REMOTE PROCESSING AND MEMORY UTILIZATION

• US 20130290473A1
• Filed: 08/13/2012
• Published: 10/31/2013
• Est. Priority Date: 08/09/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Eisley, Noel A., Heidelberger, Philip, Kahle, James A., Senger, Robert M., Steinmacher-Burow, Burkhard, Sugawara, Yutaka, Chen, Dong, Petrini, Fabrizio
• Granted Patent: US 10,152,450 B2
• Time in Patent Office: Days
• US Class Current: 709/216
• CPC Class Codes: G06F 15/17331 Distributed shared memory [DSM], e.g. remote direct memory access [RDMA]
• CPC Class Codes: G06F 9/544 Buffers; Shared memory; Pipes
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/133 Protocols for remote procedure calls [RPC]

Abstract

According to one embodiment of the present invention, a system for operating memory includes a first node coupled to a second node by a network, the system configured to perform a method including receiving the remote transaction message from the second node in a processing element in the first node via the network, wherein the remote transaction message bypasses a main processor in the first node as it is transmitted to the processing element. In addition, the method includes accessing, by the processing element, data from a location in a memory in the first node based on the remote transaction message, and performing, by the processing element, computations based on the data and the remote transaction message.

Claims

1. A method for operating a memory system comprising a first node coupled to a second node by a network, the method comprising:

receiving a remote transaction message from the second node in a processing element in the first node via the network, wherein the remote transaction message bypasses a main processor in the first node as it is transmitted to the processing element;

accessing, by the processing element, data from a location in a memory in the first node based on the remote transaction message; and

performing, by the processing element, computations based on the data and the remote transaction message.

2. The method of claim 1, wherein receiving the remote transaction message comprises receiving a remote transaction message that includes at least one of a selected instruction or a pointer to the selected instruction in the memory in the first node, the selected instruction using data from the memory or from the remote transaction message.

3. The method of claim 1, comprising transmitting, from the processing element, a return value, a signal or a second remote transaction message following execution of the remote transaction message.

4. The method of claim 3, comprising receiving the return value at the second node and sending a second remote transaction message to the first node based on receiving the return value or signal following execution of the remote transaction message.

5. The method of claim 1, wherein receiving the remote transaction message comprises receiving from a second main processor or a second processing element in the second node from a second network interface in the second node at the processing element via a first network interface in the first node.

6. The method of claim 1, wherein receiving the remote transaction message in the processing element in the first node comprises receiving the remote transaction message in a processing element located in an active buffered memory device in the first node.

7. The method of claim 1, wherein receiving the remote transaction message comprises receiving, substantially simultaneously, a plurality of remote transaction messages from the second node at a plurality of nodes including the first node, via the network.

8. The method of claim 1, wherein receiving the remote transaction message comprises receiving a remote transaction message comprising an atomic transaction.

9. The method of claim 1, wherein receiving the remote transaction message comprises receiving, from the second node, a first transaction message and a second transaction message to the first node and a third node, respectively, via the network, wherein a result used in the second node is based on at least one result from execution of the first transaction message and the second transaction message.

10. The method of claim 1, wherein accessing, by the processing element, the location in the memory in the first node based on the remote transaction message comprises translating, by the processing element, a virtual address to a real address and accessing the data at the location in the memory using the real address.

11. A method for operating a system comprising a first node coupled to a second node by a network, the method comprising:

receiving a remote transaction message from the second node in a processing element in the first node via the network, wherein the remote transaction message bypasses a main processor in the first node as it is transmitted to the processing element;

accessing, by the processing element, a location in a memory in the first node based on the remote transaction message; and

performing, by the processing element, computations on data based on the remote transaction message, the data comprising at least one of;

data from the location in the memory and data from within the remote transaction message.

12. The method of claim 11, wherein receiving the remote transaction message comprises receiving a remote transaction message that includes at least one of a selected instruction or a pointer to the selected instruction in the memory in the first node, the selected instruction using data from the memory or from the remote transaction message.

13. The method of claim 11, comprising transmitting, from the processing element, a return value, a signal or a second remote transaction message following execution of the remote transaction message.

14. The method of claim 13, comprising receiving a second remote transaction message at the first node from the second node based on receiving the return value or signal at the second node following execution of the remote transaction message.

15. The method of claim 11, wherein receiving the remote transaction message comprises receiving the remote transaction message from a second main processor or a second processing element in the second node from a second network interface in the second node at the processing element via a first network interface in the first node.

16. The method of claim 11, wherein receiving the remote transaction message in the processing element in the first node comprises, receiving the remote transaction message in a processing element located in an active buffered memory device in the first node.

17. The method of claim 11, wherein receiving the remote transaction message comprises receiving, substantially simultaneously, a plurality of remote transaction messages from the second node at a plurality of nodes, including the first node, via the network.

18. The method of claim 11, wherein receiving the remote transaction message comprises receiving a remote transaction message comprising an atomic transaction.

19. The method of claim 11, wherein receiving the remote transaction message comprises receiving, from the second node, a first transaction message and a second transaction message to the first node and a third node, respectively, via the network, wherein a result used in the second node is based on at least one result from execution of the first transaction message and the second transaction message.

20. The method of claim 11, wherein accessing, by the processing element, the location in the memory in the first node based on the remote transaction message comprises translating, by the processing element, a virtual address to a real address and accessing the data at the location in the memory using the real address.