International Business Machines Corporation

Sedonasys Systems Ltd

Assaf Nitzan

Ehood Garmiza

Haim Helman

Orit Nissan-Messing

Eyal Perek

Ram Elron

Steven Robert Hetzler

Binny Sher Gill

James Lee Hafner

Daniel Felix Smith

Wendy Ann Belluomini

Ornan Alexander Gerstel

Oren S Li-On

Edi Shmueli

Eyal Lotem

Venu Nayar

Moty Cohen

Eliyahu Weissbrem

Eyal Lotern

Oren Li-on

Rami Elron

Migrating data traffic from legacy networks to packet networks by inserting a first circuit emulation device (CEM) at a first endpoint and a second circuit emulation device (CEM) at a second endpoint of a connection in the legacy network, duplicating data traffic provided to the first CEM by routing one copy of the data traffic over the connection in the legacy network to the second CEM and one copy of the data traffic over a packet network from the first CEM to a comparison server, duplicating data traffic provided to the second CEM over the legacy network from the first CEM by routing one copy of the data traffic to customer communication equipment and one copy of the data traffic to the comparison server, comparing the data traffic routed to the comparison server and validating that the packet network can be used to replace the connection in the legacy network.

System and method for network migration with minimal traffic impact

A method includes computing, in a local storage system having a local volume with a plurality of local regions, respective local checksum signatures over the local regions, and computing, in a remote storage system having a remote volume with remote regions in a one-to-one correspondence with the local regions, respective remote checksum signatures over the remote regions. A given remote region is identified, the given remote region having a given remote signature and a corresponding local region with a given local signature that does not match the given remote signature. The data in the given remote region is then replaced with data from the corresponding local region.

Synchronization for initialization of a remote mirror storage facility

Methods, apparatus and computer program products implement embodiments of the present invention that include detecting, by a storage system, a change in a mapping of a logical volume to one or more host ports of a host computer communicating with the storage system via a storage area network (SAN). Subsequent to detecting the change, first and second input/output (I/O) requests for the logical volume are received from a given host port, and a first unit attention message is conveyed to the given host port in response to the first I/O request. A second unit attention message is conveyed to the given host port upon determining that the storage system received the second I/O request within a specific time period commencing upon receiving the first I/O request. However, the second I/O request can be performed if the storage system received the second I/O request subsequent to the specific time period.

Safely mapping and unmapping host SCSI volumes

Methods, apparatus and computer program products implement embodiments of the present invention that include a storage device configured to store a logical volume mapped between a host computer and a computer having multiple initiator ports, and a processor configured to detect a unit attention condition for the logical volume. Examples of unit attention conditions for the logical volume include management operations on the logical volume and host requests for the logical volume. Upon determining, based on the unique serial number, that the processor has previously conveyed the unit attention code to any of the multiple initiator ports, to reconvey an input/output (I/O) request, receive a response for the I/O request, and convey the result to the host computer.

Unit attention processing in proxy and owner storage systems

Methods, apparatus and computer program products implement embodiments of the present invention that include a storage device configured to store a logical volume mapped between a host computer and a computer having multiple initiator ports, and a processor configured to detect a unit attention condition for the logical volume. Examples of unit attention conditions for the logical volume include management operations on the logical volume and host requests for the logical volume. Upon detecting a unit attention condition, the processor can to generate a unit attention code and a respective unique serial number in response to detecting the unit attention condition, and upon receiving an input/output (I/O) request for the logical volume from a given initiator port, the processor can convey, to the given initiator port, the unit attention code and the respective unique serial number in response to the I/O request.

A method includes computing, in a local storage system having a local volume with a plurality of local regions, respective local signatures over the local regions, and computing, in a remote storage system having a remote volume with remote regions in a one-to-one correspondence with the local regions, respective remote signatures over the remote regions. A given remote region is identified, the given remote region having a given remote signature and a corresponding local region with a given local signature that does not match the given remote signature. The data in the given remote region is then replaced with data from the corresponding local region.

In a local storage system having a local volume with a plurality of local regions, respective local signatures are computed over the local regions. In a remote storage system having a remote volume with remote regions in a one-to-one correspondence with the local regions, respective remote signatures are computed over the remote regions. A given remote region is identified, the given remote region having a given remote signature and a corresponding local region with a given local signature that does not match the given remote signature. The data in the given remote region is then replaced with data from the corresponding local region.

A method, including defining a snapshot referencing a source partition of a storage volume on a storage device, and receiving a request to write a block of data to the source partition. Upon receiving the write request, a delayed instant copy operation is initiated by allocating a target partition on the storage device and replacing, in the storage volume, the source partition with the target partition. A definition of a condition for completion of the delayed instant copy operation is received, and the delayed instant copy operation is completed upon the condition being met.

Delayed instant copy operation for short-lived snapshots

Embodiments of the invention relate to throttling accesses to a flash memory device. The flash memory device is part of a storage system that includes the flash memory device and a second memory device. The throttling is performed by logic that is external to the flash memory device and includes calculating a throttling factor responsive to an estimated remaining lifespan of the flash memory device. It is determined whether the throttling factor exceeds a threshold. Data is written to the flash memory device in response to determining that the throttling factor does not exceed the threshold. Data is written to the second memory device in response to determining that the throttling factor exceeds the threshold.

Growing community of inventors

Haifa, Israel

Assaf Nitzan