# US 2016/0253101 A1 · Memory Access and Detecting Memory Failures using Dynamically Replicated Memory

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

- Number: US 2016/0253101 A1
- Document ID: US20160253101A1
- Status: Published application
- Kind: A1
- Application number: 15/149,999
- Filed: 2016-05-09
- Published: 2016-09-01
- Earliest claimed: 2009-11-18
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Engin Ipek; Douglas Christopher Burger; Thomas Moscibroda; Edmund Bernard Nightingale; Jeremy P. Condit
- Web page: https://patentgrep.com/patent/US20160253101A1

## CPC

- G06F 3/065: {Replication mechanisms}
- G06F 3/061: {Improving I/O performance}
- G06F 3/0619: {in relation to data integrity, e.g. data losses, bit errors}
- G06F 3/064: {Management of blocks}
- G06F 3/0653: {Monitoring storage devices or systems}
- G06F 3/0673: {Single storage device}
- G06F 3/0683: {Plurality of storage devices}
- G06F 11/1008: {in individual solid state devices (G06F11/1004 takes precedence)}
- G06F 11/1666: {where the redundant component is memory or memory area}
- G06F 12/1027: using associative or pseudo-associative address translation means, e.g. translation look-aside buffer [TLB]
- G06F 11/1076: {Parity data used in redundant arrays of independent storages, e.g. in RAID systems}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Dynamically replicated memory is usable to allocate new memory space from failed memory pages by pairing compatible failed memory pages to reuse otherwise unusable failed memory pages. Dynamically replicating memory involves detecting and recording memory faults, reclaiming failed memory pages for later use, recovering from detected memory faults, and scheduling access to replicated memory pages.

## Claims

### Claim 1

(canceled)

### Claim 2

One or more computer-readable storage media storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:
receiving a memory access request for data from a memory controller, the data being included in a set of faulty memory pages including at least a first memory page and a second memory page, wherein the first memory page includes a failed memory block that is located at a relative position different from a failed memory block of the second memory page; and
sending a first data request for data of the first memory page and a second data request for data of the second memory page.

### Claim 3 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising receiving and consuming the data of the first memory page at the memory controller without waiting for the data of the second memory page.

### Claim 4 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising:
checking a parity bit of the data of the first memory page; and
waiting to receive the data of the second memory page in response to the parity bit of the data of the first memory page signaling a memory fault.

### Claim 5 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising storing the data of the first memory page in a memory if the data of the first memory page is non-faulty and storing the data of the second memory page in the memory if the data of the second memory page is non-faulty.

### Claim 6 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising replicating the first memory page to create the second memory page.

### Claim 7 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising canceling the second data request for the second memory page in response to receiving the data of the first memory page and determining that the data of the first memory page is fault-free.

### Claim 8 (depends on 2)

The one or more computer-readable storage media of claim 2, wherein the first memory page comprises phase change memory.

### Claim 9 (depends on 2)

The one or more computer-readable storage media of claim 2, the acts further comprising dynamically replicating the first memory page as errors are found.

### Claim 10

A system comprising:
memory and one or more processors;
a memory controller module stored in the memory and executable by the one or more processors to make memory write requests for data included in a set of faulty memory pages including at least a primary memory page and a duplicate memory page, at least one failed memory block of the primary memory page being located at a relative position different from a failed memory block of the duplicate memory page; and
a transaction queue module to store the memory write requests.

### Claim 11 (depends on 10)

The system of claim 10, wherein the memory controller module is further configured to insert a first data write request for the primary memory page in the transaction queue and to insert a second data write request for the duplicate memory page in the transaction queue.

### Claim 12 (depends on 11)

The system of claim 11, wherein the memory controller module is further configured to remove the second data write request for the duplicate memory page from the transaction queue in response to a completion of a checker-read.

### Claim 13 (depends on 10)

The system of claim 10, wherein the memory controller is further configured to send a first memory access request and a second memory access request to the primary memory page and the duplicate memory page respectively, receive and consume data of the primary memory page without waiting for data of the duplicate memory page.

### Claim 14 (depends on 10)

The system of claim 10, wherein the memory comprises phase change memory.

### Claim 15

A method comprising:
receiving an incoming memory access request for data from a memory controller, the data being included in a set of faulty memory pages including at least a first memory page and a second memory page, at least one failed memory block of the first memory page being located at a relative position different from a failed memory block of the second memory page;
placing a first memory access request for data of the first memory page and a second memory access request for data of the second memory page;
receiving the data of the first memory page at the memory controller; and
waiting to receive the data of the second memory page in response to determining that the data of the first memory page is faulty.

### Claim 16 (depends on 15)

The method of claim 15, wherein determining that the data of the first memory page is faulty comprises determining that a parity bit of the data of the first memory page signals a memory fault.

### Claim 17 (depends on 15)

The method of claim 15, further comprising storing the data of the second memory page in a memory when the data of the second memory page is non-faulty.

### Claim 18 (depends on 15)

The method of claim 15, further comprising replicating the first memory page to create the second memory page.

### Claim 19 (depends on 15)

The method of claim 15, wherein the first memory page and the second memory page are located on different memory banks.

### Claim 20 (depends on 15)

The method of claim 15, wherein the first memory page comprises phase change memory.

### Claim 21 (depends on 15)

The method of claim 15, further comprising dynamically replicating the first memory page as errors are found.
