Failover for failed secondary in cross-zone replicated block storage devices

Patent metadata

Field	Value
Publication number	US-11237751-B2
Application number	US-201916579687-A
Country	US
Kind code	B2
Filing date	Sep 23, 2019
Priority date	Sep 23, 2019
Publication date	Feb 1, 2022
Grant date	Feb 1, 2022

Abstract

The present disclosure generally relates to creating virtualized block storage devices whose data is replicated across isolated computing systems to lower risk of data loss even in wide-scale events, such as natural disasters. The virtualized device can include at least two volumes, each of which is implemented in a distinct computing system. In the case of a failed volume, a new volume can be created and populated with data from the surviving volume. During population, new writes can continue to be replicated to the new volume. The population process can write data from the surviving volume to the new volume “under” new writes, such that the population process does not overwrite data included in the new writes.

First claim

What is claimed is:

1. A system to provide redundancy in a virtualized storage device replicated across at least a first and second zone of a plurality of zones, wherein the plurality of zones are in communication via a communication network but are electrically and physically isolated from one another at least partly by being geographically distanced from one another within a geographic region and having independent electrical power, and wherein the virtualized storage device comprises a first volume in the first zone and a second volume in the second zone, each of the first and second volumes configured to provide a distinct copy of data of the virtualized storage device, the system comprising:
a first computing system associated with the first zone, the first computing system including at least a first computing device and a second computing device that are collectively configured to implement the first volume to provide a first copy of the data of the virtualized storage device by providing redundant endpoints for reading from and writing to the first volume;
a second computing system associated with the second zone, the second computing system including at least a first computing device and a second computing device that are collectively configured to implement the second volume to provide a second copy of the data of the virtualized storage device by providing redundant endpoints for reading from and writing to the second volume; and
one or more computing devices implementing a multi-zone control plane service configured to:
detect a failure of the second volume;
select a zone from the plurality of zones in which to create a new volume for the virtualized storage device;
generate the new volume within the selected zone, wherein the new volume is implemented by at least two computing devices within the selected zone, and wherein the new volume provides a third copy of the data of the virtualized storage device;
instruct the first volume to maintain a record of data in the first volume at a point in time; and
instruct the new volume to populate the new volume with the data in the first volume at the point in time using the record of data in the first volume at the point in time;

wherein the at least two computing devices implementing the new volume are configured to replicate writes received at the first volume subsequent to the point in time, and to populate the new volume with the data in the first volume at the point in time using the record of data in the first volume at the point in time without overwriting replicated writes received subsequent to the point in time at least partly by copying, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have not been written to subsequent to the point in time and declining to copy, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have been written to subsequent to the point in time.

2. The system of claim 1 , wherein the virtualized storage device represents a block storage device of a virtual machine instance.

3. The system of claim 1 , wherein the first volume is designated by the multi-zone control plane service as a primary volume for the virtualized storage device, the primary volume having authority to accept writes to the virtualized storage device.

4. The system of claim 1 , wherein the first computing device implementing the first volume is designated as a primary computing device for the first volume having authority to accept writes to the first volume, and wherein the first computing device is configured to:
subsequent to the point in time and prior to completion of population of the new volume with data from the record:
obtain a request to write data to the virtualized storage device;
store within the first volume; and
replicate the data to one or more secondary volumes of the virtualized storage device.

5. The system of claim 4 , wherein the one or more computing devices implementing the multi-zone control plane service are further configured to notify the first computing device implementing the first volume that the new volume is a secondary volume of the virtualized storage device.

6. A computer-implemented method to provide redundancy in a virtualized storage device replicated across at least a first and second zone of a plurality of zones, wherein the plurality of zones are in communication via a communication network but are isolated from one another, and wherein the virtualized storage device comprises a first volume in the first zone and a second volume in the second zone, the first volume configured to provide a first copy of data of the virtualized storage device and the second volume configured to provide a second copy of the data of the virtualized storage device, the computer-implemented method comprising:
detecting a failure of the second volume of the virtualized storage device;
selecting a zone from the plurality of zones in which to create a new volume for the virtualized storage device;
generating the new volume within the selected zone, wherein the new volume is implemented by at least two computing devices within the selected zone, and wherein the new volume provides a third copy of the data of the virtualized storage device;
causing the first volume to maintain a record of data in the first volume at a point in time; and
causing the new volume to:
replicate writes to the virtualized storage device obtained at the first volume subsequent to the point in time; and

populate the new volume with the data in the first volume at the point in time using the record of data in the first volume at the point in time without overwriting replicated writes received subsequent to the point in time at least partly by copying, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have not been written to subsequent to the point in time without copying, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have been written to subsequent to the point in time.

7. The method of claim 6 , wherein selecting the zone from the plurality of zones in which to create the new volume for the virtualized storage device comprises:
detecting that the second zone has not failed; and
assigning the second zone as the selected zone.

8. The method of claim 6 , wherein selecting the zone from the plurality of zones in which to create the new volume for the virtualized storage device comprises:
detecting that the second zone has failed; and
selecting the zone in which to create the new volume from among a set comprising the plurality of zones but excluding the first zone and the second zone.

9. The method of claim 6 , wherein writes to the virtualized storage device are assigned sequential write numbers, and wherein causing the new volume to populate data within the new volume from the record of data in the first volume at the point in time without overwriting replicas of the writes obtained at the first volume subsequent to the point in time at least partly by copying, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have not been written to subsequent to the point in time without copying, from the first volume to the new volume, data in the record that corresponds to locations of the first volume that have been written to subsequent to the point in time comprises causing the new volume to decline to overwrite data within the new volume associated with a write number higher than a write number associated with data from the record of data in the first volume at

Assignees

• Amazon Tech Inc

Inventors

• Muniswamy-Reddy Kiran-Kumar
• Wei Danny
• Kusters Norbert Paul
• Seguy Romain Benoit
• Upadhyaya Rahul
• Jain Divya Ashok Kumar
• Zaharchuk William
• Lin Wells

Classifications

• G06F3/065 Primary

  Replication mechanisms · CPC title

• G06F3/064

  Management of blocks · CPC title

• G06F3/0664

  at device level, e.g. emulation of a storage device or system · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F11/203

  using migration · CPC title