Block storage with volume locality placement based on performance requirements

Patent metadata

Field	Value
Publication number	US-11836359-B2
Application number	US-202016747707-A
Country	US
Kind code	B2
Filing date	Jan 21, 2020
Priority date	Apr 30, 2018
Publication date	Dec 5, 2023
Grant date	Dec 5, 2023

Abstract

A block storage service can ensure volumes are placed in a same region as an attached virtual machine instance for performance and durability guarantees. A region can reference multiple things, but one example is that a volume is within a same spine as a virtual machine in order to meet performance guarantees. Each region can have a buffer of server computers held in reserve for volumes having a type where performance guarantees are required. If performance guarantees cannot be met, a rejection is transmitted to the customer. In another embodiment, the customer can provide a list in priority order of different volume types so that if a desired volume type cannot be placed, then alternative volume types can be used.

First claim

What is claimed is:

1. A system comprising:
a first server computer hosting a virtual machine instance;
a second server computer configured to host storage volumes;
at least one server computer in a control plane configured to:
receive a request to associate a storage volume with the virtual machine instance; and
select the second server computer to host the storage volume associated with the virtual machine instance based at least in part on a proximity of the second server computer to the first server computer hosting the virtual machine instance,

wherein the first server computer and second server computer are within a data center of a cloud provider, and
wherein the data center includes server computers interconnected in a spine-leaf network topology and the second server computer is selected to be within a same spine as the first server computer in the spine-leaf network topology.

2. The system of claim 1 , wherein the proximity of the second server computer to the first server computer is selected to ensure that communications between the virtual machine instance and the storage volume are limited to traversing a predetermined number of switches.

3. The system of claim 2 , wherein the predetermined number of switches are between 1 and 5 switches.

4. The system of claim 1 , wherein, due to the proximity between the virtual machine instance and the storage volume, a minimum latency is guaranteed for communications there between.

5. A computer-implemented method, comprising:
receiving a request to associate a volume with a virtual machine hosted by a cloud provider; and
launching the volume on a storage device within the cloud provider, the storage device selected based at least in part on a proximity between the storage device and a server computer hosting the virtual machine,
wherein the storage device and the server computer hosting the virtual machine are within a data center of the cloud provider, and
wherein the data center includes server computers interconnected in a spine-leaf network topology and the storage device is selected to be within a same spine in the spine-leaf network topology as the server computer hosting the virtual machine.

6. The computer-implemented method of claim 5 , wherein the server computer hosting the virtual machine is a first server computer and the storage device is a second server computer, the first and second server computers being coupled together through a network of routers within the data center.

7. The computer-implemented method of claim 5 , wherein the proximity between the storage device and the server computer hosting the virtual machine ensures that performance characteristics are met including any one of the following: a minimum Input/Output Operations Per Second (IOPS), a minimum latency or a minimum throughput.

8. The computer-implemented method of claim 5 , wherein the volume is a first volume, and wherein the computer-implemented method further includes:
monitoring usage of volumes within the cloud provider; and
moving a second volume having usage below a threshold level to free up storage space for the first volume, wherein the second volume is moved from a first data center including the server computer hosting the virtual machine to a second data center.

9. The computer-implemented method of claim 5 , wherein if the volume on the storage device is not within the proximity which is needed to meet performance requirements, the request to associate the volume is rejected.

10. The computer-implemented method of claim 5 , wherein the request includes a list of alternative levels of performance characteristics in a priority order for communications between the storage device and the server computer hosting the virtual machine.

11. The computer-implemented method of claim 10 , wherein if a first performance characteristic on the list is not achievable, then a next level of performance characteristics on the list is analyzed and a different locality of the storage device is chosen to meet the next level of performance characteristics.

12. The computer-implemented method of claim 5 , wherein the proximity of the storage device to the server computer hosting the virtual machine is selected to ensure that communications between the virtual machine and the volume are limited to traversing a predetermined number of switches in a network.

13. The computer-implemented method of claim 12 , wherein the predetermined number of switches are between 1 and 5 switches.

14. A method, comprising:
receiving a request to associate a volume hosted by a cloud provider with a virtual machine hosted by the cloud provider; and
launching the volume on a storage device, the storage device selected based at least in part on a number of switches that are traversed for communications between the storage device and a server computer hosting the virtual machine,
wherein the storage device and the server computer hosting the virtual machine are within a data center of the cloud provider, and
wherein the data center includes server computers interconnected in a spine-leaf network topology and the storage device is selected to be within a same spine in the spine-leaf network topology as the server computer hosting the virtual machine.

15. The method of claim 14 , wherein the volume is a first volume, the method further including moving a second volume having usage below a threshold level to free up storage space for the first volume, wherein the second volume is moved from a first data center including the server computer hosting the virtual machine to a second data center.

16. The method of claim 14 , further including communicating between the storage device and the server computer hosting the virtual machine with a minimum guaranteed latency.

Assignees

• Amazon Tech Inc

Inventors

• Greenwood Christopher Magee
• Brennan Patrick E
• Flaherty Mitchell Gannon
• Guo Yilin
• Herndon Jr Gary Michael
• Venugopal Sriram
• Yu Linfeng
• Lin Wells

Classifications

• G06F3/0631 Primary

  by allocating resources to storage systems · CPC title

• G06F3/067

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

• G06F3/0611

  in relation to response time · CPC title

• G06F3/0613

  in relation to throughput · CPC title

• G06F3/0647

  Migration mechanisms · CPC title