Use get_multi instead of get for datastore reads

[achals]

Signed-off-by: Achal Shah [email protected]

What this PR does / why we need it:

get is a wrapper around get_multi, and get_multi reduces the network costs involved.

Which issue(s) this PR fixes:

Fixes #1759

Does this PR introduce a user-facing change?:

Use a batch lookup API for Google Datastore. Users should see lower latencies for `get_online_features` when using Datastore as the online store.

[codecov-commenter]

Codecov Report

Merging #1814 (3a53cfb) into master (5857a55) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1814      +/-   ##
==========================================
- Coverage   84.85%   84.84%   -0.01%     
==========================================
  Files          92       92              
  Lines        6828     6844      +16     
==========================================
+ Hits         5794     5807      +13     
- Misses       1034     1037       +3     

Flag	Coverage Δ
integrationtests	84.77% <100.00%> (-0.01%)	⬇️
unittests	63.90% <10.00%> (-0.04%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
sdk/python/feast/infra/online_stores/datastore.py	93.49% <100.00%> (+0.39%)	⬆️
sdk/python/tests/doctest/test_all.py	92.72% <0.00%> (-3.20%)	⬇️
sdk/python/feast/infra/provider.py	88.49% <0.00%> (-0.60%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5857a55...3a53cfb. Read the comment docs.

[woop]

What is the performance difference between these two approaches? Have you run a benchmark?

[achals]

What is the performance difference between these two approaches? Have you run a benchmark?

retrieving 10 records, using pytest-benchmark. Single retreival seems to be strictly slower, which is not surprising since get is implemented as a wrapper over get_multi.

----------------------------------------------------------------------------------------- benchmark: 2 tests ----------------------------------------------------------------------------------------
Name (time in s)                              Min               Max              Mean            StdDev            Median               IQR            Outliers     OPS            Rounds  Iterations
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_online_retrieval_datastore_batch      2.8716 (1.0)      3.2226 (1.0)      2.9943 (1.0)      0.0986 (1.0)      2.9681 (1.0)      0.1220 (1.0)           8;0  0.3340 (1.0)          30           5
test_online_retrieval_datastore_single     4.0147 (1.40)     4.6170 (1.43)     4.2247 (1.41)     0.1446 (1.47)     4.2366 (1.43)     0.1492 (1.22)          9;3  0.2367 (0.71)         30           5
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

[felixwang9817]

/lgtm

[feast-ci-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: achals, felixwang9817

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [achals]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[feast-ci-bot]

New changes are detected. LGTM label has been removed.