# US 2026/0089095 A1 · Using headers with forwarding tags and transport layer protocol parameters to forward results of endpoint processing unit computation through a network

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0089095 A1
- Document ID: US20260089095A1
- Status: Published application
- Kind: A1
- Application number: 19/227,470
- Filed: 2025-06-04
- Published: 2026-03-26
- Earliest claimed: 2024-09-21
- Assignees: Delos Data Inc.
- Applicants: Delos Data Inc.
- Inventors: Daniel P. Daly; Edward V. E. Doe; Alain J. E. Gravel
- Web page: https://patentgrep.com/patent/US20260089095A1

## CPC

- H04L 45/24: Multipath
- G06F 9/4881: {Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues}
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 9/5083: {Techniques for rebalancing the load in a distributed system}
- G06F 9/547: {Remote procedure calls [RPC]; Web services}
- H04L 45/123: {Evaluation of link metrics (techniques for monitoring network metrics H04L43/08)}
- H04L 47/33: using forward notification
- H04L 67/10: in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46)
- H04L 67/1001: for accessing one among a plurality of replicated servers
- H04L 67/60: Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources (admission control or resource allocation H04L47/70)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Some embodiments provide a method of forwarding data messages between endpoint processing units (EPUs) that are connected through a network and that perform computations to collectively execute a distributed application. The method generates, for each computation of each source EPU, a data message flow including one or more data messages that store a result of the computation. The generation includes specifying for each data message a header including (i) a tag that is not a network address and that is used for forwarding the data message through the network and (ii) one or more fields storing one or more transport layer protocol (TLP) parameters. The method uses the TLP parameters as layer 4 (L4) parameters to establish forwarding of each data message flow between the source EPU and the destination for the flow.

## Claims

### Claim 1

A method of forwarding data messages between endpoint processing units (EPUs) that are connected through a network and that perform computations to collectively execute a distributed application, the method comprising:
generating, for each computation of each source EPU, a data message flow comprising one or more data messages that store a result of the computation, said generating comprising specifying for each data message a header comprising (i) a tag that is not a network address and that is used for forwarding the data message through the network and (ii) one or more fields storing one or more transport layer protocol (TLP) parameters;
using the TLP parameters as layer 4 (L4) parameters to establish forwarding of each data message flow between the source EPU and the destination for the flow.

### Claim 2 (depends on 1)

The method of claim 1, wherein the EPUs are graphics processing units (GPUs).

### Claim 3 (depends on 1)

The method of claim 1, wherein the EPUs comprise at least one of graphics processing units (GPUs), tensor processing units (TPUs) and central processing units (CPUs).

### Claim 4 (depends on 1)

The method of claim 1, wherein using the TLP parameters comprises using the TLP parameters to request start of the forwarding of each flow from the source of the flow and to accept the start of the flow from the destination of the flow.

### Claim 5 (depends on 1)

The method of claim 1, wherein using the TLP parameters comprises using the TLP parameters to ensure reliable completion of the forwarding of each flow.

### Claim 6 (depends on 5)

The method of claim 5, wherein using the TLP parameters to ensure the reliable completion of each flow comprises:
using the TLP parameters to receive reception acknowledgments from a destination of the flow to signify the reception of different sets of segments of a result that are carried in different sets of data messages of the flow;
using the TLP parameters to identify any segment for which a reception acknowledge has not been received from the destination of the flow; and
using the TLP parameters comprises using the TLP parameters to specify completion of the forwarding of each flow.

### Claim 7 (depends on 1)

The method of claim 1, wherein using the TLP parameters to specify completion of the forwarding of each flow comprises using the TLP parameters to identify a data message in the flow as the data message that stores a last data segment in a plurality of result segments that is carried by the flow.

### Claim 8 (depends on 1)

The method of claim 1, wherein using the TLP parameters comprises using the TLP parameters to establish forwarding of each data message flow between source network interfaces of source EPUs that perform computations that produce results and destination network interfaces of destination EPUs that receive results of the computations in data message flows forwarded through the network by the source destination network interfaces.

### Claim 9 (depends on 1)

The method of claim 1, wherein the data message is a revised Ethernet frame as the data message's header stores a forwarding tag in the MAC field instead of storing a MAC address.

### Claim 10 (depends on 9)

The method of claim 9, wherein the forwarding elements comprise switches with data plane circuits that are custom designed and manufactured to perform forwarding operations based on the tags.

### Claim 11 (depends on 10)

The method of claim 10, wherein the forwarding elements further comprise off-the-shelf third-party switches that perform forwarding operations by performing match-action operations based on values stored in the MAC address fields of the data messages, said third party switches configured not to perform L2 learning operations and manually configured with match-action records that store tag values in match fields of the match-action records.

### Claim 12 (depends on 1)

The method of claim 1, wherein the tag does not comprise a layer 2, layer 3 or layer 4 network address.

### Claim 13

A non-transitory machine readable medium storing a program of forwarding data messages between graphics processing units (GPUs) that are connected through a network and that perform computations to collectively execute a distributed application, the program comprising sets of instructions for:
specifying, for each computation of each source GPU, a data message flow comprising one or more data messages that store a result of the computation, said generating comprising specifying for each data message a header comprising (i) a tag that is not a network address and that is used for forwarding the data message through the network and (ii) one or more fields storing one or more transport layer protocol (TLP) parameters;
using the TLP parameters as layer 4 (L4) parameters to establish forwarding of each data message flow between the source EPU and the destination for the flow.

### Claim 14 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the set of instructions for using the TLP parameters comprises a set of instructions for using the TLP parameters to request start of the forwarding of each flow from the source of the flow and to accept the start of the flow from the destination of the flow.

### Claim 15 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the set of instructions for using the TLP parameters comprises a set of instructions for using the TLP parameters to ensure reliable completion of the forwarding of each flow.

### Claim 16 (depends on 15)

The non-transitory machine readable medium of claim 15, wherein the set of instructions for using the TLP parameters to ensure the reliable completion of each flow comprises sets of instructions for:
using the TLP parameters to receive reception acknowledgments from a destination of the flow to signify the reception of different sets of segments of a result that are carried in different sets of data messages of the flow; and
using the TLP parameters to identify any segment for which a reception acknowledge has not been received from the destination of the flow.

### Claim 17 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the set of instructions for using the TLP parameters comprises a set of instructions for using the TLP parameters to specify completion of the forwarding of each flow.

### Claim 18 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the set of instructions for using the TLP parameters to specify completion of the forwarding of each flow comprises a set of instructions for using the TLP parameters to identify a data message in the flow as the data message that stores a last data segment in a plurality of result segments that is carried by the flow.

### Claim 19 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the set of instructions for using the TLP parameters comprises a set of instructions for using the TLP parameters to establish forwarding of each data message flow between source network interfaces of source EPUs that perform computations that produce results and destination network interfaces of destination EPUs that receive results of the computations in data message flows forwarded through the network by the source destination network interfaces.

### Claim 20 (depends on 13)

The non-transitory machine readable medium of claim 13, wherein the tag does not comprise a layer 2, layer 3 or layer 4 network address.
