mirror of
https://github.com/lux-apps/papers.git
synced 2026-10-30 18:15:21 +00:00
zeekayandHanzo Dev
e2cb62c540
I resolved these files toward a June commit that called the billion-order figures fabricated, and that commit was itself the mistaken one. The July campaign it predates does not assert its numbers, it checks them: a GPU-versus-CPU parity run matching exactly, 38,000,000 fills on both sides at 1,000,000 books, depth 32, 32 takers each, on hardware this estate owns. It also names the error in the earlier framing. Consensus determinism was taken to force a single-thread launch and forfeit the GPU. It does not: determinism constrains ordering WITHIN a book and says nothing across books, so one thread per book — each matching its taker stream in strict price-time priority, bit-for-bit re-executable, millions of threads — saturates the accelerator with every book deterministic. Restored at the later versions across all 27 files: the accounting section, the CLOB matcher, lightspeed-dex, lp-306, lp-308, lp-137, p3q-corona, the precompile engineering sections and the FHE orderbook application. Paper builds, 23 pages, no errors. Co-authored-by: Hanzo Dev <[email protected]>
32 lines
2.1 KiB
TeX
32 lines
2.1 KiB
TeX
\section{Future Work}
|
|
\label{sec:future}
|
|
|
|
The architecture is built and the three axes are measured; the following items are named,
|
|
not yet built, and stated as such.
|
|
|
|
\begin{enumerate}
|
|
\item \textbf{A cheaper matching representation.} A \texttt{Q64.64} fixed-point matching
|
|
path (4\,KB arenas, $64/64$ division) trades precision and ABI width for a far shorter
|
|
critical path and a far smaller working set, on top of the billions-of-orders/sec the
|
|
\texttt{uint256} path already delivers (Section~\ref{subsec:billion-matching}). It is not
|
|
yet built; it carries a precision/ABI tradeoff that must be specified before it can replace
|
|
the \texttt{uint256} path under the same parity gate.
|
|
\item \textbf{A compact arena.} The current per-book arena is large and mostly empty, which
|
|
causes a cache cliff beyond roughly 50--75\,k markets (Section~\ref{sec:gpu-matching}). A
|
|
compact arena is the lever for very high market counts.
|
|
\item \textbf{The production-batch \texttt{starkfri} AIR.} The measured 986 verifies/s is
|
|
for a toy AIR; the production DEX-batch AIR (boundary and transition constraints over many
|
|
trace rows) is scaffolded and not yet measured, and will verify slower
|
|
(Section~\ref{sec:rollup}).
|
|
\item \textbf{Decentralizing the D-Chain.} The D-Chain is a single-operator venue today.
|
|
Decentralizing it to many validators requires committing the matching representation at the
|
|
state layer for cross-architecture agreement and moving finality from P3Q attestation to
|
|
\texttt{starkfri} validity proofs (Sections~\ref{sec:determinism}, \ref{sec:security}).
|
|
\item \textbf{Private order flow at scale.} Homomorphic bootstrapping is orders of magnitude
|
|
below the attestation rate (Section~\ref{sec:rollup}); making private order flow
|
|
competitive with the public path is an accelerator-bound problem on the FHE side.
|
|
\item \textbf{Derivative markets.} Perpetual futures and options settle naturally on the
|
|
CLOB with oracle-fed mark prices and a liquidation engine; these are product surfaces above
|
|
the matching and settlement primitives described here.
|
|
\end{enumerate}
|