# How to Send Data Streams Independently Without Connection Overload > A method for sending multiple streams of data between devices using independent, reliable paths that keep data in the correct order without needing a constant connection. - **Patent:** US 7904576 - **Original title:** Reliable datagram via independent source destination resources - **Owner:** Hewlett Packard Development Co LP - **Granted:** 2011 - **Status:** Active - **Times cited:** 42 - **Field:** telecommunications, semiconductors, software ## What it does This patent describes a system for moving data between a source and a destination using multiple independent 'Source and Destination Resources' (SDRs). Each SDR acts as a dedicated, reliable channel that guarantees the data arrives in the exact order it was sent, even though the overall system remains 'connectionless' from the perspective of the applications. By using multiple independent SDRs, the system can assign different priority levels to different data streams. For example, a high-priority stream for real-time control data can be processed faster than a lower-priority stream for background logs, all while ensuring that each stream maintains its own internal sequence integrity. ## What it does NOT cover - Does not cover systems that require a traditional persistent TCP connection between the source and destination. - Does not cover methods where data ordering is not guaranteed at the destination. - Does not cover communication architectures that rely on a single, shared resource for all data traffic. - Does not cover data transmission methods that do not use the specific 'unit of work' abstraction defined in the claims. ## The clever bit The system achieves 'reliable' and 'ordered' delivery while remaining 'connectionless' to the application. It offloads the complexity of sequencing and retransmission to the SDR layer, allowing applications to send data without managing the underlying communication state. ## Real-world examples 1. High-performance computing clusters 2. Enterprise data center interconnects 3. Distributed storage area networks ## Why it matters In large-scale data centers and high-performance computing, managing traffic is difficult. If all data uses one path, a slow process can block everything. This patent provides a way to isolate streams so that one application's data flow does not interfere with another's, which is essential for maintaining predictable performance in complex networked systems. ## Frequently asked questions ### What does How to Send Data Streams Independently Without Connection Overload cover? A method for sending multiple streams of data between devices using independent, reliable paths that keep data in the correct order without needing a constant connection. ### Who owns patent US 7904576? Hewlett Packard Development Co LP owns this patent, granted in 2011. ### When does this patent expire? This patent is expected to expire on January 7, 2028, when the invention enters the public domain. ### What is patent US 7904576 cited by? This patent has been cited by 42 later patents that build on its ideas. ### What problem does this patent solve? In large-scale data centers and high-performance computing, managing traffic is difficult. If all data uses one path, a slow process can block everything. This patent provides a way to isolate streams so that one application's data flow does not interfere with another's, which is essential for maintaining predictable performance in complex networked systems. ### What does this patent NOT cover? Does not cover systems that require a traditional persistent TCP connection between the source and destination. **Full plain-English explainer:** https://patentbrief.org/patent/us/7904576/reliable-datagram-via-independent-source-destination-resources **Original patent:** https://patents.google.com/patent/US7904576 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How to Split Computer Tasks Between Different Types of Processors](https://patentbrief.org/patent/us/9477524/aws-cloudformation) — A system that uses a physical backplane to connect two different types of computer processors—one for general tasks and one for real-time tasks—to improve efficiency. - [How Software Services Pass Messages Across Different Network Zones](https://patentbrief.org/patent/us/6073184/method-of-transmitting-a-notification-to-a-receiver-from-plural-notification-services-in-a-distributed-application-network-and-a-network-for-implementing-the-method) — A method for letting different parts of a large software network communicate by linking their notification services together like a relay race. - [How IBM's Storage Controllers Keep Data Backups in the Right Order](https://patentbrief.org/patent/us/5682513/cache-queue-entry-linking-for-dasd-record-updates) — A method for storage controllers to track and sequence data updates in a specific order, ensuring that remote backups remain consistent with the original data during a system failure. - [How Servers Share an IP Address for High Availability](https://patentbrief.org/patent/us/9571603/redundancy-network-protocol-system) — Cisco's 2017 patent describes a system where multiple servers can share a single virtual IP address to ensure an application stays online even if one server fails. - [How Distributed Servers Find Data Using Location Pointers](https://patentbrief.org/patent/us/7233978/method-and-apparatus-for-managing-location-information-in-a-network-separate-from-the-data-to-which-the-location-information-pertains) — A system that uses a network of specialized servers to track where specific data is located, allowing computers to find information by asking a server for its address rather than searching every machine.