Abstract: A system and method for improved remote direct memory access (RDMA) for multi-host network interface controllers (NIC), the method including: allocating a first key to a first host, the first key corresponding to a first address of a memory device of the first host; and allocating the first key to a second host, wherein the second host is an RDMA NIC (rNIC) configured to offload at least a portion of storage operations from the first host.

Abstract: Systems and methods for providing input/output (I/O) determinism. An I/O instruction and at least one service level indicator are received, wherein the at least one service level indicator includes a required time for executing the I/O instruction. It is determined whether the I/O instruction can be executed on the storage within the required time. When it is not determined that the I/O instruction can be executed on the storage within the required time, a notification is sent.

Abstract: A computerized method for enabling a client device seamless access to a plurality of remote storage devices connected to the client device via a communication network. The method comprises receiving a plurality of physical addresses by a controller communicatively coupled to the client device and to a plurality of storage servers, each of the plurality of storage servers communicatively coupled to at least one storage device, the plurality of physical addresses enabling access by the controller to the remote storage devices. A single virtual storage device having a logical address space is generated on the device, wherein each of the plurality of physical addresses is mapped by the controller to a unique logical address of the virtual storage device.

Abstract: A client device for accessing remote storage devices. The client device includes: a processing unit communicatively connected to: a network interface and a memory; the network interface configured to communicatively connect the client device to a network comprising the client device and a storage server, the storage server comprising one or more storage devices, a remote direct memory access network interface controller (rNIC), and a primary processing unit; the memory containing instructions, that, when executed by the processing unit, configure the client device to: send an input/output (I/O) request for a storage device of the one or more storage devices; receive an indication from the rNIC based on the sent I/O request, the indication including an interrupt generated by the rNIC in response to the I/O request.

Abstract: A network interface controller (NIC) for use in a server controlling a plurality of storage devices provides remote access to a plurality of clients each accessing one or more storage devices of the plurality of storage devices. The NIC is configured to receive from at least a client device an input/output (I/O) request for a storage device of the plurality of storage devices and store the I/O request in a secondary memory thereof. The I/O request is sent to the storage device and a response respective of the I/O request intended for the primary processing unit of the server is received by the NIC. The client device is sent an indication that the I/O request is completed. In some exemplary embodiments, the NIC determines if the client device has appropriate I/O permissions to access a specific data block.

Abstract: Systems and methods for providing input/output (I/O) determinism. An I/O instruction and at least one service level indicator are received, wherein the at least one service level indicator includes a required time for executing the I/O instruction. It is determined whether the I/O instruction can be executed on the storage within the required time. When it is not determined that the I/O instruction can be executed on the storage within the required time, a notification is sent.

Abstract: A system and method for providing remote direct memory access failover. The method includes receiving, at a storage server including a network storage device, an access instruction over a first communication path between a client device and the storage server, wherein the network storage device includes a first memory and a second memory, wherein the first memory stores a first access queue; storing the access instruction in the first access queue; storing a second access queue in the second memory when the first communication path has failed, wherein the second access queue is at least a portion of the first access queue including the access instruction; determining a status of the access instruction when a second communication path is established with the client device; and sending, to the client device, a notification indicating the determined status of the access instruction via the second communication path.

Abstract: A system and method for reduced bandwidth remote storage between a client device and a storage device storing at least a first data block and a first exclusive or (XOR) product. The method includes: receiving, from the client device, a second exclusive XOR product, wherein the second XOR product is determined based on the first data block and a third data block; reading, from the storage device, the first XOR product, wherein the first XOR product is determined based on the first data block and at least a second data block; determining a third XOR product based on the first XOR product and the second XOR product; and storing, in the storage device, the third XOR product.

Abstract: A system and method for increased efficiency thin provisioning of data storage. The method includes: receiving a request for storage allocation from a client device; and allocating, to the client device, at least a portion of a first storage of the plurality of storages when a garbage collector of the first storage is turned off.

Abstract: A system and method for dynamic caching by a client device having remote memory access to a server. The system includes: a processing circuitry; and at least one memory, the at least one memory containing instructions that, when executed by the processing circuitry, configure the system to: configure a network interface of the client device to: request a memory allocation of at least a portion of a storage of the client device; receive, in real-time, the requested memory allocation of the client device storage; and store, in the allocated at least a portion of the client device storage, at least a first portion of cached metadata, wherein the cached metadata corresponds to at least an access operation between the client device and the server.

Abstract: A system and method for providing data redundancy. The method includes receiving a write instruction, the write instruction including data to be written to a primary remote direct access (RDMA) storage and a primary block address corresponding to a designated primary block of the primary RDMA storage, wherein the designated primary block is associated with a primary lock stored in the primary RDMA storage; determining whether the primary lock and at least one secondary lock are obtained, wherein each secondary lock is associated with a secondary block of a distinct secondary RDMA storage; and writing the data included in the write instruction to the designated primary data block and each secondary data block associated with the obtained at least one secondary lock, when it is determined that the primary lock and the at least one secondary lock have been obtained.

Abstract: To improve upon some of the characteristics of current storage systems in general and block data storage systems in particular, exemplary embodiments combine state-of-the art networking techniques with state-of-the-art data storage elements in a novel way. To accomplish this combination in a highly effective way, it is proposed to combine networking remote direct memory access (RDMA) technique and storage-oriented memory mapped input output (MMIO) technique in a system to provide direct access from a remote storage client to a remote storage system with little to no central processing unit (CPU) intervention of the remote storage server. In some embodiments, this technique may reduce the required CPU intervention on the client side. These reductions of CPU intervention potentially reduce latency while providing performance improvements, and/or providing more data transfer bandwidth and/or throughput and/or more operations per second compared to other systems with equivalent hardware.

Abstract: A computerized method for enabling a client device seamless access to a plurality of remote storage devices connected to the client device via a communication network. The method comprises receiving a plurality of physical addresses by a controller communicatively coupled to the client device and to a plurality of storage servers, each of the plurality of storage servers communicatively coupled to at least one storage device, the plurality of physical addresses enabling access by the controller to the remote storage devices. A single virtual storage device having a logical address space is generated on the device, wherein each of the plurality of physical addresses is mapped by the controller to a unique logical address of the virtual storage device.

Abstract: A network interface controller (NIC) for use in a server controlling a plurality of storage devices provides remote access to a plurality of clients each accessing one or more storage devices of the plurality of storage devices. The NIC is configured to receive from at least a client device an input/output (I/O) request for a storage device of the plurality of storage devices and store the I/O request in a secondary memory thereof. The I/O request is sent to the storage device and a response respective of the I/O request intended for the primary processing unit of the server is received by the NIC. The client device is sent an indication that the I/O request is completed. In some exemplary embodiments, the NIC determines if the client device has appropriate I/O permissions to access a specific data block.

Abstract: To improve upon some of the characteristics of current storage systems in general and block data storage systems in particular, exemplary embodiments combine state-of-the art networking techniques with state-of-the-art data storage elements in a novel way. To accomplish this combination in a highly effective way, it is proposed to combine networking remote direct memory access (RDMA) technique and storage-oriented memory mapped input output (MMIO) technique in a system to provide direct access from a remote storage client to a remote storage system with little to no central processing unit (CPU) intervention of the remote storage server. In some embodiments, this technique may reduce the required CPU intervention on the client side. These reductions of CPU intervention potentially reduce latency while providing performance improvements, and/or providing more data transfer bandwidth and/or throughput and/or more operations per second compared to other systems with equivalent hardware.

Abstract: A controller system allows direct access to a client device accessing remote data storage devices communicatively connected to the controller. A virtual block metadata (VBM) is generated by the controller system, the VBM including a plurality of maps. Each map includes a plurality of virtual addresses, each virtual address mapped to a physical address, each physical address on a remote storage device of the plurality of storage devices. The client device is sent a portion of the VBM, the portion including the map of the virtual addresses sent to the client. The controller receives, over a first network interface, read/write instructions from the client device for the a physical address corresponding to the virtual address allocated to the client device.

Abstract: A system for controlling a multi-tiered storage device (MTSD) includes a first group of storage devices and a second group of storage devices, wherein each storage device of the first group of storage devices has a higher endurance than each storage device of a second group of storage devices of the MTSD. A block of data is received by a controller of the MTSD, and is written to a storage device of the first group of storage devices. Upon determination that the block of data has been written to infrequently, the block of data is written to a storage device of the second group of storage devices. The block of data may then be erased from the storage device of the first group of storage devices. In some embodiments, a storage device from the first group is associated with the second group, upon determination that the storage has degraded.

Abstract: According to a disclosed embodiment of the invention, an improved cluster-based collection of computers (nodes) is realized using conventional computer hardware. Software is provided that enables at least one virtual machine to be presented to guest operating systems, wherein each node participating with the virtual machine has its own emulator or virtual machine monitor. VM memory coherency and I/O coherency are provided by hooks, which result in the manipulation of internal processor structures. A private network provides communication among the nodes.

Abstract: According to a disclosed embodiment of the invention, an improved cluster-based collection of computers (nodes) is realized using conventional computer hardware. Software is provided that enables at least one virtual machine to be presented to guest operating systems, wherein each node participating with the virtual machine has its own emulator or virtual machine monitor. VM memory coherency and I/O coherency are provided by hooks, which result in the manipulation of internal processor structures. A private network provides communication among the nodes.

Abstract: According to a disclosed embodiment of the invention, an improved cluster-based collection of computers (nodes) is realized using conventional computer hardware. Software is provided that enables at least one virtual machine to be presented to guest operating systems, wherein each node participating with the virtual machine has its own emulator or virtual machine monitor. VM memory coherency and I/O coherency are provided by hooks, which result in the manipulation of internal processor structures. A private network provides communication among the nodes.