n |    .   .
      s |   . .
      u.| ..
      ------------------------------------>
        |    Available Bandwidth exhausted
      (*) Metric switchover point from Consumed Power to Available
   Power.

2.4.1.3 Need to advertise both available power and consumed power

   Thus the above sections have shown that both the available power and
   the consumed power MUST be advertised so that case (c) can be
   deciphered and the switchover of the curves be done and the
   appropriate router be chosen for the rest of the bandwidth to be
   switched over to.

   Thus there will exist Consumed-Power to Available Bandwidth ratio and
   the Available Power to Available Bandwidth ratio. Both the ratios are
   computed and the lower value chosen. The Available Power can be
   judged from the calibration process such as the one carried out by
   independent test organizations as in [12]. An example of their
   calibration is referred to in [12].

   Here given below is the formula for calculating the Available Power
   to Available Bandwidth ratio also called the Available POWER metric.

   Available
   POWER metric    = Available Power consumed per XG
                      (normalized bandwidth) port
   for a given       ----------------------------------
   Port on a LC      Available Bandwidth on that port

2.4.2 Power to Available Bandwidth ratio in a TLV

   As per [RFC3630] the Link TLV can be used to carry this power to
   available Bandwidth ratio with an additional sub-TLV of the link TLV.
   The sub-type number 11 is recommended to be defined for this purpose.

 


Shankar Raman et.al,     Expires September 2013                [Page 14]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   [RFC 3630] states in section 2.2.1 and we QUOTE ...

   2.2.1  Link TLV

   The Link TLV describes a single link.  It is constructed of a set of
   sub-TLVs.  There are no ordering requirements for the sub-TLVs.

   Only one Link TLV shall be carried in each LSA, allowing for fine
   granularity changes in topology.

   The Link TLV is type 2, and the length is variable.

   The following sub-TLVs of the Link TLV are defined:

            1 - Link type (1 octet)
            2 - Link ID (4 octets)
            3 - Local interface IP address (4 octets)
            4 - Remote interface IP address (4 octets)
            5 - Traffic engineering metric (4 octets)
            6 - Maximum bandwidth (4 octets)
            7 - Maximum reservable bandwidth (4 octets)
            8 - Unreserved bandwidth (32 octets)
            9 - Administrative group (4 octets)
           10 - Power-to-Multicast-replication-capacity (4 octets)
           11 - Consumed-Power-to-Available-Bandwidth (4 octets)
           12 - Available-Power-to-Available-Bandwidth (4 octets)

   This memo defines sub-Types 1 through 9.  See the IANA Considerations
   in [RFC3630] section for allocation of new sub-Types.

   The Link Type and Link ID sub-TLVs are mandatory, i.e., must appear
   exactly once.  All other sub-TLVs defined here may occur at most
   once.  These restrictions need not apply to future sub-TLVs.
   Unrecognized sub-TLVs are ignored.

   Various values below use the (32 bit) IEEE Floating Point format. For
   quick reference, this format is as follows:

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |S|    Exponent   |                  Fraction                   |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   S is the sign, Exponent is the exponent base 2 in "excess 127"
   notation, and Fraction is the mantissa - 1, with an implied binary
   point in front of it.  Thus, the above represents the value:

 


Shankar Raman et.al,     Expires September 2013                [Page 15]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


      (-1)**(S) * 2**(Exponent-127) * (1 + Fraction)

   It is proposed that we use the Power-to-Available-Bandwidth ratio as
   a 32 bit IEEE floating Point format field for the purpose of this
   document.

3. Conclusion 

   Routers may have step levels in which they increase power consumption
   when they additively are loaded with more large bandwidth consuming
   multicast or unicast streams. Calibrating these levels may be useful
   for implementing this scheme. It is possible that such calibrated
   thresholds can be used for advertising the PWRLINK ratios in the OSPF
   LSA advertisements. This would be useful for bringing down the
   frequency of updates or advertisements from a line-card about its
   PWRLINK ratio. When power consumption meanders within a certain given
   interval these ratios need not be re-advertised even if further
   unicast and/or multicast streams are added to it. The incentive is to
   recognize a linecard that does not drastically change power
   consumption even if large bandwidth streams are added onto it for
   forwarding and thus give it credit for its power optimal functioning.
   If a router tends to consume the highest level of power even when
   carrying low amounts of unicast and multicast streams on its line
   card, it would automatically have a poor ratio when compared to a
   router that efficiently uses power when considering the Available
   Bandwidth being observed. The best case would be a low power
   consuming line-card or a router filled with such line cards that does
   not leave its power interval no matter how much ever capacity is
   sought to be used on it. But that would be an ideal condition but it
   is definitely an idealistic scenario towards which the router
   manufacturers should look at. 

















 


Shankar Raman et.al,     Expires September 2013                [Page 16]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


3  Security Considerations

   <Security considerations text>


4  IANA Considerations

   New requirements are required from IANA for a new type in the Link
   TLV in order to carry the PWRLINK metric as well. This is needed for
   both Consumed Power Ratio and Available Power Ratio.


5  References

5.1  Normative References


5.2  Informative References

              [1] G. Appenzeller, Sizing router buffers, Doctoral
              Thesis, Department of Electrical Engineering, Stanford
              University, 2005.

              [2] A. P. Bianzino, C. Chaudet, D. Rossi and J. L.
              Rougier, A survey of green networking research, IEEE
              Communications and Surveys Tutorials, preprint.

              [3] J. Baliga, K. Hinton and R. S. Tucker, Energy
              consumption of the internet, Proc. of joint international
              conference on optical internet, June 2007, pp. 1-3.

              [4] J. Chabarek, J. Sommers, P. Barford, C. Estan, D.
              Tsiang and S. Wright, Power awareness in network design
              and routing, Proc. of the IEEE INFOCOM 2008, April 2008,
              pp. 457-465.

              [5] B. Venkat et.al, Constructing disjoint and partially
              disjoint InterAS TE-LSPs, USPTO Patent 7751318, Cisco
              Systems, 2010.

              [6] M. Xia et. al., Greening the optical backbone network:
              A traffic engineering approach, IEEE ICC Proceedings, May
              2010, pp. 1-5.

              [7] W. Lu and S. Sahni, Low-power TCAMs for very large
              forwarding tables, IEEE/ACM Transactions on Computer
              Networks, June 2010, vol. 18, no. 3, pp. 948-959.

 


Shankar Raman et.al,     Expires September 2013                [Page 17]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


              [8] B. Zhang, Routing Area Open Meeting, Proceedings of
              the IETF 81, Quebec, Canada, July 2011.

              [9] M.J.S Raman, V.Balaji Venkat, G.Raina, Reducing Power
              consumption using the Border Gateway Protocol, IARIA
              conferences ENERGY 2012.

              [10] A.Cianfrani et al., An OSPF enhancement for energy
              saving in IP Networks, IEEE INFOCOM 2011 Workshop on Green
              Communications and Networking



Authors' Addresses



   Shankar Raman
   Department of Computer Science and Engineering,
   IIT Madras
   Chennai - 600036
   TamilNadu
   India.

   EMail: mjsraman@cse.iitm.ac.in



   Balaji Venkat Venkataswami
   Department of Electrical Engineering
   IIT Madras
   Chennai - 600036
   TamilNadu
   India.

   EMail: balajivenkat299@gmail.com



   Prof.Gaurav Raina
   Department of Electrical Engineering
   IIT Madras
   Chennai - 600036
   TamilNadu
   India.

   EMail: gaurav@ee.iitm.ac.in

 


Shankar Raman et.al,     Expires September 2013                [Page 18]

INTERNET DRAFT Power based Topologies using CSPF in OSPF      March 2013


   Vasan Srini
   Department of Computer Science and Engineering
   IIT Madras
   Chennai - 600036
   TamilNadu
   India.

   EMail: vasan.vs@gmail.com











































Shankar Raman et.al,     Expires September 2013                [Page 19]