The present disclosure generally relates to efficient block usage after ungraceful shutdown (UGSD) events. After a UGSD event, a host device is alerted by the data storage device that a QLC block that was being used prior to the UGSD event is experiencing an ongoing block recovery and that the block is not yet available to accept new data. The block is then checked to determine whether the block can continue to be used for the programming that was occurring at the time of the UGSD event. Once a determination is made, the data storage device notifies the host device so that normal operations may continue. Additionally, the amount of free blocks available for programming is monitored during UGSD events so that the host device can be warned if a power loss halt is triggered.

There are provided a memory system and an operating method thereof. A memory system includes: a plurality of storage regions, each including a plurality of memory cells; and a controller configured to provide a plurality of read retry sets, determine an applying order of the plurality of read retry sets based on characteristics of a read error occurred in a first storage region among the plurality of storage regions, and apply at least one of the read retry sets, based on the applying order, for a read retry operation performed on the first storage region.

A system, program product, and method for processing synchronized memory repairs. The method includes identifying a faulty memory row from a plurality of functioning memory rows in a memory array. The method also includes executing memory row repair operations directed toward the faulty memory row and identifying a repair row to operationally replace the faulty memory row. The method also includes creating a multiple hot state within a memory decoder. The memory decoder includes logic circuitry for executing operation of the plurality of functioning memory rows. The method further includes activating a wordline of the identified repair row through the multiple hot state, and executing one or more memory operations on the identified repair row though the memory decoder. Accordingly, the embodiments disclosed herein facilitate synchronization of the repair row and functioning memory rows within the memory array, as well as any associated peripheral signals.

A system comprising: a first subsystem comprising at least one first processor, and a second subsystem comprising one or more second processors. A first program is arranged to run on the at least one first processor, the first program being configured to send data from the first subsystem to the second subsystem. A second program is arranged to run on the one more second processors, the second program being configured to operate on the data content from the first subsystem. The first program is configured to set a checkpoint at one or more points in time. At each checkpoint it records in memory of the first subsystem i) a program state of the second program, comprising a state of one or more registers on each of the second processors at the time of the checkpoint, and ii) a copy of the data content sent to the second subsystem since the respective checkpoint.

Detecting execution hazards in offloaded operations is disclosed. A second offload operation is compared to a first offload operation that precedes the second offload operation. It is determined whether the second offload operation creates an execution hazard on an offload target device based on the comparison of the second offload operation to the first offload operation. If the execution hazard is detected, an error handling operation may be performed. In some examples, the offload operations are processing-in-memory operations.

When a computer boots up, a Basic Input/Output System (BIOS) configures system memory to have a crash memory area within the system address map, which can be used by a processor to dump crash memory data. When an error event occurs, the processor can initiate a dump to the crash memory area. Any desired data can be placed into the crash memory area, but typical data can include a state of registers in the processor. The processor then sets a flag, such as an external pin, indicating that the crash memory data is ready to be read. The flag can be read by a secure processor, which then reads the crash memory area at normal memory access speeds using the system bus. For example, the secure processor can access the crash memory area using Direct Memory Access (DMA) reads over a PCIe system bus.

Technologies for connecting data cables in a data center are disclosed. In the illustrative embodiment, racks of the data center are grouped into different zones based on the distance from the racks in a given zone to a network switch. All of the racks in a given zone are connected to the network switch using data cables of the same length. In some embodiments, certain physical resources such as storage may be placed in racks that are in zones closer to the network switch and therefore use shorter data cables with lower latency. An orchestrator server may, in some embodiments, schedule workloads or create virtual servers based on the different zones and corresponding latency of different physical resources.

Embodiments are generally directed apparatuses, methods, techniques and so forth to select two or more processing units of the plurality of processing units to process a workload, and configure a circuit switch to link the two or more processing units to process the workload, the two or more processing units each linked to each other via paths of communication and the circuit switch.

An error associated with a read operation corresponding to a target memory die of a memory sub-system is detected. In response to detecting the error, a first read throughput level of the memory sub-system is identified. The first read throughput level is adjusted to a second read throughput level. A read retry operation associated with the target memory die is executed at the second read throughput level.

One embodiment can provide an apparatus. The apparatus can include a persistent flush (PF) cache and a PF-tracking logic coupled to the PF cache. The PF-tracking logic is to: in response to receiving, from a media controller, an acknowledgment to a write request, determine whether the PF cache includes an entry corresponding to the media controller; in response to the PF cache not including the entry corresponding to the media controller, allocate an entry in the PF cache for the media controller; in response to receiving a persistence checkpoint, identify a media controller from a plurality of media controllers based on entries stored in the PF cache; issue a persistent flush request to the identified media controller to persist write requests received by the identified media controller; and remove an entry corresponding to the identified media controller from the PF cache subsequent to issuing the persistent flush request.