@@ -20,7 +20,7 @@ About this Task
20
20
---------------
21
21
22
22
- You can use :ref:`field customizations <rm-field-customizations>` to
23
- convert string columns to datetime in Relational Migrator.
23
+ convert string columns to datetime fields in Relational Migrator.
24
24
25
25
- CDC jobs that have time series configurations do not create the
26
26
time series collections until the CDC job completes.
@@ -41,20 +41,19 @@ Before you Begin
41
41
- If the time series data in your sync job does not require any
42
42
complex transformations you may benefit from migrating your time
43
43
series data as a separate sync job. This can make it easier to
44
- preload a large time series collection and once completed migrate
45
- the rest of your data.
44
+ preload a large time series collection and migrate the rest of your data after the initial job completes.
46
45
47
46
- If your migrating larges amounts of data, consider using
48
47
:ref:`table filters <rm-table-filters>` to batch your sync jobs
49
48
based on logical conditions. For example, you might import one year
50
- of historical data at a time using table filters. This allows a large
51
- job to be broken into more manageable segments.
49
+ of historical data at a time using table filters. This allows you to break down a large
50
+ job into more manageable segments.
52
51
53
52
- If a long-running sync job fails unexpectedly (for example, due to a
54
53
database outage) and is not able to automatically recover, you may
55
54
be able to use :ref:`table filters <rm-table-filters>` to create a
56
55
new sync job that picks up from where the previous one left off. This
57
- approach saves you from re-importing your previous data from scratch .
56
+ approach lets you avoid having to repeat the entire sync job .
58
57
59
58
Steps
60
59
-----
@@ -90,15 +89,15 @@ Steps
90
89
91
90
#. (Optional) Select a :guilabel:`Meta Field`.
92
91
93
- ``Meta Field`` is the name of the field which contains metadata
92
+ ``Meta Field`` is the name of the field that contains metadata
94
93
in each time series document. The metadata in the specified
95
- field should be data that is used to label a unique series of
94
+ field is used to label a unique series of
96
95
documents. The metadata should rarely, if ever, change.
97
96
98
97
#. Select a :guilabel:`Granularity`.
99
98
100
99
``Granularity`` should be a value that most closely matches
101
- the time between consecutive incoming timestamps. This
100
+ the time between consecutive incoming timestamps. Setting the correct granularity
102
101
improves performance by optimizing how MongoDB stores data in the
103
102
collection. Possible values are ``seconds``, ``minutes``, and
104
103
``hours``.
0 commit comments