Data storage system management techniques and metrics

Patent metadata

Field	Value
Publication number	US-11513684-B1
Application number	US-202117355242-A
Country	US
Kind code	B1
Filing date	Jun 23, 2021
Priority date	Jun 23, 2021
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

Data storage management may include configuring a stretched volume in a metro cluster configuration using a first volume and a second volume both configured as a same logical volume, L1; receiving I/O operations at the metro cluster configuration directed to L1, wherein the I/O operations include a write operations that are received at a first system of the metro cluster configuration and that write first data to the first volume; viewing values of a metric regarding data replication performed for the first volume to replicate data between systems of the metro cluster configuration for writes directed to the first volume; responsive to said viewing, determining whether the first metric values denote a problem has occurred in the metro cluster configuration at a first point in time; and responsive to determining that the problem has occurred at the first point in time, performing one or more corrective actions.

First claim

What is claimed is:

1. A method of performing data storage management comprising:
configuring a stretched volume in a metro cluster configuration including a first data storage system, DS 1 , and a second data storage system, DS 2 , wherein the stretched volume is configured using a first volume, V 1 , on DS 1 and a second volume, V 2 , on DS 2 , wherein V 1 and V 2 are configured as a same logical volume, L 1 , and exposed to a host as the same logical volume, L 1 , over a plurality of paths including a first path, P 1 , and a second path, P 2 , wherein P 1 is between the host and DS 1 , and wherein P 2 is between the host and DS 2 ;
receiving a plurality of I/O operations at the metro cluster configuration directed to the same logical volume L 1 , wherein the plurality of I/O operations includes a first plurality of write operations that are received at DS 1 and that write first data to V 1 ;
viewing values of one or more metrics regarding data replication performed for V 1 at a plurality of points in time, wherein the data replication is from DS 1 to DS 2 to replicate the first data written by the first plurality of write operations whereby the replicated first data is subsequently written to V 2 on DS 2 , wherein the one or more metrics include at least one of: an average mirroring overhead latency for V 1 indicating latency experienced when performing the data replication that replicates writes directed to V 1 from DS 1 to DS 2 , a mirrored write I/O rate denoting a number of write I/O operations in a specified unit of time for write I/Os that are directed to V 1 and replicated from DS 1 to DS 2 , and an average mirrored write I/O bandwidth denoting an I/O bandwidth for write data transmitted when replicating write I/Os directed to V 1 from DS 1 to DS 2 ;
responsive to said viewing, determining whether one or more of the values of the one or more metrics denote that a problem has occurred in the metro cluster configuration at a first of the plurality of points in time; and
responsive to determining that the problem has occurred at the first point in time, performing one or more corrective actions.

2. The method of claim 1 , further comprising:
responsive to determining that the one or more values for the one or more metrics indicate that write data directed to V 1 is not being replicated from DS 1 to DS 2 at corresponding points in time, determining that the problem that has occurred at the first point in time is replication failure when replicating write data directed to V 1 from DS 1 to DS 2 since, at the first point in time, V 1 is stretched and included in a configuration for the stretched volume.

3. The method of claim 2 , wherein the one or more corrective actions performed in response to determining that the problem is replication failure where write data directed to V 1 is not being replicated from DS 1 to DS 2 include one or more of: placing DS 2 back online, rebooting DS 2 , and fixing or replacing a failed link used for replicating data from DS 1 to DS 2 .

4. The method claim 1 , wherein V 1 and V 2 are configured for synchronous data replication, wherein the data replication performed for V 1 at the plurality of points in time is synchronous data replication, and wherein the one or more metrics are regarding synchronous data replication performed for V 1 .

5. A method of performing data storage management comprising:
configuring a stretched volume in a metro cluster configuration including a first data storage system, DS 1 , and a second data storage system, DS 2 , wherein the stretched volume is configured using a first volume, V 1 , on DS 1 and a second volume, V 2 , on DS 2 , wherein V 1 and V 2 are configured as a same logical volume, L 1 , and exposed to a host as the same logical volume, L 1 , over a plurality of paths including a first path, P 1 , and a second path, P 2 , wherein P 1 is between the host and DS 1 , and wherein P 2 is between the host and DS 2 ;
receiving a plurality of I/O operations at the metro cluster configuration directed to the same logical volume L 1 , wherein the plurality of I/O operations includes a first plurality of write operations that are received at DS 1 and that write first data to V 1 ;
viewing values of a first metric regarding data replication performed for V 1 at a plurality of points in time, wherein the data replication is from DS 1 to DS 2 to replicate the first data written by the first plurality of write operations whereby the replicated first data is subsequently written to V 2 on DS 2 ;
responsive to said viewing, determining whether one or more of the values of the first metric denote that a problem has occurred in the metro cluster configuration at a first of the plurality of points in time; and
responsive to determining that the problem has occurred at the first point in time, performing one or more corrective actions, wherein at the first point in time, V 1 is configured as an unstretched or local resource and not included in the configuration of the stretched volume, and wherein the method further comprises:
responsive to determining that the one or more values for the first metric indicate that write data directed to V 1 is not being replicated from DS 1 to DS 2 at corresponding points in time, determining that the one or more values of the first metric are due to a valid configuration change and do not denote an occurrence of the problem in the metro cluster configuration since, at the first point in time, V 1 is unstretched and not included in the configuration for the stretched volume.

6. A non-transitory computer readable medium comprising code stored thereon that, when executed, performs a method of performing data storage management comprising:
configuring a plurality of stretched volumes in a metro cluster configuration including a first data storage system, DS 1 , and a second data storage system, DS 2 , wherein each stretched volume of the plurality of stretched volumes is configured using a corresponding one of a plurality of volume pairs, wherein each of the plurality of volume pairs for one of the plurality of stretched volumes is configured as one of a plurality of logical volumes, wherein said each volume pair includes a first volume, V 1 , on DS 1 and a second volume, V 2 , on DS 2 , wherein V 1 and V 2 are configured as said one logical volume, and exposed to a host as said one logical volume, over a plurality of paths including a first path, P 1 , and a second path, P 2 , wherein P 1 is between the host and DS 1 , and wherein P 2 is between the host and DS 2 ;
receiving a plurality of I/O operations at the metro cluster configuration directed to the plurality of stretched volumes;
viewing values of one or more metrics regarding data replication performed for V 1 at a plurality of points in time, wherein the data replication is from DS 1 to DS 2 to replicate the first data written by the first plurality of write operations whereby the replicated first data is subsequently written to V 2 on DS 2 , wherein the one or more metrics include at least one of: an average mirroring overhead latency for the plurality of stretched volumes indicating latency experienced when performing the data replication that replicates writes directed to the plurality of stretched volumes from DS 1 to DS 2 ; a mirrored write I/O rate denoting a number of write I/O operations in a specified unit of time for write I/Os that are directed to the plurality of stretched volumes and replicated from DS 1 to DS 2 ; an average mirrored write I/O bandwidth denoting an I/O bandwidth for write data transmitted when replicating write I/Os, that are directed to the plurality of stretched volumes, from DS 1 to DS 2 ; and an average mirrored write I/O latency for write I/Os that are directed to the

Assignees

• Emc Ip Holding Co Llc

Inventors

• Tylik Dmitry
• Gelotti Carole Ann
• Haravu Nagasimha G
• Blitzer Aharon
• Izikson Alexander
• Harduf Yuval
• Taylor Alan L
• Shem Tov Marina

Classifications

• G06F3/0683

  Plurality of storage devices · CPC title

• G06F3/0665

  at area level, e.g. provisioning of virtual or logical volumes · CPC title

• G06F3/065

  Replication mechanisms · CPC title

• G06F3/0614

  Improving the reliability of storage systems · CPC title

• G06F3/0604 Primary

  Improving or facilitating administration, e.g. storage management · CPC title