International Business Machines Corporation

Alol Antony Crasta

Christoph Raisch

Kishorekumar G Pillai

Harshal S Patil

Nishant Ranjan

Roger Gregory Hathorn

John R Flanagan

Christopher J Colonna

Patricia G Driever

Mooheng Zee

Mikel William Welsh

Raghavan Devanathan

Provided are a computer program product, system, and method embodiments for reverting from a new security association to a previous security association in response to an error during a rekey operation. The responder maintains a first security association with the initiator having a first key to use to encrypt and decrypt messages transmitted with the initiator. The responder receives a message from the initiator for a rekey operation to establish a second security association with the initiator using a second key. The responder queues Input/Output (I/O) for transmission using the second key after completing the rekey operation. After activating the second security association, the responder receives a revert message from the initiator to revert back to using the first security association and first key in response to a failure of the rekey operation.

Reverting from a new security association to a previous security association in response to an error during a rekey operation

Provided are a computer program product, system, and method embodiments for reverting from a new security association to a previous security association in response to an error during a rekey operation. An initiator maintains a first security association with the responder having a first key to use to encrypt and decrypt data transmitted with the responder. The initiator initiates a rekey operation to establish a second security association with the responder using a second key. The initiator detects a failure of the rekey operation after the responder started using the second key for transmissions. A revert message is sent to the responder to revert back to using the first security association and first key in response to detecting the failure of the rekey operation.

Aspects of the invention include receiving an input/output (I/O) request that includes a data stream from a host processor. The receiving is at a network adapter of a storage controller that manages storage for the host processor. The storage controller includes a storage buffer to store data received from the host processor before migrating it to the storage. The storage controller also includes a data cache. It is determined whether the storage buffer has enough free space to store the received data stream. Based at least in part on determining that the storage buffer has enough free space to store the received data stream, the received data stream is stored by the network adapter in the storage. Based at least in part on determining that the storage buffer does not have enough free space to store the received data stream, the received data stream is stored in the data cache.

Guaranteed delivery in receiver side overcommitted communication adapters

A method, computer program product, and system includes a processor(s) connecting a first computer system to a boot swarm, initiating formation of a peer to peer network. The processor(s) receive a request from a second computer system, a request for a file. The processor(s) configure the second computer system, including implementing a client application hosted from a resource in the first computer system, to facilitate the second computer system joining the peer to peer network. The processor(s) determine immediate peer(s) in the peer to peer network available to provide the file to the second computer system. The processor(s) generate a magnet link that includes a listing of address(es) of the immediate peer(s), ranking address(es) from best source to worst source for downloading the file. The processor(s) provide the second computer system with the magnet link to utilize in downloading the file from a peer.

Network booting in a peer-to-peer environment using dynamic magnet links

A computer system includes a first server including a first data management module and a second server including a second data management module, and a first and second storage controller. The first and second storage controllers are interconnected by a network configured for reliable data multicasting. The computer system further includes a cross system manager for controlling one or more core storage functions on each of the first and second storage controllers. The computer system further includes computer program instructions to, responsive to a core storage function triggered by the cross system manager: Create a multicast group for each of the first server and the second server in the network, wherein the multicast group is created by a reliable multicast mechanism, and Multicast data write operations of each of the first server and the second server to the first storage controller and the second storage controller of the multicast group.

Updating data objects on a system

Aspects include creating a network multicast group in a storage area network (SAN) for a first computer system and a second computer system that are connected by the SAN and that each include a local cache for a shared storage device. Prior to one of the first or second computer system writing write data to the shared storage device: the write data is written to the local cache of the one of the first or second computer system; the write data is sent to the multicast group; based on the send operation completing, the write data is written to the shared storage device; and based on the write operation completing, the write completion is signaled to an operating system or hypervisor of the first and second computer systems.

Growing community of inventors

Bengaluru, India

Alol Antony Crasta