Abstract: A method includes, in a computing system in which one or more workloads access memory pages in a memory, defining multiple memory-page lists, and specifying for each memory-page list a respective different scanning period. Access frequencies, with which the memory pages are accessed, are estimated continually by periodically checking the memory pages on each memory-page list in accordance with the scanning period specified for that memory-page list, and re-assigning the memory pages to the memory-page lists based on the estimated access frequencies. One or more of the memory pages are evicted from the memory based on a history of assignments of the memory pages to the memory-page lists.

Abstract: A method includes, in a computing system in which one or more workloads access memory pages in a memory, defining multiple memory-page lists, and specifying for each memory-page list a respective different scanning period. Access frequencies, with which the memory pages are accessed, are estimated continually by periodically checking the memory pages on each memory-page list in accordance with the scanning period specified for that memory-page list, and re-assigning the memory pages to the memory-page lists based on the estimated access frequencies. One or more of the memory pages are evicted from the memory based on a history of assignments of the memory pages to the memory-page lists.

Abstract: A method includes running a Virtual Machine (VM) on a first compute node in a plurality of compute nodes that communicate with one another over a communication network. The VM is migrated from the first compute node to a second compute node in the plurality by generating, for memory pages accessed by the VM, page transfer state of one or more local memory pages that are accessed locally on the first compute node, and of one or more externalized memory pages whose access is not confined to the first node. Based on the page transfer state, the migrated VM is provided with access to the memory pages, including both the local and the externalized memory pages, on the second compute node.

Abstract: A method for operating a memory (28) includes storing data in a group of analog memory cells (32) of the memory as respective first voltage levels. After storing the data, second voltage levels are read from the respective analog memory cells. The second voltage levels are affected by cross-coupling interference causing the second voltage levels to differ from the respective first voltage levels. Cross-coupling coefficients, which quantify the cross-coupling interference among the analog memory cells, are estimated by processing the second voltage levels. The data stored in the group of analog memory cells is reconstructed from the read second voltage levels using the estimated cross-coupling coefficients.

Abstract: A method includes running a Virtual Machine (VM) on a first compute node in a plurality of compute nodes that communicate with one another over a communication network. The VM is migrated from the first compute node to a second compute node in the plurality by generating, for memory pages accessed by the VM, page transfer state of one or more local memory pages that are accessed locally on the first compute node, and of one or more externalized memory pages whose access is not confined to the first node. Based on the page transfer state, the migrated VM is provided with access to the memory pages, including both the local and the externalized memory pages, on the second compute node.

Abstract: A method includes storing data in an analog memory cell by writing an analog value into the memory cell. After storing the data, the data stored in the memory cell is read by discharging electrical current to flow through the memory cell, during a predefined time interval, while applying a variable voltage to a gate of the memory cell. A fraction of the predefined time interval, during which the variable voltage allows the electrical current to flow through the memory cell, is estimated. The stored data is estimated based on the estimated fraction.

Abstract: A method includes, in a plurality of compute nodes that communicate with one another over a communication network, running one or more Virtual Machines (VMs) that access storage blocks stored on non-volatile storage devices coupled to at least some of the compute nodes. One or more of the storage blocks accessed by a given VM, which runs on a first compute node, are cached in a volatile memory of a second compute node that is different from the first compute node. The cached storage blocks are served to the given VM.

Abstract: A method includes running on multiple compute nodes respective memory sharing agents that communicate with one another over a communication network. One or more local Virtual Machines (VMs), which access memory pages, run on a given compute node. Using the memory sharing agents, the memory pages that are accessed by the local VMs are stored on at least two of the compute nodes, and the stored memory pages are served to the local VMs.

Abstract: A data storage method includes, in a memory controller that accepts memory access commands from a host for execution in one or more memory units, holding a definition of a policy to be applied by the memory controller in the execution of the memory access commands in the memory units. The policy is reported from the memory controller to the host so as to cause the host to format memory access commands based on the reported policy.

Abstract: In a system that includes a host and a memory controller that is separate from the host and stores data for the host in a non-volatile memory, a method for data storage includes transferring from the memory controller to the host one or more source blocks from the non-volatile memory for compaction. The source blocks are compacted in the host by copying valid data from the source blocks into one or more destination blocks. The destination blocks are transferred from the host to the memory controller, and the destination blocks are stored by the memory controller in the non-volatile memory.

Abstract: A method includes storing data in an analog memory cell by writing an analog value into the memory cell. After storing the data, the data stored in the memory cell is read by discharging electrical current to flow through the memory cell, during a predefined time interval, while applying a variable voltage to a gate of the memory cell. A fraction of the predefined time interval, during which the variable voltage allows the electrical current to flow through the memory cell, is estimated. The stored data is estimated based on the estimated fraction.

Abstract: A method for operating a memory (28) includes storing data in a group of analog memory cells (32) of the memory as respective first voltage levels. After storing the data, second voltage levels are read from the respective analog memory cells. The second voltage levels are affected by cross-coupling interference causing the second voltage levels to differ from the respective first voltage levels. Cross-coupling coefficients, which quantify the cross-coupling interference among the analog memory cells, are estimated by processing the second voltage levels. The data stored in the group of analog memory cells is reconstructed from the read second voltage levels using the estimated cross-coupling coefficients.

Abstract: A method includes storing data in an analog memory cell by writing an analog value into the memory cell. After storing the data, the data stored in the memory cell is read by discharging electrical current to flow through the memory cell, during a predefined time interval, while applying a variable voltage to a gate of the memory cell. A fraction of the predefined time interval, during which the variable voltage allows the electrical current to flow through the memory cell, is estimated. The stored data is estimated based on the estimated fraction.

Abstract: A method for data storage in a memory that includes a plurality of analog memory cells includes estimating respective achievable storage capacities of the analog memory cells. The memory cells are assigned respective storage configurations defining quantities of data to be stored in the memory cells based on the estimated achievable capacities. The data is stored in the memory cells in accordance with the respective assigned storage configurations. The achievable storage capacities of the analog memory cells are re-estimated after the memory has been installed in a host system and used for storing the data in the host system. The storage configurations are modified responsively to the re-estimated achievable capacities.

Abstract: In a passive optical network, dynamic bandwidth allocation and queue management methods and algorithms, designed to avoid fragmentation loss, guarantee that a length of a grant issued by an OLT will match precisely the count for bytes to be transmitted to an ONU. The methods include determining an ONU uplink transmission egress based on a three-stage test, and various embodiments of methods for ONU report 700 threshold setting.

Abstract: A method for operating a memory (28) includes storing data in a group of analog memory cells (32) of the memory as respective first voltage levels. After storing the data, second voltage levels are read from the respective analog memory cells. The second voltage levels are affected by cross-coupling interference causing the second voltage levels to differ from the respective first voltage levels. Cross-coupling coefficients, which quantify the cross-coupling interference among the analog memory cells, are estimated by processing the second voltage levels. The data stored in the group of analog memory cells is reconstructed from the read second voltage levels using the estimated cross-coupling coefficients.

Abstract: A method for data storage includes, in a memory that includes at least N memory units, each memory unit including memory blocks, defining superblocks, each superblock including a respective set of N of the memory blocks that are allocated respectively in N different ones of the memory units, such that compaction of all the memory blocks in a given superblock is performed without any intervening programming operation in the given superblock. Data is stored in the memory by computing redundancy information for a selected portion of the data, and storing the selected portion and the redundancy information in the N memory blocks of a selected superblock.

Abstract: A method for operating a memory (28) includes storing data in a group of analog memory cells (32) of the memory as respective first voltage levels. After storing the data, second voltage levels are read from the respective analog memory cells. The second voltage levels are affected by cross-coupling interference causing the second voltage levels to differ from the respective first voltage levels. Cross-coupling coefficients, which quantify the cross-coupling interference among the analog memory cells, are estimated by processing the second voltage levels. The data stored in the group of analog memory cells is reconstructed from the read second voltage levels using the estimated cross-coupling coefficients.

Abstract: A method for registration of multiple entities belonging to a specific optical network unit (ONU). In one embodiment, the multiple entity registration method comprises checking by an optical line terminal (OLT) if a registration request message received from the specific ONU belongs to a certain grant, and based on the check result, registering an entity as either a first or as an additional entity of the specific ONU. In another embodiment, the method comprises checking by an OLT of a reserved value of a flags field inside a registration request message, and based on the check result, registering an entity as either a first or as an additional entity of the specific ONU. The knowledge by an OLT that multiple entities belong to a specific ONU is used for grant optimization and packet data flow optimization.

Abstract: A method includes, in a storage device that includes a non-volatile memory having a physical storage space, receiving data items associated with respective logical addresses assigned in a logical address space that is larger than the physical storage space. The logical addresses of the data items are translated into respective physical storage locations in the non-volatile memory. The data items are stored in the respective physical storage locations.