Slices (cluster) v1

Description

This page provides information about the slices in the cluster.

  1. Limitations of the queries, store and I/O columns
  2. Volume of records
  3. Clusters, nodes and slices

Columns

Name Type
node int2
slice int2
slice:type varchar
idle interval
store:blocks:sorted int8
store:blocks:unsorted int8
store:blocks:total int8
store:rows:sorted int8
store:rows:unsorted int8
store:rows:total int8
i/o:bytes read:disk int8
i/o:bytes read:network int8
i/o:rows:inserted int8
i/o:rows:deleted int8
i/o:rows:returned int8
i/o:bytes processed:in memory int8
i/o:bytes processed:on disk int8

Column Descriptions

node

The node ID. Nodes count from 0 and are contiguous.

slice

The slice ID. Slices count from 0 and are contiguous and unique across all nodes.

There is a single special case “slice”, which used to be number 6411 but now seems to be a value which varies by query and which is slightly above 12800, and this is in fact the leader node.

slice:type

This column indicates the slice types.

For ra3 type nodes, there are two slice types, full and partial.

For the other node types, there is only the one slice type, full.

In AWS parlance, a full slice is a data slice, and partial slice is a compute slice.

A full slice is the normal, usual, it’s-always-been-like-this slice, which is fully-featured, participates in all step types, owns part of each table, can access disk, S3, RMS, etc.

A partial slice is a recent addition and is used to help ameliorate a drawback of elastic resize.

idle

Duration since the start of most recent table access in stl_delete, stl_insert and stl_scan, which is to say, the time since the table last had rows deleted, inserted, or was scanned. If empty, there are no records in any of these system tables within the time period being examined by the page.

store:blocks:sorted

The number of sorted blocks.

store:blocks:unsorted

The number of unsorted blocks.

store:blocks:total

The total number of blocks.

store:rows:sorted

The number of sorted rows.

store:rows:unsorted

The number of unsorted rows.

store:rows:total

The total number of rows.

i/o:bytes read:disk

This column then shows the total number of bytes read from disk, as best I can judge the types indicated in stl_scan.

i/o:bytes read:network

This column then shows the total number of bytes read from network, as best I can judge the types indicated in stl_scan.

Importantly, it only counts the receive side of network activity - the step is scan, after all, not broadcast, so we’re not counting bytes twice.

i/o:rows:inserted

The number of rows inserted into the table.

For tables with all distribution, this is the physical number of rows (i.e. one per node), not the logical number of rows.

i/o:rows:deleted

The number of rows deleted from the table.

For tables with all distribution, this is the physical number of rows (i.e. one per node), not the logical number of rows.

i/o:rows:returned

It is the leader node returns rows to the SQL client, so for this page, this column is the number of rows returned to the leader node.

i/o:bytes processed:in memory

This column then shows the total number of bytes processed by the stl_aggr, stl_hash, stl_save, stl_sort and stl_unique steps, when running in memory rather than on disk.

i/o:bytes processed:on disk

This column then shows the total number of bytes processed by the stl_aggr, stl_hash, stl_save, stl_sort and stl_unique steps, when running on disk rather than in memory.