IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0251148
(2011-09-30)
|
등록번호 |
US-8312101
(2012-11-13)
|
발명자
/ 주소 |
- McCanne, Steven
- Demmer, Michael J.
|
출원인 / 주소 |
- Riverbed Technology, Inc.
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
1 인용 특허 :
56 |
초록
▼
Transactions are accelerated by the transaction handlers by storing segments of data used in the transactions in persistent segment storage accessible to the server side transaction handler and in persistent segment storage accessible to the client side transaction handler. When data is to be sent b
Transactions are accelerated by the transaction handlers by storing segments of data used in the transactions in persistent segment storage accessible to the server side transaction handler and in persistent segment storage accessible to the client side transaction handler. When data is to be sent between the transaction handlers, the sending transaction handler compares the segments of the data to be sent with segments stored in its persistent segment storage and replaces segments of data with references to entries in its persistent segment storage that match or closely match the segments of data to be replaced. The receiving transaction store reconstructs the data sent by replacing segment references with corresponding segment data from its persistent segment storage.
대표청구항
▼
1. A system for accelerating communication between network hosts, comprising: a plurality of network devices comprising at least a first network device, a second network device and a third network device;referenced data storage, in each of the first, second and third network devices, the referenced
1. A system for accelerating communication between network hosts, comprising: a plurality of network devices comprising at least a first network device, a second network device and a third network device;referenced data storage, in each of the first, second and third network devices, the referenced data storage for storing referenced data that is referenced by globally unique references, a reference being globally unique in that it is distinguishable from other references that might be used to reference different referenced data;interception logic, in the first network device, for intercepting communications between a first network host and a second network host such that the communication is via the first network device and the second network device;transforming logic, in the first network device, for transforming data of a communication between the first host and the second host including at least substituting some of the data of the communication between the first network device to the second network device with globally unique references that reference data in the referenced data storage of at least one of the first, second or third network devices; anda reference resolver, in the second network device, that resolves received globally unique references that are unknown to the second network device by transmitting a request message to resolve the unknown reference of data, sent to at least one of the plurality of network devices other than the second network device. 2. The system of claim 1, wherein each of first, second and third network devices comprises a distinct referenced data storage, distinct interception logic, distinct transforming logic and a distinct reference resolver, and wherein the referenced data storage for a given network device contains referenced data usable for accelerating communications between the given network device and two or more other network devices. 3. The system of claim 1, wherein globally unique references are labels and reference data is stored as data sequences with each data sequence having an associated label. 4. The system of claim 1, wherein the transforming logic is configured to transform data of the communication between the first host and the second host by substituting globally unique references for data sequences, wherein the substituted globally unique references are associated with the substituted data sequences in the referenced data storage of the first host. 5. The system of claim 4, further comprising logic for storing data sequences to be replaced, storing into the referenced data storage, wherein a first data sequence is stored into the referenced data storage of the first network device with its globally unique reference prior to transforming the data of the first data sequence, thereby assuring a match. 6. The system of claim 1, wherein the first network device is a client-side transaction accelerator associated with the first host, the second network device is a server-side transaction accelerator associated with the second host, and wherein the communication is a client-server transaction between the first host and the second host. 7. The system of claim 6, further comprising a third host, wherein the third host is a server associated with the third network device, and wherein the transforming logic of the first network device is configured to use a globally unique reference in transactions both between the first host and the second host and between the first host and the third host when both the second host and third host are to receive, from the first host, referenced data in common. 8. The system of claim 1, wherein the globally unique references are such that a reference is globally unique in that it is distinguishable, during an expected lifetime of the reference, from other references that might be used during that expected lifetime to reference referenced data. 9. The system of claim 1, wherein the globally unique references are such that one data sequence can be used in more than one unrelated transaction and the globally unique references are globally unique within a group of conditions, but could be duplicated for different data sequences that would occur under a distinct group of conditions. 10. The system of claim 9, wherein the group of conditions includes a timeframe and a given globally unique reference is unique within one timeframe but could be duplicated for reuse in another timeframe, the system further comprising logic for deleting globally unique references and referenced data from the referenced data storage when a timeframe for lifetime of those referenced data items expires. 11. The system of claim 1, wherein the referenced data storage includes more than one copy of a given referenced data item, each with distinct references, with duplication based on parameters about the transactions in which the referenced data items are used. 12. The system of claim 1, wherein the referenced data storage includes only one copy of a given referenced data item that is used in more than one transaction with more than one host. 13. The system of claim 1, wherein a given referenced data item and its globally unique reference is used over multiple network devices and multiple transactions with different hosts. 14. The system of claim 1, wherein a value of a globally unique reference is a hash function of the data that the globally unique reference is substituted for. 15. A method for accelerating communication between network hosts in a network having a plurality of network devices comprising at least a first network device, a second network device and a third network device, the method comprising: storing referenced data that is referenced by globally unique references, a reference being globally unique in that it is distinguishable from other references that might be used to reference different referenced data;wherein each of the first, second, and third network devices is associated with a referenced data storage;intercepting communications between a first network host and a second network host such that the communication is via the first network device and the second network device;transforming data of a communication between the first host and the second host including at least substituting some of the data of the communication between the first network device to the second network device with globally unique references that reference data in the referenced data storage of at least one of the first, second or third network devices; andreceiving and resolving globally unique references that are unknown to the second network device by transmitting a request message to resolve the unknown reference of data, sent to at least one of the plurality of network devices other than the second network device. 16. The method of claim 15, wherein: each of first, second and third network devices comprises a distinct referenced data storage and is capable of intercepting communications, transforming data, and receiving and resolving globally unique references; andthe referenced data storage for a given network device contains referenced data usable for accelerating communications between the given network device and two or more other network devices. 17. The method of claim 15, wherein globally unique references are labels and reference data is stored as data sequences with each data sequence having an associated label. 18. The method of claim 15, wherein transforming data of the communication between the first host and the second host is performed by substituting globally unique references for data sequences, wherein the substituted globally unique references are associated with the substituted data sequences in the referenced data storage of the first host. 19. The method of claim 18, further comprising storing a first data sequence into the referenced data storage of the first network device with its globally unique reference prior to transforming the data of the first data sequence, thereby assuring a match. 20. The method of claim 15, wherein the first network device is a client-side transaction accelerator associated with the first host, the second network device is a server-side transaction accelerator associated with the second host, and wherein the communication is a client-server transaction between the first host and the second host. 21. The method of claim 20, further comprising a third host, wherein the third host is a server associated with the third network device, and the first network device uses a globally unique reference in transactions both between the first host and the second host and between the first host and the third host when both the second host and third host are to receive, from the first host, referenced data in common. 22. The method of claim 15, wherein the globally unique references are such that a reference is globally unique in that it is distinguishable, during an expected lifetime of the reference, from other references that might be used during that expected lifetime to reference referenced data. 23. The method of claim 15, wherein the globally unique references are such that one data sequence can be used in more than one unrelated transaction and the globally unique references are globally unique within a group of conditions, but could be duplicated for different data sequences that would occur under a distinct group of conditions. 24. The method of claim 23, wherein the group of conditions includes a timeframe and a given globally unique reference is unique within one timeframe but could be duplicated for reuse in another timeframe, the system further comprising logic for deleting globally unique references and referenced data from the referenced data storage when a timeframe for lifetime of those referenced data items expires. 25. The method of claim 15, wherein the referenced data storage includes more than one copy of a given referenced data item, each with distinct references, with duplication based on parameters about the transactions in which the referenced data items are used. 26. The method of claim 15, wherein the referenced data storage includes only one copy of a given referenced data item that is used in more than one transaction with more than one host. 27. The method of claim 15, wherein a given referenced data item and its globally unique reference is used over multiple network devices and multiple transactions with different hosts. 28. The method of claim 15, wherein a value of a globally unique reference is a hash function of the data that the globally unique reference is substituted for.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.