Peer-to-peer data message matching

Patent metadata

Field	Value
Publication number	US-12200048-B1
Application number	US-202318238187-A
Country	US
Kind code	B1
Filing date	Aug 25, 2023
Priority date	Aug 25, 2023
Publication date	Jan 14, 2025
Grant date	Jan 14, 2025

Abstract

The disclosed embodiments relate to systems/methods for data message matching in a peer-to-peer exchange system. Within the peer-to-peer exchange system, a dedicated processing instance attempts to match incoming data messages to resting message previously converted to the market data by other dedicated processing instances within the peer-to-peer exchange system. The market data stored within mirrored data structures on the dedicated processing instances, such that each instance may individually and independently attempt to match its respective data messages to all currently resting messages. When the incoming data message at least partially fails to be matched, the dedicated processing instance may determine whether to convert the unmatched portion to a resting message within the market data.

First claim

What is claimed is:

1. A computer-implemented method of data message matching in a peer-to-peer system, the method including:
receiving, by a processor and from a first participant, a current data message at a dedicated processing instance for the first participant, the dedicated processing instance for the first participant including dedicated compute resources for match operations specific to the first participant;
accessing, in a data structure at the dedicated processing instance, a participant-specific exchange relationship mapping;
determining, by the processor and based on the participant-specific exchange relationship mapping, that the first participant has an exchange relationship with one or more relationship participants;
determining, by the processor and based on the participant-specific exchange relationship mapping, that the first participant lacks an exchange relationship with one or more blocked participants;
attempting to match, using a mirrored market datastore of the data structure at the dedicated processing instance and without attempting to match to data messages associated with the one or more blocked participants, the current data message to one or more provider data messages from a first provider and/or one or more previous participant data messages associated with corresponding ones of the one or more relationship participants having an exchange relationship with the first participant, the one or more previous participant data messages converted to market data based on match-failure market data conversion preferences by the corresponding ones of the one or more relationship participants, the mirrored market datastore including global market data mirrored across the peer-to-peer system;
determining, by the processor at the dedicated processing instance, whether the first participant has a match-failure market data conversion preference for the current data message;
at a time the attempt to match fails at least in part and the first participant has a match-failure market data conversion preference for the current data message:
converting, by the processor, the current data message to updated market data; and
globally broadcasting the updated market data from the dedicated processing instance to other dedicated processing instances for other participants in the peer-to-peer system, the other participants in the peer-to-peer system including both the relationship participants and the one or more blocked participants; and

cancelling, at a time the attempt to match fails and the first participant lacks a match-failure market data conversion preference for the current data message, the current data message without distribution to the other dedicated processing instances.

2. The computer-implemented method of claim 1 , further including maintaining a mirrored state of the mirrored market datastore by receiving, at the dedicated processing instance for the first participant, globally broadcast data messages from dedicated processing instances for other participants in the peer-to-peer system.

3. The computer-implemented method of claim 1 , further including receiving a peer-to-peer message by receiving, at the dedicated processing instance for the first participant, a globally broadcast data message from a specific one of the dedicated processing instances for other participants, the globally broadcast data message including the peer-to-peer message, the peer-to-peer message directed from a specific one of the dedicated processing instances for other participants individually to the first participant.

4. The computer-implemented method of claim 1 , wherein the participant-specific exchange relationship mapping excludes relationship details for participant relationships within the peer-to-peer system that do not include the first participant.

5. The computer-implemented method of claim 1 , wherein:
the data structure of the dedicated processing instance for the first participant further includes a visibility profile for the mirrored market datastore;
the method further includes:
receiving, at the dedicated processing instance for the first participant, a request from a client interface of the first participant to view market data within the mirrored market datastore; and
determining, by the processor and based on the visibility profile, to exclude the market data for a matchable message associated with a visibility-control participant, the visibility-control participant having an exchange relationship with the first participant, the visibility-control participant denying visibility of the market data for the matchable message for the first participant.

6. The computer-implemented method of claim 1 , wherein the dedicated processing instance for the first participant is hosted on a compute resource unit, the dedicated compute resources maintained by capping an allowed number of dedicated processing instances hosted on the compute resource unit.

7. The computer-implemented method of claim 6 , wherein the compute resource unit includes:
a cloud server;
a serverless compute reservation;
a virtual machine; and/or
a computer node within a distributed ledger network.

8. The computer-implemented method of claim 1 , wherein:
attempting to match the current data message includes:
first determining, by the processor and based on credit threshold data within the data structure of the dedicated processing instance for the first participant, that fully matching the current data message with an otherwise fully matchable data message results in a credit threshold being exceeded for a credit-threshold participant of the relationship participants; and
second determining, by the processor and based on the first determining, to partially fail to match the current data message.

9. The computer-implemented method of claim 1 , further including:
receiving, by the processor, incoming market data for an unmatched data message from one of the other dedicated processing instances for other participants in the peer-to-peer system; and
pseudonymizing, by the processor, the incoming market data by storing the incoming market data within the mirrored market datastore using a pseudonymous identifier for the one of the other dedicated processing instances for other participants in the peer-to-peer system.

10. The computer-implemented method of claim 1 , further including:
generating, at a time that the attempt to match at least in part succeeds, an exchange request data message for a matched data message matched to the current data message; and
receiving, prior to completion of a match exchange based on the matched data message, chase market data including a request for a competing exchange associated with the matched data message;
accessing, responsive to the chase market data, a chase protocol to determine a backoff priority;
immediately reissuing, at a time when the dedicated processing instance for the first participant has a top priority within the chase protocol, the exchange request data message for the matched data message; and
executing, at a time when the dedicated processing instance for the first participant lacks a top priority within the chase protocol, a backoff delay before attempting to reissue the exchange request data message for the matched data message.

11. The computer-implemented method of claim 10 , wherein:
determining the backoff priority includes determining a priority within a deterministic priority scheme defined within the data structure of the dedicated processing instance for the first participant;
determining the backoff priority includes determining a randomized priority;
executing the backoff delay includes executing a pre-defined delay define

Assignees

• Chicago Mercantile Exchange Inc

Inventors

• Merold Michael S
• Bucaj Mark
• Nevistich Igor Douglas
• Wenzel Richard
• Sheves Michal

Classifications

• G06Q40/04 Primary

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title

• G06Q40/03

  Credit; Loans; Processing thereof · CPC title

• H04L67/104 Primary

  Peer-to-peer [P2P] networks · CPC title