Remove linear qualification

This commit is contained in:
Quentin 2022-09-28 15:30:02 +02:00
parent ca4ad80447
commit 59afb0d32b
Signed by: quentin
GPG key ID: E9602264D639FF68

View file

@ -319,8 +319,8 @@ metadata engine and thus focus only on 16-byte objects.
It appears that the performances of our metadata engine are acceptable, as we
have a comfortable margin compared to Minio (Minio is between 3x and 4x times
slower per batch). We also note that, past 200k objects, Minio batch
completion time is constant as Garage's one remains linear: it could be
interesting to know if Garage batch's completion time would cross Minio's one
completion time is constant as Garage's one is still increasing in the observed range:
it could be interesting to know if Garage batch's completion time would cross Minio's one
for a very large number of objects. If we reason per object, both Minio and
Garage performances remain very good: it takes respectively around 20ms and
5ms to create an object. At 100 Mbps, if you upload a 10MB file, the
@ -333,10 +333,11 @@ Next, we focus on Garage's data only to better see its specific behavior:
![Showing the time to send 128 batches of 8192 objects for Garage only](1million.png)
Two effects are now more visible: 1. batch completion time is linear with the
Two effects are now more visible: 1. increasing batch completion time with the
number of objects in the bucket and 2. measurements are dispersed, at least
more than Minio. We discussed the first point previously but not the second
one on measurement dispersion. This instability could be an issue as it could
more than Minio. We don't know for sure if this increasing batch completion
time is linear or logarithmic as we don't have enough datapoint; additinal
measurements are needed. Concercning the observed instability, it could
be a symptom of what we saw with some other experiments in this machine:
sometimes it freezes under heavy I/O operations. Such freezes could lead to
request timeouts and failures. If it occurs on our testing computer, it will
@ -351,8 +352,8 @@ cluster at [deuxfleurs.fr](https://deuxfleurs) smoothly manages a bucket with
116k objects. This bucket contains real data: it is used by our Matrix instance
to store people's media files (profile pictures, shared pictures, videos,
audios, documents...). Thanks to this benchmark, we have identified two points
of vigilance: putting object duration seems linear with the number of existing
objects in the cluster, and we have some volatility in our measured data that
of vigilance: batch duration increases with the number of existing
objects in the cluster in the observed range, and we have some volatility in our measured data that
could be a symptom of our system freezing under the load. Despite these two
points, we are confident that Garage could scale way above 1M+ objects, but it
remains to be proved!