Package org.platanios.tensorflow.proto

Class GPUOptions.Experimental

  • All Implemented Interfaces:
    com.google.protobuf.Message, com.google.protobuf.MessageLite, com.google.protobuf.MessageLiteOrBuilder, com.google.protobuf.MessageOrBuilder, java.io.Serializable, GPUOptions.ExperimentalOrBuilder
    Enclosing class:
    GPUOptions
    public static final class GPUOptions.Experimental
extends com.google.protobuf.GeneratedMessageV3
implements GPUOptions.ExperimentalOrBuilder
    Protobuf type org.platanios.tensorflow.proto.GPUOptions.Experimental
    See Also:
    Serialized Form

    • Field Detail

      • VIRTUAL_DEVICES_FIELD_NUMBER

        public static final int VIRTUAL_DEVICES_FIELD_NUMBER
        See Also:
        Constant Field Values

      • USE_UNIFIED_MEMORY_FIELD_NUMBER

        public static final int USE_UNIFIED_MEMORY_FIELD_NUMBER
        See Also:
        Constant Field Values

      • NUM_DEV_TO_DEV_COPY_STREAMS_FIELD_NUMBER

        public static final int NUM_DEV_TO_DEV_COPY_STREAMS_FIELD_NUMBER
        See Also:
        Constant Field Values

      • COLLECTIVE_RING_ORDER_FIELD_NUMBER

        public static final int COLLECTIVE_RING_ORDER_FIELD_NUMBER
        See Also:
        Constant Field Values

      • TIMESTAMPED_ALLOCATOR_FIELD_NUMBER

        public static final int TIMESTAMPED_ALLOCATOR_FIELD_NUMBER
        See Also:
        Constant Field Values

      • KERNEL_TRACKER_MAX_INTERVAL_FIELD_NUMBER

        public static final int KERNEL_TRACKER_MAX_INTERVAL_FIELD_NUMBER
        See Also:
        Constant Field Values

      • KERNEL_TRACKER_MAX_BYTES_FIELD_NUMBER

        public static final int KERNEL_TRACKER_MAX_BYTES_FIELD_NUMBER
        See Also:
        Constant Field Values

      • KERNEL_TRACKER_MAX_PENDING_FIELD_NUMBER

        public static final int KERNEL_TRACKER_MAX_PENDING_FIELD_NUMBER
        See Also:
        Constant Field Values

    • Method Detail

      • newInstance

        protected java.lang.Object newInstance​(com.google.protobuf.GeneratedMessageV3.UnusedPrivateParameter unused)
        Overrides:
        newInstance in class com.google.protobuf.GeneratedMessageV3

      • getUnknownFields

        public final com.google.protobuf.UnknownFieldSet getUnknownFields()
        Specified by:
        getUnknownFields in interface com.google.protobuf.MessageOrBuilder
        Overrides:
        getUnknownFields in class com.google.protobuf.GeneratedMessageV3

      • getDescriptor

        public static final com.google.protobuf.Descriptors.Descriptor getDescriptor()

      • internalGetFieldAccessorTable

        protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable internalGetFieldAccessorTable()
        Specified by:
        internalGetFieldAccessorTable in class com.google.protobuf.GeneratedMessageV3

      • getVirtualDevicesList

        public java.util.List<GPUOptions.Experimental.VirtualDevices> getVirtualDevicesList()
         The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
        repeated .org.platanios.tensorflow.proto.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;
        Specified by:
        getVirtualDevicesList in interface GPUOptions.ExperimentalOrBuilder

      • getVirtualDevicesOrBuilderList

        public java.util.List<? extends GPUOptions.Experimental.VirtualDevicesOrBuilder> getVirtualDevicesOrBuilderList()
         The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
        repeated .org.platanios.tensorflow.proto.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;
        Specified by:
        getVirtualDevicesOrBuilderList in interface GPUOptions.ExperimentalOrBuilder

      • getVirtualDevicesCount

        public int getVirtualDevicesCount()
         The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
        repeated .org.platanios.tensorflow.proto.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;
        Specified by:
        getVirtualDevicesCount in interface GPUOptions.ExperimentalOrBuilder

      • getVirtualDevices

        public GPUOptions.Experimental.VirtualDevices getVirtualDevices​(int index)
         The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
        repeated .org.platanios.tensorflow.proto.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;
        Specified by:
        getVirtualDevices in interface GPUOptions.ExperimentalOrBuilder

      • getVirtualDevicesOrBuilder

        public GPUOptions.Experimental.VirtualDevicesOrBuilder getVirtualDevicesOrBuilder​(int index)
         The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
        repeated .org.platanios.tensorflow.proto.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;
        Specified by:
        getVirtualDevicesOrBuilder in interface GPUOptions.ExperimentalOrBuilder

      • getUseUnifiedMemory

        public boolean 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;
        Specified by:
        getUseUnifiedMemory in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The useUnifiedMemory.

      • getNumDevToDevCopyStreams

        public 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;
        Specified by:
        getNumDevToDevCopyStreams in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The numDevToDevCopyStreams.

      • getCollectiveRingOrder

        public java.lang.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;
        Specified by:
        getCollectiveRingOrder in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The collectiveRingOrder.

      • getCollectiveRingOrderBytes

        public 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;
        Specified by:
        getCollectiveRingOrderBytes in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The bytes for collectiveRingOrder.

      • getTimestampedAllocator

        public boolean 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;
        Specified by:
        getTimestampedAllocator in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The timestampedAllocator.

      • getKernelTrackerMaxInterval

        public 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;
        Specified by:
        getKernelTrackerMaxInterval in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The kernelTrackerMaxInterval.

      • getKernelTrackerMaxBytes

        public 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;
        Specified by:
        getKernelTrackerMaxBytes in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The kernelTrackerMaxBytes.

      • getKernelTrackerMaxPending

        public 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;
        Specified by:
        getKernelTrackerMaxPending in interface GPUOptions.ExperimentalOrBuilder
        Returns:
        The kernelTrackerMaxPending.

      • isInitialized

        public final boolean isInitialized()
        Specified by:
        isInitialized in interface com.google.protobuf.MessageLiteOrBuilder
        Overrides:
        isInitialized in class com.google.protobuf.GeneratedMessageV3

      • writeTo

        public void writeTo​(com.google.protobuf.CodedOutputStream output)
             throws java.io.IOException
        Specified by:
        writeTo in interface com.google.protobuf.MessageLite
        Overrides:
        writeTo in class com.google.protobuf.GeneratedMessageV3
        Throws:
        java.io.IOException

      • getSerializedSize

        public int getSerializedSize()
        Specified by:
        getSerializedSize in interface com.google.protobuf.MessageLite
        Overrides:
        getSerializedSize in class com.google.protobuf.GeneratedMessageV3

      • equals

        public boolean equals​(java.lang.Object obj)
        Specified by:
        equals in interface com.google.protobuf.Message
        Overrides:
        equals in class com.google.protobuf.AbstractMessage

      • hashCode

        public int hashCode()
        Specified by:
        hashCode in interface com.google.protobuf.Message
        Overrides:
        hashCode in class com.google.protobuf.AbstractMessage

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(java.nio.ByteBuffer data)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(java.nio.ByteBuffer data,
                                                com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(com.google.protobuf.ByteString data)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(com.google.protobuf.ByteString data,
                                                com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(byte[] data)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(byte[] data,
                                                com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                         throws com.google.protobuf.InvalidProtocolBufferException
        Throws:
        com.google.protobuf.InvalidProtocolBufferException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(java.io.InputStream input)
                                         throws java.io.IOException
        Throws:
        java.io.IOException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(java.io.InputStream input,
                                                com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                         throws java.io.IOException
        Throws:
        java.io.IOException

      • parseDelimitedFrom

        public static GPUOptions.Experimental parseDelimitedFrom​(java.io.InputStream input)
                                                  throws java.io.IOException
        Throws:
        java.io.IOException

      • parseDelimitedFrom

        public static GPUOptions.Experimental parseDelimitedFrom​(java.io.InputStream input,
                                                         com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                                  throws java.io.IOException
        Throws:
        java.io.IOException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(com.google.protobuf.CodedInputStream input)
                                         throws java.io.IOException
        Throws:
        java.io.IOException

      • parseFrom

        public static GPUOptions.Experimental parseFrom​(com.google.protobuf.CodedInputStream input,
                                                com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                         throws java.io.IOException
        Throws:
        java.io.IOException

      • newBuilderForType

        public GPUOptions.Experimental.Builder newBuilderForType()
        Specified by:
        newBuilderForType in interface com.google.protobuf.Message
        Specified by:
        newBuilderForType in interface com.google.protobuf.MessageLite

      • toBuilder

        public GPUOptions.Experimental.Builder toBuilder()
        Specified by:
        toBuilder in interface com.google.protobuf.Message
        Specified by:
        toBuilder in interface com.google.protobuf.MessageLite

      • newBuilderForType

        protected GPUOptions.Experimental.Builder newBuilderForType​(com.google.protobuf.GeneratedMessageV3.BuilderParent parent)
        Specified by:
        newBuilderForType in class com.google.protobuf.GeneratedMessageV3

      • getParserForType

        public com.google.protobuf.Parser<GPUOptions.Experimental> getParserForType()
        Specified by:
        getParserForType in interface com.google.protobuf.Message
        Specified by:
        getParserForType in interface com.google.protobuf.MessageLite
        Overrides:
        getParserForType in class com.google.protobuf.GeneratedMessageV3

      • getDefaultInstanceForType

        public GPUOptions.Experimental getDefaultInstanceForType()
        Specified by:
        getDefaultInstanceForType in interface com.google.protobuf.MessageLiteOrBuilder
        Specified by:
        getDefaultInstanceForType in interface com.google.protobuf.MessageOrBuilder