GPUOptions.ExperimentalOrBuilder

interfaccia statica pubblica GPUOptions.ExperimentalOrBuilder

Sottoclassi indirette conosciute

GPUOptions.Experimental , GPUOptions.Experimental.Builder

Metodi pubblici

stringa astratta	getCollectiveRingOrder () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
astratto com.google.protobuf.ByteString	getCollectiveRingOrderBytes () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
astratto int	getKernelTrackerMaxBytes () If kernel_tracker_max_bytes = n > 0, then a tracking event is inserted after every series of kernels allocating a sum of memory >= n.
astratto int	getKernelTrackerMaxInterval () Parameters for GPUKernelTracker.
astratto int	getKernelTrackerMaxPending () If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time.
astratto int	getNumDevToDevCopyStreams () If > 1, the number of device-to-device copy streams to create for each GPUDevice.
booleano astratto	getTimestampedAllocator () If true then extra work is done by GPUDevice and GPUBFCAllocator to keep track of when GPU memory is freed and when kernels actually complete so that we can know when a nominally free memory chunk is really not subject to pending use.
booleano astratto	getUseUnifiedMemory () If true, uses CUDA unified memory for memory allocations.
astratto GPUOptions.Experimental.VirtualDevices	getVirtualDevices (indice int) The multi virtual device settings.
astratto int	getVirtualDevicesCount () The multi virtual device settings.
Elenco astratto< GPUOptions.Experimental.VirtualDevices >	getVirtualDevicesList () The multi virtual device settings.
astratto GPUOptions.Experimental.VirtualDevicesOrBuilder	getVirtualDevicesOrBuilder (indice int) The multi virtual device settings.
Elenco astratto<? estende GPUOptions.Experimental.VirtualDevicesOrBuilder >	getVirtualDevicesOrBuilderList () The multi virtual device settings.

Metodi pubblici

public abstract String getCollectiveRingOrder ()

 If non-empty, defines a good GPU ring order on a single worker based on
 device interconnect.  This assumes that all workers have the same GPU
 topology.  Specify as a comma-separated string, e.g. "3,2,1,0,7,6,5,4".
 This ring order is used by the RingReducer implementation of
 CollectiveReduce, and serves as an override to automatic ring order
 generation in OrderTaskDeviceMap() during CollectiveParam resolution.

string collective_ring_order = 4;

estratto pubblico com.google.protobuf.ByteString getCollectiveRingOrderBytes ()

 If non-empty, defines a good GPU ring order on a single worker based on
 device interconnect.  This assumes that all workers have the same GPU
 topology.  Specify as a comma-separated string, e.g. "3,2,1,0,7,6,5,4".
 This ring order is used by the RingReducer implementation of
 CollectiveReduce, and serves as an override to automatic ring order
 generation in OrderTaskDeviceMap() during CollectiveParam resolution.

string collective_ring_order = 4;

public abstract int getKernelTrackerMaxBytes ()

 If kernel_tracker_max_bytes = n > 0, then a tracking event is
 inserted after every series of kernels allocating a sum of
 memory >= n.  If one kernel allocates b * n bytes, then one
 event will be inserted after it, but it will count as b against
 the pending limit.

int32 kernel_tracker_max_bytes = 8;

public abstract int getKernelTrackerMaxInterval ()

 Parameters for GPUKernelTracker.  By default no kernel tracking is done.
 Note that timestamped_allocator is only effective if some tracking is
 specified.
 If kernel_tracker_max_interval = n > 0, then a tracking event
 is inserted after every n kernels without an event.

int32 kernel_tracker_max_interval = 7;

public abstract int getKernelTrackerMaxPending ()

 If kernel_tracker_max_pending > 0 then no more than this many
 tracking events can be outstanding at a time.  An attempt to
 launch an additional kernel will stall until an event
 completes.

int32 kernel_tracker_max_pending = 9;

public abstract int getNumDevToDevCopyStreams ()

 If > 1, the number of device-to-device copy streams to create
 for each GPUDevice.  Default value is 0, which is automatically
 converted to 1.

int32 num_dev_to_dev_copy_streams = 3;

pubblico astratto booleano getTimestampedAllocator ()

 If true then extra work is done by GPUDevice and GPUBFCAllocator to
 keep track of when GPU memory is freed and when kernels actually
 complete so that we can know when a nominally free memory chunk
 is really not subject to pending use.

bool timestamped_allocator = 5;

pubblico astratto booleano getUseUnifiedMemory ()

 If true, uses CUDA unified memory for memory allocations. If
 per_process_gpu_memory_fraction option is greater than 1.0, then unified
 memory is used regardless of the value for this field. See comments for
 per_process_gpu_memory_fraction field for more details and requirements
 of the unified memory. This option is useful to oversubscribe memory if
 multiple processes are sharing a single GPU while individually using less
 than 1.0 per process memory fraction.

bool use_unified_memory = 2;