UDFS_STATISTICS

Records aggregate information about file-system and object-store operations.

Records aggregate information about file-system and object-store operations. For access through LibHDFS++, the table records information about metadata but not data.

An operation can be made up of many individual read, write, or retry requests. SUCCESSFUL_OPERATIONS and FAILED_OPERATIONS count operations; the other counters count individual requests. When an operation finishes, one of the OPERATIONS counters is incremented once, but several other counters could be incremented several times each.

Column Name Data Type Description
FILESYSTEM VARCHAR Name of the file system, such as S3 or Libhdfs++.
SUCCESSFUL_OPERATIONS INTEGER Number of successful file-system operations.
FAILED_OPERATIONS INTEGER Number of failed file-system operations.
RETRIES INTEGER Number of retry events.
METADATA_READS INTEGER Number of requests to read metadata. For example, S3 list bucket and HEAD requests are metadata reads.
METADATA_WRITES INTEGER Number of requests to write metadata. For example, S3 POST and DELETE requests are metadata writes.
DATA_READS INTEGER Number of read operations, such as S3 GET requests to download files.
DATA_WRITES INTEGER Number of write operations, such as S3 PUT requests to upload files.
DOWNSTREAM_BYTES INTEGER Number of bytes received.
UPSTREAM_BYTES INTEGER Number of bytes sent.
OPEN_FILES INTEGER Number of files that are currently open.
MAPPED_FILES INTEGER Number of currently-mapped files on S3 file systems. This value shows the number of streaming connections for reading data from S3. This value will be 0 for non-S3 file systems.
READING INTEGER The number of currently-running read operations.
WRITING INTEGER The number of currently-running writer operations.

Examples

The following query gets the total number of metadata RPCs for Libhdfs++ operations:

=> SELECT SUM(metadata_reads) FROM UDFS_STATISTICS WHERE filesystem = 'Libhdfs++';