US-20260126925-A1 - MEMORY SYSTEM WITH MULTIPLE STRIPING OF RAID GROUPS AND METHOD FOR PERFORMING THE SAME

Abstract

A method comprises receiving data. The method comprises determining redundancy data for the received data. The method comprises storing the received data and the redundancy data in one or more memory modules of a group of memory modules. The method comprises receiving a read data request. The method comprises reconstructing the requested data, based on an erase data operation preventing a read data operation on the one or more memory modules where the requested data is stored. A time delay to reconstruct the requested data is less than a time period where a memory module of the group of memory modules is in an erase state.

Inventors

• Jon C.R. Bennett

Assignees

• INNOVATIONS IN MEMORY LLC

Dates

Publication Date

20260507

Application Date

20251002

Claims (1)

1. 1 . A method comprising: receiving data; determining redundancy data for the received data; storing the received data and the redundancy data in one or more memory modules of a group of memory modules; receiving a read data request, and reconstructing the requested data, based on an erase data operation preventing a read data operation on the one or more memory modules where the requested data is stored, wherein a time delay to reconstruct the requested data is less than a time period where a memory module of the group of memory modules is in an erase state.

Description

This application is a continuation of U.S. patent application Ser. No. 18/608,221 filed on Mar. 18, 2024, which is a continuation of U.S. patent application Ser. No. 18/178,629, filed on Mar. 6, 2023, now U.S. Pat. No. 11,960,743 issued Mar. 27, 2024, which is continuation of U.S. patent application Ser. No. 17/321,189, filed on May 14, 2021, now U.S. Pat. No. 11,599,285 issued Mar. 7, 2023, which is a continuation of U.S. patent application Ser. No. 16/444,176, filed on Jun. 18, 2019, now U.S. Pat. No. 11,010,076 issued May 18, 2021, which is a continuation of U.S. patent application Ser. No. 15/467,515, filed on Mar. 23, 2017, now U.S. Pat. No. 10,372,366 issued Aug. 6, 2019, which is a continuation of U.S. patent application Ser. No. 12/901,224, filed on Oct. 8, 2010, now U.S. Pat. No. 9,632,870, issued Apr. 25, 2017, which claims the benefit of U.S. Provisional Application No. 61/250,216 filed on Oct. 9, 2009. U.S. patent application Ser. No. 12/901,224, filed on Oct. 8, 2010, now U.S. Pat. No. 9,632,870, issued Apr. 25, 2017, is a continuation-in-part of U.S. patent application Ser. No. 12/079,364, filed on Mar. 26, 2008, now U.S. Pat. No. 8,200,887, issued Jun. 12, 2012, which claims the benefit of U.S. Provisional Application No. 60/920,737, filed on Mar. 29, 2007, each of which are all incorporated herein by reference in their entireties. TECHNICAL FIELD The present application relates to computer memory systems, and a method of using the same. BACKGROUND Computer memory systems may be either of the persistent or non-persistent type. Examples of persistent memory types are magnetic cores, disk drives, tape drives and semiconductor FLASH memories. Non-persistent memory types may be semiconductor memories such as DRAM or the like. Non-persistent memory types typically have rapid access times for both reading and writing of data and are used as computer main memory or cache memory. The data is retained in such memories by means which require a supply of power, and the information stored therein may be lost if the power is interrupted. Systems of non-persistent memory usually have a back-up power supply, which may be a capacitive storage device for short duration power interruptions, or back-up power supplies using batteries, generators, or the like for longer term data retention. Persistent storage devices, such as disk, tape or FLASH memory retain stored data even if the power source is removed from the device, and are often used to back up the non-persistent data storage devices, and for longer term data storage where the cost or reliability of providing continuous power is not practical. Additionally, since larger amounts of data are stored in the persistent data storage devices, the technologies developed have been oriented towards the reduction of the cost per bit of storage, rather than access speed. Thus, many computing systems use a variety of memory types to perform different storage functions, where immediately needed data is stored in non-persistent storage, and may be backed up in persistent storage, while less frequently accessed data, and large groupings of data are stored in persistent storage. Computer data base systems, which may be termed data centers, or distributed data systems such as the Internet and the storage devices associated therewith may store vast amounts of data. Certain aspects of this architecture are currently being termed “cloud” computing. Today, such data storage requirements may exceed 1000 Terabytes (TB), and are expected to continue to grow. Many of these data sets are substantially larger than the capability of non-persistent storage to immediately access, and the response time of the servers in a data center when servicing a request from a client computer may be a serious bottleneck in system performance. Much of this restriction is a result of the data-access-time latency of the persistent storage media. For tape systems, the linear tape must be translated so that the data portion to be read or written is positioned at the reading or writing heads. Similarly, for a disk, the head must be positioned so as to be over the data track where the desired sector of data is located, and then the disk controller waits until the sector rotates under the positioned head. Any of these operations is substantially slower than reading or writing to non-persistent memory devices. Such limitations are particularly severe where data single memory locations having a random location in the data base need to be read, written or modified. The time between a request for data stored in a memory and the retrieval of data from the memory may be called the latency. FLASH memories, amongst the presently used persistent memory technologies, has a lower latency than mechanical devices such as disks, but has significantly more latency than the non-persistent memory types in current use. The price of FLASH memory and similar solid state technologies has traditionally been governed by a princi