IP Library Granted Patent US 8,732,360
Granted Patent B2
US 8,732,360 · App. 12/276,010 · Granted May 20, 2014

System and method for accessing memory

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,732,360
App. No.
12/276,010
Granted
May 20, 2014
Kind
B2
Abstract

A storage system and method for storing information in memory nodes. The storage or memory nodes include a communication buffer. Flow of information to the storage nodes is controlled based upon constraints on the communication buffer. In one embodiment, communications between a master controller and a storage node have a determined maximum latency.

Claims (28)

1. A storage system comprising:

a first storage node for storing information wherein said first storage node includes a communication buffer;

a master controller for controlling flow to said first storage node based upon constraints of said communication buffer, wherein communication between said master controller and said storage node has a determined maximum latency;

wherein said master controller tracks a state of a buffer count of said first storage node by decrementing said buffer count when a request that involves a response is forwarded to said first memory node and incrementing said buffer count when a request response is received from said first memory node,

further wherein a size of a communication buffer is determined by multiplying the maximal bandwidth between said master controller and a nearest storage node by the latency from said master controller to a farthest storage node, and

further wherein said master controller is configured to forward a request that does not require a response to the first storage node according to the individual characteristics of said first storage node.

2. The storage system of claim 1 wherein said first storage node has the same number of buffers as a second storage node.

3. The storage system of claim 1 wherein utilization of said first storage node is greater than utilization of a second storage node.

4. The storage system of claim 1 wherein said first storage node and a second storage node are organized in a chain configuration.

5. A storage method comprising:

obtaining information on a distribution of and traffic flow between a plurality of memory nodes;

determining a buffer size within at least one of said plurality of memory nodes by multiplying the maximal bandwidth between a master controller and a nearest storage node by the latency from said master controller to a farthest storage node, and;

counting requests and responses to at least one of said plurality of memory nodes, wherein said counting comprises decrementing a buffer count corresponding to the number of buffers in a memory node of said plurality of memory nodes when a request is sent out; and

forwarding requests to said plurality of memory nodes based upon said buffer count if said requests require a response, and forwarding requests to said plurality of memory nodes based on characteristics of said individual nodes when said requests do not require a response.

6. The storage method of claim 5 wherein said information on said distribution indicates said plurality of memory nodes are organized in a chain configuration.

7. The storage method of claim 6 wherein said buffer size is determined for one of said plurality of said memory nodes that is closest to a master controller.

8. The storage method of claim 6 wherein requests and responses are counted for one of said plurality of said memory nodes that is closest to a master controller.

9. The storage method of claim 6 wherein said determining a buffer size is ascertained by examining the nature of a bus topology coupling said plurality of memory nodes.

10. The storage method of claim 6 further comprising utilizing a first one of said plurality of said memory nodes that is closest to a master controller the same as or more than other of said plurality of memory nodes.

11. A storage system comprising:

a plurality of storage nodes for storing information wherein said plurality of storage nodes are organized in a chained network and each of said plurality of storage node includes a respective communication buffer comprising a respective plurality of buffers; and

a master controller for controlling a flow of requests requiring responses to said plurality of storage nodes based upon a respective count of said respective plurality of buffers within said plurality of storage nodes, and a flow of requests which do not require responses to said plurality of storage nodes based upon individual characteristics of said storage nodes, wherein communication between said master controller and said plurality storage nodes has a determined maximum latency,

wherein a size a communication buffer is determined by multiplying the maximal bandwidth between said master controller and a nearest storage node by the latency from said master controller to a farthest storage node, and

wherein said master controller is operable to track a state of a respective buffer count of said respective plurality of buffers by decrementing said respective buffer count when a request that involves a response is forwarded to a respective storage node and incrementing said buffer count when a request response is received from said respective storage node.

12. The storage system of claim 11 wherein said master controller has information on the distribution of said plurality of storage nodes within a network and how traffic flows within said chained network.

13. The storage system of claim 12 wherein said master controller forwards requests that require responses and said requests that require responses are forwarded to said plurality of memory nodes according to a count of buffer space available in at least one of said plurality of memory nodes.

14. The storage system of claim 12 wherein communication protocol tokens are not utilized for scheduling requests to said plurality of memory nodes.

15. The storage system of claim 11 wherein said master controller counts requests to and responses from at least one of said plurality of memory nodes in said network.

Assignments (7)
CORRECTIVE ASSIGNMENT TO CORRECT THE 8647899 PREVIOUSLY RECORDED ON REEL 035240 FRAME 0429. ASSIGNOR(S) HEREBY CONFIRMS THE SECURITY INTERST. Recorded Nov 3, 2020
From: CYPRESS SEMICONDUCTOR CORPORATION; SPANSION LLC
To: MORGAN STANLEY SENIOR FUNDING, INC.
Reel/Frame 058002/0470 →
PARTIAL RELEASE OF SECURITY INTEREST IN PATENTS Recorded Aug 11, 2016
From: MORGAN STANLEY SENIOR FUNDING, INC., AS COLLATERAL AGENT
To: CYPRESS SEMICONDUCTOR CORPORATION; SPANSION LLC
Reel/Frame 039708/0001 →
SECURITY INTEREST Recorded Mar 21, 2015
From: CYPRESS SEMICONDUCTOR CORPORATION; SPANSION LLC
To: MORGAN STANLEY SENIOR FUNDING, INC.
Reel/Frame 035240/0429 →
RELEASE OF SECURITY INTEREST Recorded Mar 13, 2015
From: BARCLAYS BANK PLC
To: SPANSION LLC; SPANSION INC.; SPANSION TECHNOLOGY LLC
Reel/Frame 035201/0159 →
SECURITY AGREEMENT Recorded Aug 6, 2010
From: SPANSION LLC; SPANSION INC.
To: BARCLAYS BANK PLC
Reel/Frame 024802/0054 →
CORRECTIVE ASSIGNMENT TO CORRECT THE ASSIGNEE PREVIOUSLY RECORDED ON REEL 024551 FRAME 0662. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNEE NAME FROM SPANSION LLC TO SPANSION LLC AND HITACHI, LTD.. Recorded Jun 28, 2010
From: ISAAC, ROGER DWAIN; MIURA, SEIJI
To: SPANSION LLC; HITACHI, LTD.
Reel/Frame 024605/0408 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 17, 2010
From: ISAAC, ROGER DWAIN; MIURA, SEIJI
To: SPANSION LLC
Reel/Frame 024551/0662 →