LogicalBytesPerSector 与 PhysicalBytesPerSector
LogicalBytesPerSector vs PhysicalBytesPerSector
当我执行命令 fsutil fsinfo sectorInfo c:
时,我收到以下输出:
LogicalBytesPerSector : 512
PhysicalBytesPerSectorForAtomicity : 4096
PhysicalBytesPerSectorForPerformance : 4096
FileSystemEffectivePhysicalBytesPerSectorForAtomicity : 4096 Device
Alignment : Aligned (0x000)
Partition alignment on device : Aligned
(0x000) No Seek Penalty Trim Supported Not DAX capable Not
Thinly-Provisioned
我担心 LogicalBytesPerSector
和 PhysicalBytesPerSectorForAtomicity
之间的区别。
由于LogicalBytesPerSector
小于PhysicalBytesPerSectorForAtomicity
,我写的每个512字节的块实际上是原子写的吗?另一方面,我写的每个 4096 字节的块实际上是原子写的吗?
另外,512字节的read/write会不会很慢,好像我理解的很好,系统需要读取4096个物理字节才能写入我的512字节?
does every block of 512 bytes I will wrote will be wrote atomically ? on the other way does every block of 4096 Bytes I will wrote will be wrote atomically?
4096 将自动写入。
512 将自动写入并可能写入其他数据以完成 4096。
Also does the read/wrote of 512 bytes will be not too much slow as, if I understand well, the system will need to read 4096 Physical bytes just to wrote my 512 bytes?
是的,至少由 PhysicalBytesPerSectorForPerformance
写。由于更有效的命令利用,更多可能会更快。
当我执行命令 fsutil fsinfo sectorInfo c:
时,我收到以下输出:
LogicalBytesPerSector : 512
PhysicalBytesPerSectorForAtomicity : 4096
PhysicalBytesPerSectorForPerformance : 4096
FileSystemEffectivePhysicalBytesPerSectorForAtomicity : 4096 Device
Alignment : Aligned (0x000)
Partition alignment on device : Aligned
(0x000) No Seek Penalty Trim Supported Not DAX capable Not
Thinly-Provisioned
我担心 LogicalBytesPerSector
和 PhysicalBytesPerSectorForAtomicity
之间的区别。
由于LogicalBytesPerSector
小于PhysicalBytesPerSectorForAtomicity
,我写的每个512字节的块实际上是原子写的吗?另一方面,我写的每个 4096 字节的块实际上是原子写的吗?
另外,512字节的read/write会不会很慢,好像我理解的很好,系统需要读取4096个物理字节才能写入我的512字节?
does every block of 512 bytes I will wrote will be wrote atomically ? on the other way does every block of 4096 Bytes I will wrote will be wrote atomically?
4096 将自动写入。
512 将自动写入并可能写入其他数据以完成 4096。
Also does the read/wrote of 512 bytes will be not too much slow as, if I understand well, the system will need to read 4096 Physical bytes just to wrote my 512 bytes?
是的,至少由 PhysicalBytesPerSectorForPerformance
写。由于更有效的命令利用,更多可能会更快。