draft-tls-client-puzzles.xml

<?xml version="1.0" encoding="UTF-8"?>

<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
<!ENTITY RFC2119 SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml">
<!ENTITY RFC5754 SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.5754.xml">
<!ENTITY RFC8019 SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.8019.xml">
<!ENTITY RFC8446 SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.8446.xml">
<!ENTITY RFC8701 SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.8701.xml">
<!ENTITY NYGRENCLIENTPUZZLES SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml3/reference.I-D.nygren-tls-client-puzzles.xml">
<!ENTITY NIRCLIENTPUZZLES SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc/bibxml3/reference.I-D.nir-tls-puzzles.xml">
]>

<?rfc rfcedstyle="yes"?>
<?rfc toc="yes"?>
<?rfc tocindent="yes"?>
<?rfc sortrefs="yes"?>
<?rfc symrefs="yes"?>
<?rfc strict="yes"?>
<?rfc comments="yes"?>
<?rfc compact="yes"?>
<?rfc inline="yes"?>
<?rfc text-list-symbols="-o*+"?>

<rfc docName="draft-venhoek-tls-client-puzzles-00" category="std" ipr="trust200902">

  <front>
    <title abbrev="Client Puzzles">TLS Client Puzzles Extension</title>

    <author initials="D." surname="Venhoek" fullname="David Venhoek">
      <organization>Tweede Golf B.V.</organization>
      <address>
        <email>david@venhoek.nl</email>
      </address>
    </author>
    <author initials="W." surname="Bokslag" fullname="Wouter Bokslag">
      <organization>Midnight Blue</organization>
      <address>
        <email>w.bokslag@midnightblue.nl</email>
      </address>
    </author>
    <author initials="M." surname="Schoolderman" fullname="Marc Schoolderman">
      <organization>Tweede Golf B.V.</organization>
      <address>
        <email>marc@tweedegolf.com</email>
      </address>
    </author>
    <author initials="E." surname="Nygren" fullname="Erik Nygren">
      <organization>Akamai Technologies</organization>
      <address>
        <email>erik+ietf@nygren.org</email>
        <uri>http://erik.nygren.org/</uri>
      </address>
    </author>
    <author initials="S." surname="Erb" fullname="Samuel Erb">
      <organization>Akamai Technologies</organization>
      <address>
        <email>serb@akamai.com</email>
      </address>
    </author>
    <author initials="A." surname="Biryukov" fullname="Alex Biryukov">
      <organization>University of Luxembourg</organization>
      <address>
        <email>alex.biryukov@uni.lu</email>
      </address>
    </author>
    <author initials="D." surname="Khovratovich" fullname="Dmitry Khovratovich">
      <organization>University of Luxembourg</organization>
      <address>
        <email>khovratovich@gmail.com</email>
      </address>
    </author>
	  <author initials="A." surname="Juels" fullname="Ari Juels">
      <organization>Cornell Tech and Cornell University</organization>
      <address>
        <email>juels@cornell.edu</email>
      </address>
    </author>

    <date year="2024" month="September" day="12"/>

    <area>General</area>

    <workgroup>Transport Layer Security</workgroup>
    
    <keyword>Internet-Draft</keyword>

    <abstract>


<t>Client puzzles allow a TLS server to defend itself against asymmetric DDoS attacks.
In particular, it allows a server to request clients perform a selected amount
of computation prior to the server performing expensive cryptographic operations.
This allows servers to employ a layered defense that represents an improvement
over pure rate-limiting strategies.</t>

<t>Client puzzles are implemented as an extension to TLS 1.3
<xref target="RFC8446"/> wherein a server can issue a HelloRetryRequest
containing the puzzle as an extension.  The client must then
resend its ClientHello with the puzzle results in the extension.</t>



    </abstract>


  </front>

  <middle>


<section anchor="overview" title="Overview and rationale">

<t>Adversaries can exploit the design of the TLS protocol to craft
powerful asymmetric DDOS attacks.  Once an attacker has opened a TCP
connection, the attacker can transmit effectively static content that
causes the server to perform expensive cryptographic operations. Rate
limiting offers one possible defense against this type of attack;
however, pure rate limiting systems represent an incomplete solution:</t>

<t><list style="numbers">
  <t>Rate limiting systems work best when a small number of bots are
attacking a single server. Rate limiting is much more difficult when
a large number of bots are directing small amounts of traffic to
each member of a large distributed pool of servers.</t>
  <t>Rate limiting systems encounter problems where a mixture of “good”
and “bad” clients are hidden behind a single NAT or Proxy IP
address and thus are all stuck being treated on equal footing.</t>
  <t>Rate limiting schemes often penalize well-behaved good clients
(which try to complete handshakes and may limit their number of
retries) much more heavily than they penalize attacking bad clients
(which may try to disguise themselves as good clients, but which
otherwise are not constrained to behave in any particular way).</t>
</list></t>

<t>Client puzzles are complementary to rate-limiting and give servers
another option than just rejecting some fraction of requests. A server
can provide a puzzle (of varying and server-selected complexity) to a
client as part of a HelloRetryRequest extension. The client must
choose to either abandon the connection or solve the puzzle and resend
its ClientHello with a solution to the puzzle. Puzzles are designed to
have asymmetric complexity such that it is much cheaper for the server to
generate and validate puzzles than it is for clients to solve them.</t>

<t>Client puzzle systems may be inherently “unfair” to clients that run 
with limited resources (such as mobile devices with batteries and slow CPUs).
However, client puzzle schemes will typically only be evoked when a server is under
attack and would otherwise be rejecting some fraction of requests.
The overwhelming majority of transactions will never involve a
client puzzle.  Indeed, if client puzzles are successful in forcing
adversaries to use a new attack vector, the presence of client puzzles
will be completely transparent to end users.</t>

<t>It is likely that not all clients will choose to support this
extension.  During attack scenarios, servers will still have the
option to apply traditional rate limiting schemes (perhaps with
different parameters) to clients not supporting this extension or
using a version of TLS prior to 1.3.</t>

</section>
<section anchor="notational-conventions" title="Notational Conventions">

<t>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
<xref target="RFC2119"/>.</t>

<t>Messages are formatted with the notation as described within <xref target="RFC8446"/>.</t>

</section>
<section anchor="handshake-changes" title="Handshake Changes">

<t>Client puzzles are implemented as a new ClientPuzzleExtension to TLS
1.3 <xref target="RFC8446"/>.  A client supporting the
ClientPuzzleExtension MUST indicate support by sending a
ClientPuzzleExtension along with their ClientHello containing a list
of puzzle types supported, but with no puzzle response.  When a server
wishes to force the client to solve a puzzle, it MAY send a
HelloRetryRequest with a ClientPuzzleExtension containing a puzzle of
a supported puzzle type and with associated parameters.  To continue
with the handshake, a client MUST resend their ClientHello with a
ClientPuzzleExtension containing a response to the puzzle.
The ClientHello must otherwise be identical to the initial ClientHello,
other than for attributes that are defined by specification to 
not be identical.</t>

<t>If a puzzle would consume too many resources, a client MAY choose to
abort the handshake with the new fatal alert “puzzle_too_hard” and terminate
the connection.</t>

<t>A typical handshake when a puzzle is issued will look like:</t>

<figure><artwork><![CDATA[
   Client                                               Server

   ClientHello              
     + ClientPuzzleExtension
     + ClientKeyShare        -------->
                             <--------       HelloRetryRequest
                                       + ClientPuzzleExtension
   ClientHello
     + ClientPuzzleExtension
     + ClientKeyShare        -------->
                                                   ServerHello
                                                ServerKeyShare
                                        {EncryptedExtensions*}
                                        {ServerConfiguration*}
                                                {Certificate*}
                                         {CertificateRequest*}
                                          {CertificateVerify*}
                             <--------              {Finished}
   {Certificate*}
   {CertificateVerify*}
   {Finished}                -------->
   [Application Data]        <------->     [Application Data]
]]></artwork></figure>

<t>Figure 1.  Message flow for a handshake with a client puzzle</t>

<t>* Indicates optional or situation-dependent messages that are not always sent.</t>

<t>{} Indicates messages protected using keys derived from the ephemeral secret.</t>

<t>[] Indicates messages protected using keys derived from the master secret.</t>

<t>Note in particular that the major cryptographic operations (starting
to use the ephemeral secret and generating the CertificateVerify) are
performed <spanx style="emph">after</spanx> the server has received and validated the
ClientPuzzleExtension response from the client.</t>

<section anchor="the-clientpuzzleextension-message" title="The ClientPuzzleExtension Message">

<t>The ClientPuzzleExtension message contains the communication necessary
for the client puzzle mechanims to work. It is used for three purposes:
<list style = "numbered">
  <t>
    In a ClientHello to indicate which puzzles are supported.
  </t>
  <t>
    In a HelloRetryRequest to provide the client with a specific puzzle to solve.
  </t>
  <t>
    In a retried ClientHello to provide the server with the solution to the puzzle.
  </t>
</list>
</t>

<figure><artwork><![CDATA[
      struct {
          ClientPuzzleType type<2..254>;
          opaque client_puzzle_challenge_response<0..2^16-1>;
      } ClientPuzzleExtension;

      enum {
         echo (0),
         sha256_cpu (1),
         sha512_cpu (2),
         birthday_puzzle (3),
         (0xFFFF)
      } ClientPuzzleType;
]]></artwork></figure>

<t>The interpretation of the fields of the ClientPuzzleExtension depends on
the purpose for which the extension field is sent. In a ClientHello, they should follow:
<list style="hanging">
  <t hangText="type:">
  When the extension is used to convey supported puzzles, this MUST contain a list of all
  puzzle types supported by the implementation. When used to transport the solution to a
  puzzle, this field MUST contain a single puzzle type, and MUST contain the type of
  the puzzle for which a solution is in the "client_puzzle_challenge_response" field.
  </t>
  <t hangText="client_puzzle_challenge_response">
  When the extension is used to convey supported puzzles, this MUST be empty. When used to
  transport a puzzle solution, it MUST contain a solution to the puzzle type indicated in the
  "type" field.
  </t>
</list>
If the server provides a HelloRetryRequest, they should follow:
<list style="hanging">
  <t hangText="type:">
  The type field MUST contain the type of the puzzle provided to the client. In HelloRetryRequest
  messages, it MUST always contain a single puzzle type.
  </t>
  <t hangText="client_puzzle_challenge_response">
  This field MUST contain a puzzle of the type indicated by the "type" field.
  </t>
</list>
Note that a server MAY choose not to provide a puzzle in a HelloRetryRequest, even if it supports
the client puzzle extension.
</t>

<t>Clients supporting the ClientPuzzleExtension MUST send a ClientPuzzleExtension listing all supported
puzzles in their initial ClientHello message. Unless provided with a puzzle, they MUST also repeat
this extension in any retried ClientHello.
</t>

<t>
A supporting server MAY send a client indicating support for the ClientPuzzleExtension a puzzle of a
type the client indicated support for. A server MUST ignore puzzle types send by the client for which it has no support.
A server MUST NOT send a puzzle for a type not supported by the client according to its initial ClientHello.
A server MAY abort a connection with clients not supporting the ClientPuzzleExtension or if there is no overlap
between the puzzle types acceptable to the server and those supported by the client.
The server MAY send a "handshake_failure" alert in such cases.
</t>

<t>
A client receiving a puzzle from the server MAY abort the connection if the puzzle difficulty is perceived to be
too high. A client MAY send a "handshake_failure" alert in such cases. If a client sends a retried ClientHello after
receiving a client puzzle from the server, it MUST provide a ClientPuzzleExtension containing a solution to that puzzle.
A server receiving a retried ClientHello without a valid solution after providing a puzzle MUST abort the connection, optionally sending
a "missing_extension" alert.
</t>

<t>
A client that receives a puzzle of a type it does not support MUST abort the connection. It MAY send an "illegal_parameter" alert
in such cases. A server that receives a puzzle solution of a type it never gave a challenge for MUST abort the connection. It MAY
send an "illegal_parameter" alert in such cases.
</t>
</section>
</section>
<section anchor="usage-by-servers" title="Usage by Servers">

<t>Servers MAY send puzzles to clients when under duress, and the 
percentage of clients receiving puzzles and the complexity
of the puzzles both MAY be selected as a function of the degree of duress.</t>

<t>Servers MAY use additional factors, such as client IP reputation
information, to determine when to send a puzzle as well as the
complexity.</t>

</section>

<section anchor="grease" title="GREASE">
<t>
To exercise both the solvers in clients, as well as the mechanisms for adding new
puzzle types, we add a greasing mechanism similar to that proposed in <xref target="RFC8701"/>.
</t>

<t>
The following puzzle types are reserved as GREASE values for the puzzle type field:
<list>
<t>0x0A0A</t>
<t>0x1A1A</t>
<t>0x2A2A</t>
<t>0x3A3A</t>
<t>0x4A4A</t>
<t>0x5A5A</t>
<t>0x6A6A</t>
<t>0x7A7A</t>
<t>0x8A8A</t>
<t>0x9A9A</t>
<t>0xAAAA</t>
<t>0xBABA</t>
<t>0xCACA</t>
<t>0xDADA</t>
<t>0xEAEA</t>
<t>0xFAFA</t>
</list>
</t>

<section anchor="grease-client" title="Client Behavior">
<t>
When indicating support for client puzzles, a client MAY behave as follows:
<list style="symbols">
<t>A client MAY select one or more of the GREASE puzzle types and advertise them in the ClientPuzzleExtension as supported.</t>
</list>
</t>

<t>
A client MUST abort the connection if it receives a GREASE type challenge. It MAY send an "illegal_parameter" alert
in such cases. Note that this can be implemented without special processing, as this matches the normal
behavior for unsupported puzzle types.
</t>
</section>

<section anchor="grease-server" title="Server Behavior">
<t>
When responding to a ClientHello, a server MAY behave as follows:
<list style="symbols">
<t>A server MAY choose to request the client to solve a puzzle even when not under duress. A server SHOULD choose
low difficulty for such puzzles so as to not unnecessarily burden clients.</t>
</list>
</t>

<t>
A server MUST NOT send puzzles for a GREASE puzzle type. A server MUST
treat received GREASE puzzle types as unsupported, and ignore them. Note
that this can be implemented without special processing on the server, as
this matches the normal behavior for unsupported puzzle types.
</t>
</section>

</section>

<section anchor="puzzles" title="Proposed Client Puzzles">

<t>Having multiple client puzzle types allows good clients a choice to
implement puzzles that match with their hardware capabilities
(although this also applies to bad clients).  It also allows
“broken” puzzles to be phased out and retired, such as when
cryptographic weaknesses are identified.</t>

<section anchor="echo-client-puzzle-type" title="Echo Client Puzzle Type">

<t>The echo ClientPuzzleType is intended to be trivial.
The client_puzzle_challenge_response data field is defined to be
a token that the client must echo back.</t>

<t>During an initial ClientHello, this MUST be empty (zero-length).
During HelloRetryRequest, the server MAY send a cookie challenge of
zero or more bytes as client_puzzle_challenge_response .  During the
retried ClientHello, the client MUST respond by resending the
identical cookie sent in the HelloRetryRequest.</t>

</section>
<section anchor="sha-256-cpu-puzzle-type" title="SHA-256 CPU Puzzle Type">

<t>This puzzle forces the client to calculate a SHA-256 <xref target="RFC5754"/> multiple times.
In particular, the server selects a difficulty and a random salt. The client solves
the puzzle by finding any nonce where a SHA-256 hash across the nonce, the salt and
a label contains difficulty leading zero bits.</t>

<figure><artwork><![CDATA[
      struct {
          uint16 difficulty;
          uint8 salt<0..2^16-1>;
      } SHA256CPUPuzzleChallenge;

      struct {
          uint64 challenge_solution;
      } SHA256CPUPuzzleResponse;
]]></artwork></figure>

<t><list style="hanging">
  <t hangText='difficulty'>
  filter affecting the time to find solution.</t>
  <t hangText='salt'>
  A server selected variable-length bytestring.</t>
  <t hangText='challenge_solution'>
  The solution response to the puzzle, as solved by the client.</t>
</list></t>

<t>To find the response, the client must find a numeric value of challenge_solution
where:</t>

<t>SHA-256(challenge_solution || salt || label) contains difficulty leading zeros.</t>

<t>where “||” denotes concatenation and where label is the NUL-terminated
value “TLS SHA256CPUPuzzle” (including the NUL terminator).</t>

<t>Clients offering to support this puzzle type SHOULD support
a difficulty value of at least 18.  [[TODO: is this a good
value?  https://en.bitcoin.it/wiki/Non-specialized_hardware_comparison has a comparison of SHA256 on various hardware.]]</t>

</section>
<section anchor="sha-512-cpu-puzzle-type" title="SHA-512 CPU Puzzle Type">

<t>The SHA-512 CPU Puzzle Type is identical to the
“SHA256 CPU Puzzle Type” except that the SHA-512 <xref target="RFC5754"/> hash function
is used instead of SHA-256.  The label used is the value “TLS SHA512CPUPuzzle”.</t>

<t>Clients offering to support this puzzle type SHOULD support difficulty
values of at least 17.  [[TODO: is this a good value?]]</t>

</section>
<section anchor="equihash-memory-hard-generalized-birthday-problem-puzzle-type" title="Equihash: Memory-hard Generalized Birthday Problem Puzzle Type">

<t>Using Equihash, the asymmetric memory-hard generalized birthday problem PoW <xref target="NDSS2016"/>,
this puzzle will force a client to use a significant amount of memory to solve. The
solution to this puzzle can be trivially verified.</t>

<figure><artwork><![CDATA[
      struct {
         uint16 n;
         uint16 k;
         uint16 difficulty;
         uint8 salt<0..2^16-1>;
     } BirthdayPuzzleChallenge;

     struct {
         uint8 V<20>;
         uint8 solution<0..2^16-1>;
     } BirthdayPuzzleResponse;
]]></artwork></figure>

<t><list style="hanging">
  <t hangText='salt'>
  A server selected variable-length bytestring.</t>
  <t hangText='n, k'>
  parameters affecting the complexity of Wagner’s algorithm.</t>
  <t hangText='difficulty'>
  secondary filter affecting the time to find solution.</t>
  <t hangText='V'>
  20 byte nonce used in solution.</t>
  <t hangText='solution'>
  list of 2^k (n/(k+1)+1)-bit nonces used in solution, referred to as xi below.</t>
</list></t>

<t>In the further text, the output of blake2b is treated as a 512-bit register with most significant bits coming from the last bytes of blake2b output (i.e. little-endian conversion).</t>

<t>To compute the response, the client must find a V and 2^k solutions such that:</t>

<t>blake2b(salt||V||x1) XOR blake2b(salt||V||x2) XOR … XOR blake2b(I||V||x(2^k)) = 0
blake2b(label||salt||V||x1||x2||…||x(2^k)) has difficulty leading zero bits.</t>

<t>where “||” denotes concatenation and where label is the NUL-terminated
value “TLS BirthdayPuzzle” (including the NUL terminator). Incomplete bytes in nonces xi are padded with zero bits, which occupy the
most significant bits.</t>

<t>The client MUST provide the solution list in an order that allows a server to
verify the solution was created using Wagner’s algorithm:</t>

<t>blake2b(salt||V||x(w<spanx style="emph">2^l+1)) XOR blake2b(salt||V||x(w</spanx>2^l+2)) XOR … XOR blake2b(I||V||x(w*2^l+2^l))
has nl/(k+1) leading zero bits for all w,l.</t>

<t>and two 2^(l-1)(n/(k+1)+1)-bit numbers Z1 and Z2 must satisfy Z1&lt;Z2 where</t>

<t>Z1 = x(w<spanx style="emph">2^l+1)||x(w</spanx>2^l+2)||…||x(w<spanx style="emph">2^l+2^(l-1)) 
Z2 = x(w</spanx>2^l+2^(l-1)+1)||x(w<spanx style="emph">2^l+2)||…||x(w</spanx>2^l+2^l)
 as in(<xref target="NDSS2016"/> section 4A, 5C).
The server MUST verify these intermediate equations.</t>

<t>A solution can be found using Wagner’s algorithm as described in <xref target="NDSS2016"/>. 
The amount of memory required to find a solution  is 2 ^ (n/(k+1)+k) bytes.
A solution  requires (k+1)2^(n/(k+1)+d) calls to the blake2b hash function.</t>

<t>Clients offering to support this puzzle type SHOULD support n, k values such that
2^(n/(k+1)+k) is at least 20MB.</t>

<t>Servers SHOULD look to minimize the value of k as 2^k blake2b hash operations will
be required to verify a solution.</t>

</section>
</section>
<section anchor="iana" title="IANA Considerations">

<t>The IANA will need to assign an extension codepoint value for ClientPuzzleExtension.</t>

<t>The IANA will need to assign an AlertDescription codepoint value for puzzle_too_hard.</t>

<t>The IANA will also need to maintain a registry of client puzzle types.</t>

</section>
<section anchor="security" title="Security Considerations">

<t>A hostile server could cause a client to consume unbounded resources.
Clients MUST bound the amount of resources (cpu/time and memory) they
will spend on a puzzle.</t>

<t>A puzzle type with economic utility could be abused by servers,
resulting in unnecessary resource usage by clients.  In the worst
case, this could open up a new class of attacks where clients might be
directed to malicious servers to get delegated work.  As such, any new
puzzle types SHOULD NOT be ones with utility for other purposes (such
as mining cryptocurrency or cracking password hashes).  Including
fixed labels in new puzzle definitions may help mitigate this risk.</t>

<t>Depeding on the structure of the puzzles, it is possible that an
attacker could send innocent clients to a hostile server and then use
those clients to solve puzzles presented by another target server that
the attacker wishes to attack.  There may be ways to defend against
this by including IP information in the puzzles (not currently proposed
in this draft), although that introduces additional issues.</t>

<t>All extensions add complexity, which could expose additional attack
surfaces on the client or the server.  Using cryptographic primitives
and patterns already in-use in TLS can help reduce (but certainly not
eliminate) this complexity.</t>

<t>An attacker that can force a server into client puzzle mode could
result in a denial of service to clients not supporting puzzles or not
having the resources to complete the puzzles.  This is not necessarily
worse than if the server was overloaded and forced to deny service to
all clients or to a random selection of clients.  By using client
puzzles, clients willing to rate-limit themselves to the rate at which
they can solve puzzles should still be able to obtain service
while the server is able to stay available for these clients.</t>

<t>It is inevitable that attackers will build hardware optimized to solve
particular puzzles.  Using common cryptographic primitives (such as
SHA-256) also means that commonly deployed clients may have hardware
assistance, although this also benefits legitimate clients.</t>

</section>
<section anchor="privacy" title="Privacy Considerations">

<t>Measuring the response time of clients to puzzles gives an indication
of the relative capabilities of clients.  This could be used as an
input for client fingerprinting.</t>

<t>Client’s support for this extension, as well as which puzzles they
support, could also be used as an input for client fingerprinting.</t>

</section>
<section anchor="acknowledgments" title="Acknowledgments">

<t>The story of client puzzles dates back to Dwork and Naor <xref target="DN92"/>
and Juels and Brainard <xref target="JB99"/>. This draft was in large part based
on the 2016 draft by Nygren et. al. <xref target="I-D.nygren-tls-client-puzzles"/>,
which in turn was partially inspired by work done by Kyle Rose in 2001, as well
as a 2001 paper by Drew Dean (Xerox PARC) and Adam Stubblefield (Rice)
<xref target="SEC2001.DEAN"/>, as well as being shaped by discussions with Eric Rescorla, 
Yoav Nir, Richard Willey, Rich Salz, Kyle Rose, Brian Sniffen, and others on
the TLS working group. An alternate approach was proposed in
<xref target="I-D.nir-tls-puzzles"/>.  Some similar mechanisms for protecting IKE
are discused in <xref target="RFC8019"/>.</t>

</section>


  </middle>

  <back>

    <references title='Normative References'>

&RFC2119;

&RFC5754;

&RFC8446;

    </references>

    <references title='Informative References'>

<reference anchor="SEC2001.DEAN" target="https://www.usenix.org/legacy/events/sec2001/full_papers/dean/dean.pdf">
  <front>
    <title>Using Client Puzzles to Protect TLS</title>
    <author initials="D." surname="Dean" fullname="Drew Dean">
      <organization>Xerox PARC</organization>
    </author>
    <author initials="A." surname="Stubblefield" fullname="Adam Stubblefield">
      <organization>Rice University</organization>
    </author>
    <date year="2001" month="August" day="11"/>
  </front>
  <seriesInfo name="Proceedings of the 10th USENIX Security Symposium" value=""/>
</reference>
<reference anchor="DN92" target="http://www.wisdom.weizmann.ac.il/~naor/PAPERS/pvp_abs.html">
  <front>
    <title>Pricing via Processing or Combatting Junk Mail</title>
    <author initials="C." surname="Dwork" fullname="Cynthia Dwork">
      <organization>IBM Research</organization>
    </author>
    <author initials="M." surname="Naor" fullname="Moni Naor">
      <organization>Weizmann Institute of Science</organization>
    </author>
    <date year="1992"/>
  </front>
  <seriesInfo name="Proceedings of Crypto'92" value=""/>
</reference>
<reference anchor="JB99" target="http://www.wisdom.weizmann.ac.il/~naor/PAPERS/pvp_abs.html">
  <front>
    <title>Client Puzzles: A Cryptographic Defense Against Connection Depletion Attacks</title>
    <author initials="A." surname="Juels" fullname="Ari Juels">
      <organization>RSA Laboratories</organization>
    </author>
    <author initials="J." surname="Brainard" fullname="John Brainard">
      <organization>RSA Laboratories</organization>
    </author>
    <date year="1999"/>
  </front>
  <seriesInfo name="Proceedings of NDSS'99" value=""/>
</reference>
<reference anchor="NDSS2016" target="https://www.internetsociety.org/sites/default/files/blogs-media/equihash-asymmetric-proof-of-work-based-generalized-birthday-problem.pdf">
  <front>
    <title>Equihash: Asymmetric proof-of-work based on the Generalized Birthday problem</title>
    <author initials="A." surname="Biryukov" fullname="Alex Biryukov">
      <organization>University of Luxembourg</organization>
    </author>
    <author initials="D." surname="Khovratovich" fullname="Dmitry Khovratovich">
      <organization>University of Luxembourg</organization>
    </author>
    <date year="2016" month="February" day="25"/>
  </front>
</reference>

&NIRCLIENTPUZZLES;

&NYGRENCLIENTPUZZLES;

&RFC8019;

&RFC8701;

    </references>



  </back>
</rfc>