Techniques for workload redistibution

A method for redistributing workload comprising:

• selecting a first port from which a first workload is to be redistributed;
• selecting a second port with available capacity to perform additional workload;
• selecting a device accessible over both the first port and the second port;
• and
• setting each of one or more paths including the first port to a passive state for the device, wherein each of said one or more paths having the passive state for said device indicates to a client that sends I/O operations to said device that said each path is not to be used for sending I/O operations to said device and that said each path is available for use in performing control operations directed to said device, wherein a first plurality of devices are accessible to a host over the first port, said first plurality of devices being associated with a plurality of quotas, a different one of the plurality of quotas being specified for each of the first plurality of devices, whereby each of said plurality of quotas specified for one of the first plurality of devices denotes an upper bound of actual workload allowable for said one device through said first port.

A non-transitory computer readable medium comprising code stored thereon for redistributing workload, that when executed, performs a method comprising:

• selecting a first port from which a first workload is to be redistributed;
• selecting a second port with available capacity to perform additional workload;
• selecting a device accessible over both the first port and the second port;
• and
• setting each of one or more paths including the first port to a passive state for the device, wherein each of said one or more paths having the passive state for said device indicates to a client that sends I/O operations to said device that said each path is not to be used for sending I/O operations to said device and that said each path is available for use in performing control operations directed to said device, wherein a first plurality of devices, including the device, are accessible to the client over the first port, said first plurality of devices being associated with a plurality of quotas, and wherein said first port is overprovisioned having a maximum capacity that is less than a sum of the plurality of quotas.

A system comprising:

• a host;
• and
• a data storage system including:
• a first port from which a first workload is to be redistributed;
• a second port with available capacity to perform additional workload;
• a device accessible to the host over both the first port and the second port;
• and
• wherein the data storage system includes a processor which is programmed to and performs processing including:
• setting each of one or more paths including the first port to a passive state for the device, wherein each of said one or more paths having the passive state for said device indicates to the host that sends I/O operations to said device that said each path is not to be used for sending I/O operations to said device and that said each path is available for use in performing control operations directed to said device, wherein a first plurality of devices, including the device, are accessible to the host over the first port, said first plurality of devices being associated with a plurality of quotas, and wherein said first port is overprovisioned having a maximum capacity that is less than a sum of the plurality of quotas.