mpc chapter: WIP
This commit is contained in:
jaseg
parent
acb6d674f7
commit
a67057f8ff
2 changed files with 56 additions and 24 deletions
|
|
@ -166,11 +166,10 @@ evaluations of a pseudorandom function such as a cryptographic hash or a cipher.
|
|||
Garbled Circuit is many times slower than performing it in the clear. Intuitively, each single-bit gate in the garbled
|
||||
circuit results in several cryptographic operations with input and output sizes of dozens or hundreds of bits.
|
||||
Practically useful functions such as AES encryption have circuit implementations measuring thousands or tens of
|
||||
thousands of gates, meaning these costs quickly escalate for practical problem sizes.
|
||||
\cite{
|
||||
thousands of gates, meaning these costs quickly escalate for practical problem sizes~\cite{
|
||||
boyarNewCombinationalLogic2010,
|
||||
songhoriTinyGarbleHighlyCompressed2015,
|
||||
}
|
||||
}.
|
||||
|
||||
% FIXME This entire connecting section
|
||||
|
||||
|
|
@ -194,14 +193,13 @@ thousands of gates, meaning these costs quickly escalate for practical problem s
|
|||
|
||||
Multiparty Computation is at the verge of being practical in some applications, but is still too computationally
|
||||
expensive for others. While some attempts at GPU-accelerating MPC primitives exist, in practice it is commonly
|
||||
implemented using CPU processing. From the performance numbers given above\todo{Give performance numbers above} we can
|
||||
see that a single, modern server-class CPU is sufficient for an useful amount of computation in MPC.
|
||||
implemented using CPU processing. The technology comes with an unavoidable increase in computational complexity since
|
||||
each single plaintext computation or gate results in several cryptographic operations.
|
||||
|
||||
A naive implementation might attempt to implement MPC using an HSM by simply offloading all cryptographic operations to
|
||||
the HSM. In practice, this is not a workable solution due to the slow processing speed of conventional HSMs.
|
||||
Conventional HSMs commonly use smartphone-class SoCs, which lag behind server CPUs in processing speed by several orders
|
||||
of magnitude.
|
||||
\todo{Cite some HSM/MPC papers here.}
|
||||
Conventional HSMs use low-power embedded processors since their encapsulation using potting and security meshes impedes
|
||||
heat transfer, limiting power dissipation.
|
||||
|
||||
In the near term, absent radical developments in either MPC theory or in the speed and power efficiency of processing
|
||||
hardware, the only feasible solution for HSM-protected MPC at any practical scale is to find a way to protect an entire
|
||||
|
|
@ -214,20 +212,21 @@ As a baseline performance target, we consider a commodity server mainboard in CE
|
|||
high-end server CPU and a large amount of RAM. As MPC systems do not usually require a great amount of storage, we can
|
||||
largely ignore storage for our size and power calculations.\todo{Refer to performance numbers from research above here}
|
||||
|
||||
As a result, we end up with a total maximum power dissipation of approximately \todo{Calculate, make table} XXX as shown
|
||||
in Table~\ref{tab_power_budget}. Dissipating this amount of power using air cooling is within the capabilities of
|
||||
commodity server cooling components\todo{Refer to datasheets}.
|
||||
As a result, we end up with a total maximum power dissipation of approximately \qty{420}{\watt} as shown in
|
||||
Table~\ref{tab_power_budget}. Dissipating this amount of power using air cooling is within the capabilities of commodity
|
||||
server cooling components~\cite{coroamaPossibleFutureTrends2025}.
|
||||
|
||||
\begin{table}
|
||||
\centering
|
||||
\begin{tabular}{r|l|r|r}
|
||||
Count & Component & Maximum Power Dissipation Each & Total\\\hline
|
||||
1 & CPU: & &\\
|
||||
16 & DDR-4 Memory modules: & &\\
|
||||
1 & Mainboard: & &\\
|
||||
1 & Power Supply: & &\\
|
||||
Count & Component & Power Dissipation (approx.) & Total\\\hline
|
||||
1 & CPU & \qty{350}{\watt}~\cite{tropgen16YearsSPEC2024}&\qty{350}{\watt}\\
|
||||
16 & Memory~\cite{kennedyDDR4DIMMsSystem2017} &\qty{2}{\watt}&\qty{32}{\watt}\\
|
||||
1 & Losses & \qty{40}{\watt}&\qty{40}{\watt}\\
|
||||
\end{tabular}
|
||||
\caption{Power budget of a modern mid-range server. Power supply power dissipation calculated at target load.}
|
||||
\caption{Power budget of a modern mid-range server. Losses were estimated at 10\%, consistent with mainboard losses
|
||||
plus losses from a 80plus platinum efficiency certified power supply (~94\% at load).
|
||||
}
|
||||
\label{tab_power_budget}
|
||||
\end{table}
|
||||
|
||||
|
|
@ -289,6 +288,8 @@ envelope diameter of an MPC accelerator application, high-airflow fans must be u
|
|||
speed. We consider this a valid tradeoff since such a system would be deployed in a datacenter where high noise levels
|
||||
are acceptable.
|
||||
|
||||
\todo{Finish sketch!}
|
||||
|
||||
\section{Outlook}
|
||||
|
||||
In this chapter we briefly introduced the challenges raised by MPC at scale, and we outlined a practical solution based
|
||||
|
|
|
|||
45
main.bib
45
main.bib
|
|
@ -1504,6 +1504,16 @@
|
|||
urldate = {2024-09-05}
|
||||
}
|
||||
|
||||
@misc{coroamaPossibleFutureTrends2025,
|
||||
title = {Past and {{Possible Future Trends}}},
|
||||
author = {Coroamă, Vlad C and Dumbrav, Oana and Hinterholzer, Simon and Progni, Kejsi and Hintemann, Ralph},
|
||||
date = {2025-04},
|
||||
url = {https://www.borderstep.de/wp-content/uploads/2025/06/EDNA-EE-of-servers-FINAL.pdf},
|
||||
langid = {english},
|
||||
organization = {International Energy Agency},
|
||||
file = {/home/jaseg/Zotero/storage/IN6J87WC/Coroamă et al. - Past and Possible Future Trends.pdf}
|
||||
}
|
||||
|
||||
@incollection{couteauSilverSilentVOLE2021,
|
||||
title = {Silver: {{Silent VOLE}} and {{Oblivious Transfer}} from {{Hardness}} of {{Decoding Structured LDPC Codes}}},
|
||||
shorttitle = {Silver},
|
||||
|
|
@ -3431,6 +3441,18 @@
|
|||
isbn = {0-07-230092-2}
|
||||
}
|
||||
|
||||
@online{kennedyDDR4DIMMsSystem2017,
|
||||
title = {{{DDR4 DIMMs}} and {{System Power Consumption}} - {{We Tested}}},
|
||||
author = {Kennedy, Patrick},
|
||||
date = {2017-01-30T13:52:03+00:00},
|
||||
url = {https://www.servethehome.com/ddr4-dimms-system-power-consumption-tested/},
|
||||
urldate = {2025-10-27},
|
||||
abstract = {We tested a low power 1U system with 16x and 4x DDR4 DIMMs to see the power consumption impact of using fewer RDIMMs in a server},
|
||||
langid = {american},
|
||||
organization = {ServeTheHome},
|
||||
file = {/home/jaseg/Zotero/storage/2FADBUGE/ddr4-dimms-system-power-consumption-tested.html}
|
||||
}
|
||||
|
||||
@online{KiCadEDA,
|
||||
title = {{{KiCad EDA}}},
|
||||
url = {https://www.kicad.org/},
|
||||
|
|
@ -5744,13 +5766,6 @@
|
|||
file = {/home/jaseg/Zotero/storage/BHBDVJAI/ki_empfehlung_kurz.en.pdf}
|
||||
}
|
||||
|
||||
@online{RecommendationsUniversityState2025,
|
||||
title = {Recommendations of the {{University}} and {{State Library Darmstadt}} for {{Labelling}} and {{Documenting AI-generated Content}}},
|
||||
date = {2025-09-22},
|
||||
url = {https://www.ulb.tu-darmstadt.de/ki-doku},
|
||||
urldate = {2025-10-24}
|
||||
}
|
||||
|
||||
@online{RefusingTechFascism,
|
||||
title = {Refusing {{Tech Fascism}} — {{Error}} 406 {{Tech Fascism Not Acceptable}}},
|
||||
url = {https://error417.expectation.fail/406/tech-fascism-not-acceptable/essay-refusing-tech-fascism-by-tante},
|
||||
|
|
@ -6935,6 +6950,22 @@ Archive 2: https://web.archive.org/web/20250510104017/https://de.linkedin.com/pu
|
|||
x-fetchedfrom = {Google Scholar}
|
||||
}
|
||||
|
||||
@inproceedings{tropgen16YearsSPEC2024,
|
||||
title = {16 {{Years}} of {{SPEC Power}}: {{An Analysis}} of X86 {{Energy Efficiency Trends}}},
|
||||
shorttitle = {16 {{Years}} of {{SPEC Power}}},
|
||||
booktitle = {2024 {{IEEE International Conference}} on {{Cluster Computing Workshops}} ({{CLUSTER Workshops}})},
|
||||
author = {Tröpgen, Hannes and Schöne, Robert and Ilsche, Thomas and Hackenberg, Daniel},
|
||||
date = {2024-09},
|
||||
pages = {76--80},
|
||||
doi = {10.1109/CLUSTERWorkshops61563.2024.00020},
|
||||
url = {https://ieeexplore.ieee.org/abstract/document/10740865},
|
||||
urldate = {2025-10-27},
|
||||
abstract = {The SPEC Power benchmark offers valuable insights into the energy efficiency of server systems, allowing comparisons across various hardware and software configurations. Benchmark results are publicly available for hundreds of systems from different vendors, published since 2007. We leverage this data to perform an analysis of trends in x86 server systems, focusing on power consumption, energy efficiency, energy proportionality and idle power consumption. Through this analysis, we aim to provide a clearer understanding of how server energy efficiency has evolved and the factors influencing these changes.},
|
||||
eventtitle = {2024 {{IEEE International Conference}} on {{Cluster Computing Workshops}} ({{CLUSTER Workshops}})},
|
||||
keywords = {Benchmark testing,Computer architecture,Computer performance,Conferences,Energy efficiency,Focusing,Hardware,High performance computing,Market research,Performance analysis,Power demand,Processor energy efficiency,Servers,Software},
|
||||
file = {/home/jaseg/Zotero/storage/T82VHY3I/10740865.html}
|
||||
}
|
||||
|
||||
@online{tschirsichHackerHinOder0100,
|
||||
title = {"{{Hacker}} Hin Oder Her": {{Die}} Elektronische {{Patientenakte}} Kommt!},
|
||||
shorttitle = {"{{Hacker}} Hin Oder Her"},
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue